Tutorial Analysis Service Tabular Model

Adventure Works Internet Sales tutorial (1400)
5/9/2019 • 3 minutes to read • Edit Online
APPLIES TO: SQL Server 2017 Analysis Services and later Azure Analysis Services Power BI
Premium
This tutorial provides lessons on how to create and deploy a tabular model at the 1400 compatibility level. If you're
new to Analysis Services and tabular modeling, completing this tutorial is the quickest way to learn how to create
and deploy a basic tabular model by using Visual Studio. Once you have the prerequisites in-place, it should take
two to three hours to complete.
What you learn

How to create a new tabular model project at the 1400 compatibility level in Visual Studio with SSDT.
How to import data from a relational database into a tabular model project workspace database.
How to create and manage relationships between tables in the model.
How to create calculated columns, measures, and Key Performance Indicators that help users analyze critical
business metrics.
How to create and manage perspectives and hierarchies that help users more easily browse model data by
providing business and application-specific viewpoints.
How to create partitions that divide table data into smaller logical parts that can be processed independent
from other partitions.
How to secure model objects and data by creating roles with user members.
How to deploy a tabular model to an Azure Analysis Services server or SQL Server 2017 Analysis
Services server by using SSDT.
Prerequisites
To complete this tutorial, you need:
An Azure Analysis Services server or a SQL Server 2017 Analysis Services server in Tabular mode. Sign up
for a free Azure Analysis Services trial and create a server or download a free SQL Server 2017 Developer
Edition.
An Azure SQL Data Warehouse with the sample AdventureWorksDW database, or an on-premises SQL
Server Data Warehouse with an AdventureWorksDW sample database. When installing an
AdventureWorksDW database to an on-premises SQL Server Data Warehouse, use the sample database
version that corresponds with your server version.
Important: If you install the sample database to an on-premises SQL Server Data Warehouse, and deploy
your model to an Azure Analysis Services server, an On-premises data gateway is required.
The latest version of SQL Server Data Tools (SSDT). Or, if you already have Visual Studio 2017, you can
download and install Microsoft Analysis Services Projects (VSIX) package. For this tutorial, references to
SSDT and Visual Studio are synonymous.
The latest version of SQL Server Management Studio (SSMS ).
A client application such as Power BI Desktop or Excel.
Scenario
This tutorial is based on Adventure Works Cycles, a fictitious company. Adventure Works is a large, multinational
manufacturing company that produces and distributes bicycles, parts, and accessories for commercial markets in
North America, Europe, and Asia. The company employs 500 workers. Additionally, Adventure Works employs
several regional sales teams throughout its market base. Your project is to create a tabular model for sales and
marketing users to analyze Internet sales data in the AdventureWorksDW database.
To complete the tutorial, you must complete various lessons. In each lesson, there are tasks. Completing each task
in order is necessary for completing the lesson. While in a particular lesson there may be several tasks that
accomplish a similar outcome, but how you complete each task is slightly different. This method shows there is
often more than one way to complete a task, and to challenge you by using skills you've learned in previous
lessons and tasks.
The purpose of the lessons is to guide you through authoring a basic tabular model by using many of the features
included in SSDT. Because each lesson builds upon the previous lesson, you should complete the lessons in order.
This tutorial does not provide lessons about managing a server in Azure portal, managing a server or database by
using SSMS, or using a client application to browse model data.
Lessons
This tutorial includes the following lessons:
LESSON ESTIMATED TIME TO COMPLETE
1 - Create a new tabular model project 10 minutes
2 - Get data 10 minutes
3 - Mark as Date Table 3 minutes
4 - Create relationships 10 minutes
5 - Create calculated columns 15 minutes
6 - Create measures 30 minutes
7 - Create Key Performance Indicators (KPI) 15 minutes
8 - Create perspectives 5 minutes
9 - Create hierarchies 20 minutes
10 - Create partitions 15 minutes
11 - Create roles 15 minutes
12 - Analyze in Excel 5 minutes
13 - Deploy 5 minutes
Supplemental lessons
These lessons are not required to complete the tutorial, but can be helpful in better understanding advanced
tabular model authoring features.
Detail Rows 10 minutes
Dynamic security 30 minutes
Ragged hierarchies 20 minutes
Next steps
To get started, see Lesson 1: Create a new tabular model project.
Create a tabular model project
Premium
In this lesson, you use Visual Studio with SQL Server Data Tools (SSDT) or Visual Studio 2017 with Microsoft
Analysis Services Projects VSIX to create a new tabular model project at the 1400 compatibility level. Once your
new project is created, you can begin adding data and authoring your model. This lesson also gives you a brief
introduction to the tabular model authoring environment in Visual Studio.
Estimated time to complete this lesson: 10 minutes
Prerequisites
This article is the first lesson in a tabular model authoring tutorial. To complete this lesson, there are several
prerequisites you need to have in-place. To learn more, see Analysis Services - Adventure Works tutorial.
Create a new tabular model project

To create a new tabular model project
1. In Visual Studio, on the File menu, click New > Project.
2. In the New Project dialog box, expand Installed > Business Intelligence > Analysis Services, and then
click Analysis Services Tabular Project.
3. In Name, type AW Internet Sales, and then specify a location for the project files.
By default, Solution Name is the same as the project name; however, you can type a different solution
name.
4. Click OK.
5. In the Tabular model designer dialog box, select Integrated workspace.
The workspace hosts a tabular model database with the same name as the project during model authoring.
Integrated workspace means Visual Studio uses a built-in instance, eliminating the need to install a separate
Analysis Services server instance just for model authoring.
6. In Compatibility level, select SQL Server 2017 / Azure Analysis Services (1400).
If you don't see SQL Server 2017 / Azure Analysis Services (1400) in the Compatibility level listbox, you're
not using the latest version of SQL Server Data Tools. To get the latest version, see Install SQL Server Data
tools.
Understanding the SSDT tabular model authoring environment

Now that you've created a new tabular model project, let's take a moment to explore the tabular model authoring
environment in Visual Studio.
After your project is created, it opens in Visual Studio. On the right side, in Tabular Model Explorer, you see a
tree view of the objects in your model. Since you haven't yet imported data, the folders are empty. You can right-
click an object folder to perform actions, similar to the menu bar. As you step through this tutorial, you use the
Tabular Model Explorer to navigate different objects in your model project.
Click the Solution Explorer tab. Here, you see your Model.bim file. If you don't see the designer window to the
left (the empty window with the Model.bim tab), in Solution Explorer, under AW Internet Sales Project,
double-click the Model.bim file. The Model.bim file contains the metadata for your model project.
Click Model.bim. In the Properties window, you see the model properties, most important of which is the
DirectQuery Mode property. This property specifies if the model is deployed in In-Memory mode (Off) or
DirectQuery mode (On). For this tutorial, you author and deploy your model in In-Memory mode.
When you create a model project, certain model properties are set automatically according to the Data Modeling
settings that can be specified in the Tools menu > Options dialog box. Data Backup, Workspace Retention, and
Workspace Server properties specify how and where the workspace database (your model authoring database) is
backed up, retained in-memory, and built. You can change these settings later if necessary, but for now, leave these
properties as they are.
In Solution Explorer, right-click AW Internet Sales (project), and then click Properties. The AW Internet
Sales Property Pages dialog box appears. You set some of these properties later when you deploy your model.
When you installed SSDT, several new menu items were added to the Visual Studio environment. Click the Model
menu. From here, you can import data, refresh workspace data, browse your model in Excel, create perspectives
and roles, select the model view, and set calculation options. Click the Table menu. From here, you can create and
manage relationships, specify date table settings, create partitions, and edit table properties. If you click the
Column menu, you can add and delete columns in a table, freeze columns, and specify sort order. SSDT also adds
some buttons to the bar. Most useful is the AutoSum feature to create a standard aggregation measure for a
selected column. Other toolbar buttons provide quick access to frequently used features and commands.
Explore some of the dialogs and locations for various features specific to authoring tabular models. While some
items are not yet active, you can get a good idea of the tabular model authoring environment.
What's next?
Lesson 2: Get data.
Get data
Premium
In this lesson, you use Get Data to connect to the AdventureWorksDW sample database, select data, preview and
filter, and then import into your model workspace.
Under the hood, Get Data is Power Query, which provides a vast array of tools for connecting to and reshaping
data for modeling and analysis. To learn more, see Power Query Documentation.
NOTE
Tasks and images in this tutorial show connecting to an AdventureWorksDW2014 database on an on-premises server. In
some cases, an AdventureWorksDW database on Azure SQL Data Warehouse may show different objects; however, they are
fundamentally the same.
Prerequisites
This article is part of a tabular modeling tutorial, which should be completed in order. Before performing the tasks
in this lesson, you should have completed the previous lesson: Lesson 1: Create a new tabular model project.
Create a connection
To create a connection to the AdventureWorksDW database
1. In Tabular Model Explorer, right-click Data Sources > Import from Data Source.
This launches Get Data, which guides you through connecting to a data source. If you don't see Tabular
Model Explorer, in Solution Explorer, double-click Model.bim to open the model in the designer.
2. In Get Data, click Database > SQL Server Database > Connect.
3. In the SQL Server Database dialog, in Server, type the name of the server where you installed the
AdventureWorksDW database, and then click Connect.
4. When prompted to enter credentials, you need to specify the credentials Analysis Services uses to connect to
the data source when importing and processing data. In Impersonation Mode, select Impersonate
Account, then enter credentials, and then click Connect. It's recommended you use an account where the
password doesn't expire.
NOTE
Using a Windows user account and password provides the most secure method of connecting to a data source.
5. In Navigator, select the AdventureWorksDW database, and then click OK. This creates the connection to
the database.
6. In Navigator, select the check box for the following tables: DimCustomer, DimDate, DimGeography,
DimProduct, DimProductCategory, DimProductSubcategory, and FactInternetSales. After selecting
the tables, click Edit.
After you click Edit, Query Editor opens. In the next section, you select only the data you want to import.
Filter the table data

Tables in the AdventureWorksDW sample database have data that isn't necessary to include in your model. When
possible, you want to filter out unnecessary data to save in-memory space used by the model. You filter out some
of the columns from tables so they're not imported into the workspace database, or the model database after it has
been deployed.
To filter the table data before importing
1. In Query Editor, select the DimCustomer table. A view of the DimCustomer table at the datasource (your
AdventureWorksDW sample database) appears.
2. Multi-select (Ctrl + click) SpanishEducation, FrenchEducation, SpanishOccupation,
FrenchOccupation, then right-click, and then click Remove Columns.
Since the values for these columns are not relevant to Internet sales analysis, there is no need to import
these columns. Eliminating unnecessary columns makes your model smaller and more efficient.
TIP
If you make a mistake, you can backup by deleting a step in APPLIED STEPS.
3. Filter the remaining tables by removing the following columns in each table:
DimDate
COLUMN
DateKey
SpanishDayNameOfWeek
FrenchDayNameOfWeek
SpanishMonthName
FrenchMonthName
DimGeography
COLUMN
SpanishCountryRegionName
FrenchCountryRegionName
IpAddressLocator
DimProduct
COLUMN
SpanishProductName
FrenchProductName
FrenchDescription
ChineseDescription
ArabicDescription
HebrewDescription
ThaiDescription
GermanDescription
JapaneseDescription
TurkishDescription
DimProductCategory
COLUMN
SpanishProductCategoryName
FrenchProductCategoryName
DimProductSubcategory
COLUMN
SpanishProductSubcategoryName
FrenchProductSubcategoryName
FactInternetSales
No columns removed.
Import the selected tables and column data

Now that you've previewed and filtered out unnecessary data, you can import the rest of the data you do want. The
wizard imports the table data along with any relationships between tables. New tables and columns are created in
the model and data that you filtered out is not be imported.
To import the selected tables and column data
1. Review your selections. If everything looks okay, click Import. The Data Processing dialog shows the status
of data being imported from your datasource into your workspace database.
2. Click Close.
Save your model project

It's important to frequently save your model project.
To save the model project
Click File > Save All.
What's next?
Lesson 3: Mark as Date Table.
Mark as Date Table
Premium
In Lesson 2: Get data, you imported a dimension table named DimDate. While in your model this table is named
DimDate, it can also be known as a Date table, in that it contains date and time data.
Whenever you use DAX time-intelligence functions, like when you create measures later, you must specify
properties which include a Date table and a unique identifier Date column in that table.
In this lesson, you mark the DimDate table as the Date table and the Date column (in the Date table) as the Date
column (unique identifier).
Before you mark the date table and date column, it's a good time to do a little housekeeping to make your model
easier to understand. Notice in the DimDate table a column named FullDateAlternateKey. This column contains
one row for every day in each calendar year included in the table. You use this column a lot in measure formulas
and in reports. But, FullDateAlternateKey isn't really a good identifier for this column. You rename it to Date,
making it easier to identify and include in formulas. Whenever possible, it's a good idea to rename objects like
tables and columns to make them easier to identify in SSDT and client reporting applications.
Estimated time to complete this lesson: Three minutes
Prerequisites
in this lesson, you should have completed the previous lesson: Lesson 2: Get data.
To rename the FullDateAlternateKey column
1. In the model designer, click the DimDate table.
2. Double-click the header for the FullDateAlternateKey column, and then rename it to Date.
To set Mark as Date Table
1. Select the Date column, and then in the Properties window, under Data Type, make sure Date is selected.
2. Click the Table menu, then click Date, and then click Mark as Date Table.
3. In the Mark as Date Table dialog box, in the Date listbox, select the Date column as the unique identifier.
It's usually selected by default. Click OK.
What's next?
Lesson 4: Create relationships.
Create relationships
Premium
In this lesson, you verify the relationships that were created automatically when you imported data and add new
relationships between different tables. A relationship is a connection between two tables that establishes how the
data in those tables should be correlated. For example, the DimProduct table and the DimProductSubcategory
table have a relationship based on the fact that each product belongs to a subcategory. To learn more, see
Relationships.
Prerequisites
in this lesson, you should have completed the previous lesson: Lesson 3: Mark as Date Table.
Review existing relationships and add new relationships

When you imported data by using Get Data, you got seven tables from the AdventureWorksDW database.
Generally, when you import data from a relational source, existing relationships are automatically imported
together with the data. In order for Get Data to automatically create relationships in the data model, there must be
relationships between tables at the data source.
Before you proceed with authoring your model, you should verify those relationships between tables were created
properly. For this tutorial, you also add three new relationships.
To review existing relationships
1. Click the Model menu > Model View > Diagram View.
The model designer now appears in Diagram View, a graphical format displaying all the tables you
imported with lines between them. The lines between tables indicate the relationships that were
automatically created when you imported the data.
NOTE
If you don't see any relationships between tables, it likely means there are no relationships between those tables at
the datasource.
Include as many of the tables as possible by using minimap controls in the lower-right corner of the model
designer. You can also click and drag tables to different locations, bringing tables closer together, or putting
them in a particular order. Moving tables does not affect the relationships between the tables. To view all
the columns in a particular table, click and drag on a table edge to expand or make it smaller.
2. Click the solid line between the DimCustomer table and the DimGeography table. The solid line between
these two tables shows this relationship is active, that is, it is used by default when calculating DAX
formulas.
Notice the GeographyKey column in the DimCustomer table and the GeographyKey column in the
DimGeography table now both each appear within a box. These columns are used in the relationship. The
relationship's properties now also appear in the Properties window.
TIP
You can also use the Manage Relationships dialog box to show the relationships between all tables in a table format.
In Tabular Model Explorer, right-click Relationships > Manage Relationships.
3. Verify the following relationships were created when each of the tables were imported from the
AdventureWorksDW database:
ACTIVE TABLE RELATED LOOKUP TABLE
Yes DimCustomer [GeographyKey] DimGeography [GeographyKey]
Yes DimProduct DimProductSubcategory

[ProductSubcategoryKey] [ProductSubcategoryKey]
Yes DimProductSubcategory DimProductCategory

[ProductCategoryKey] [ProductCategoryKey]
Yes FactInternetSales [CustomerKey] DimCustomer [CustomerKey]
Yes FactInternetSales [ProductKey] DimProduct [ProductKey]
If any of the relationships are missing, verify your model includes the following tables: DimCustomer,
DimDate, DimGeography, DimProduct, DimProductCategory, DimProductSubcategory, and
FactInternetSales. If tables from the same datasource connection are imported at separate times, any
relationships between those tables are not be created and must be created manually. If no relationships
appear, it means there are no relationships at the datasource. You can create them manually in the data
model.
Take a closer look
In Diagram View, notice an arrow, an asterisk, and a number on the lines that show the relationship between
tables.
The arrow shows the filter direction. The asterisk shows this table is the many side in the relationship's cardinality,
and the one shows this table is the one side of the relationship. If you need to edit a relationship; for example,
change the relationship's filter direction or cardinality, double-click the relationship line to open the Edit
Relationship dialog.
These features are meant for advanced data modeling and are outside the scope of this tutorial. To learn more, see
Bi-directional cross filters for tabular models in Analysis Services.
In some cases, you may need to create additional relationships between tables in your model to support certain
business logic. For this tutorial, you need to create three additional relationships between the FactInternetSales
table and the DimDate table.
To add new relationships between tables
1. In the model designer, in the FactInternetSales table, click, and hold on the OrderDate column, then drag
the cursor to the Date column in the DimDate table, and then release.
A solid line appears showing you have created an active relationship between the OrderDate column in
the Internet Sales table, and the Date column in the Date table.
NOTE
When creating relationships, the cardinality and filter direction between the primary table and the related lookup
table is automatically selected.
2. In the FactInternetSales table, click and hold on the DueDate column, then drag the cursor to the Date
column in the DimDate table, and then release.
A dotted line appears showing you have created an inactive relationship between the DueDate column in
the FactInternetSales table, and the Date column in the DimDate table. You can have multiple
relationships between tables, but only one relationship can be active at a time. Inactive relationships can be
made active to perform special aggregations in custom DAX expressions.
3. Finally, create one more relationship. In the FactInternetSales table, click and hold on the ShipDate
column, then drag the cursor to the Date column in the DimDate table, and then release.
What's next?
Lesson 5: Create calculated columns.
Create calculated columns
Premium
In this lesson, you create data in your model by adding calculated columns. You can create calculated columns (as
custom columns) when using Get Data, by using the Query Editor, or later in the model designer like you do here.
To learn more, see Calculated columns.
You create five new calculated columns in three different tables. The steps are slightly different for each task
showing there are several ways to create columns, rename them, and place them in various locations in a table.
This lesson is also where you first use Data Analysis Expressions (DAX). DAX is a special language for creating
highly customizable formula expressions for tabular models. In this tutorial, you use DAX to create calculated
columns, measures, and role filters. To learn more, see DAX in tabular models.
Prerequisites
in this lesson, you should have completed the previous lesson: Lesson 4: Create relationships.

Create a MonthCalendar calculated column in the DimDate table
1. Click the Model menu > Model View > Data View.
Calculated columns can only be created by using the model designer in Data View.
2. In the model designer, click the DimDate table (tab).
3. Right-click the CalendarQuarter column header, and then click Insert Column.
A new column named Calculated Column 1 is inserted to the left of the Calendar Quarter column.
4. In the formula bar above the table, type the following DAX formula: Auto-Complete helps you type the fully
qualified names of columns and tables, and lists the functions that are available.
=RIGHT(" " & FORMAT([MonthNumberOfYear],"#0"), 2) & " - " & [EnglishMonthName]
Values are then populated for all the rows in the calculated column. If you scroll down through the table,
you see rows can have different values for this column, based on the data in each row.
5. Rename this column to MonthCalendar.
The MonthCalendar calculated column provides a sortable name for Month.
Create a DayOfWeek calculated column in the DimDate table
1. With the DimDate table still active, click the Column menu, and then click Add Column.
2. In the formula bar, type the following formula:
=RIGHT(" " & FORMAT([DayNumberOfWeek],"#0"), 2) & " - " & [EnglishDayNameOfWeek]
When you've finished building the formula, press ENTER. The new column is added to the far right of the
table.
3. Rename the column to DayOfWeek.
4. Click the column heading, and then drag the column between the EnglishDayNameOfWeek column and
the DayNumberOfMonth column.
TIP
Moving columns in your table makes it easier to navigate.
The DayOfWeek calculated column provides a sortable name for the day of week.
Create a ProductSubcategoryName calculated column in the DimProduct table
1. In the DimProduct table, scroll to the far right of the table. Notice the right-most column is named Add
Column, click the column heading.
=RELATED('DimProductSubcategory'[EnglishProductSubcategoryName])
3. Rename the column to ProductSubcategoryName.

The ProductSubcategoryName calculated column is used to create a hierarchy in the DimProduct table, which
includes data from the EnglishProductSubcategoryName column in the DimProductSubcategory table.
Hierarchies cannot span more than one table. You create hierarchies later in Lesson 9.
Create a ProductCategoryName calculated column in the DimProduct table
1. With the DimProduct table still active, click the Column menu, and then click Add Column.
=RELATED('DimProductCategory'[EnglishProductCategoryName])
3. Rename the column to ProductCategoryName.

The ProductCategoryName calculated column is used to create a hierarchy in the DimProduct table, which
includes data from the EnglishProductCategoryName column in the DimProductCategory table. Hierarchies
cannot span more than one table.
Create a Margin calculated column in the FactInternetSales table
1. In the model designer, select the FactInternetSales table.
2. Create a new calculated column between the SalesAmount column and the TaxAmt column.
=[SalesAmount]-[TotalProductCost]
4. Rename the column to Margin.
The Margin calculated column is used to analyze profit margins for each sale.
What's next?
Lesson 6: Create measures.
Create measures
Premium
In this lesson, you create measures to be included in your model. Similar to the calculated columns you created, a
measure is a calculation created by using a DAX formula. However, unlike calculated columns, measures are
evaluated based on a user selected filter. For example, a particular column or slicer added to the Row Labels field
in a PivotTable. A value for each cell in the filter is then calculated by the applied measure. Measures are powerful,
flexible calculations that you want to include in almost all tabular models to perform dynamic calculations on
numerical data. To learn more, see Measures.
To create measures, you use the Measure Grid. By default, each table has an empty measure grid; however, you
typically do not create measures for every table. The measure grid appears below a table in the model designer
when in Data View. To hide or show the measure grid for a table, click the Table menu, and then click Show
Measure Grid.
You can create a measure by clicking an empty cell in the measure grid, and then typing a DAX formula in the
formula bar. When you click ENTER to complete the formula, the measure then appears in the cell. You can also
create measures using a standard aggregation function by clicking a column, and then clicking the AutoSum
button (∑) on the toolbar. Measures created using the AutoSum feature appear in the measure grid cell directly
beneath the column, but can be moved.
In this lesson, you create measures by both entering a DAX formula in the formula bar, and by using the AutoSum
feature.
Prerequisites
in this lesson, you should have completed the previous lesson: Lesson 5: Create calculated columns.
Create measures
To create a DaysCurrentQuarterToDate measure in the DimDate table
2. In the measure grid, click the top-left empty cell.
DaysCurrentQuarterToDate:=COUNTROWS( DATESQTD( 'DimDate'[Date]))
Notice the top-left cell now contains a measure name, DaysCurrentQuarterToDate, followed by the
result, 92. The result is not relevant at this point because no user filter has been applied.
Unlike calculated columns, with measure formulas you can type the measure name, followed by a colon,
followed by the formula expression.
To create a DaysInCurrentQuarter measure in the DimDate table
1. With the DimDate table still active in the model designer, in the measure grid, click the empty cell below
the measure you created.
DaysInCurrentQuarter:=COUNTROWS( DATESBETWEEN( 'DimDate'[Date], STARTOFQUARTER(

LASTDATE('DimDate'[Date])), ENDOFQUARTER('DimDate'[Date])))
When creating a comparison ratio between one incomplete period and the previous period. The formula
must calculate the proportion of the period that has elapsed and compare it to the same proportion in the
previous period. In this case, [DaysCurrentQuarterToDate]/[DaysInCurrentQuarter] gives the proportion
elapsed in the current period.
To create an InternetDistinctCountSalesOrder measure in the FactInternetSales table
1. Click the FactInternetSales table.
2. Click the SalesOrderNumber column heading.
3. On the toolbar, click the down-arrow next to the AutoSum (∑) button, and then select DistinctCount.
The AutoSum feature automatically creates a measure for the selected column using the DistinctCount
standard aggregation formula.
4. In the measure grid, click the new measure, and then in the Properties window, in Measure Name,
rename the measure to InternetDistinctCountSalesOrder.
To create additional measures in the FactInternetSales table
1. By using the AutoSum feature, create and name the following measures:
COLUMN MEASURE NAME AUTOSUM (∑) FORMULA
SalesOrderLineNumber InternetOrderLinesCount Count =COUNTA([SalesOrderLine

Number])
COLUMN MEASURE NAME AUTOSUM (∑) FORMULA
OrderQuantity InternetTotalUnits Sum =SUM([OrderQuantity])
DiscountAmount InternetTotalDiscountAmo Sum =SUM([DiscountAmount])

unt
TotalProductCost InternetTotalProductCost Sum =SUM([TotalProductCost])
SalesAmount InternetTotalSales Sum =SUM([SalesAmount])
Margin InternetTotalMargin Sum =SUM([Margin])
TaxAmt InternetTotalTaxAmt Sum =SUM([TaxAmt])
Freight InternetTotalFreight Sum =SUM([Freight])
2. By clicking an empty cell in the measure grid, and by using the formula bar, create, the following custom
measures in order:
InternetPreviousQuarterMargin:=CALCULATE([InternetTotalMargin],PREVIOUSQUARTER('DimDate'[Date]))
InternetCurrentQuarterMargin:=TOTALQTD([InternetTotalMargin],'DimDate'[Date])
InternetPreviousQuarterMarginProportionToQTD:=[InternetPreviousQuarterMargin]*
([DaysCurrentQuarterToDate]/[DaysInCurrentQuarter])
InternetPreviousQuarterSales:=CALCULATE([InternetTotalSales],PREVIOUSQUARTER('DimDate'[Date]))
InternetCurrentQuarterSales:=TOTALQTD([InternetTotalSales],'DimDate'[Date])
InternetPreviousQuarterSalesProportionToQTD:=[InternetPreviousQuarterSales]*
Measures created for the FactInternetSales table can be used to analyze critical financial data such as sales, costs,
and profit margin for items defined by the user selected filter.
What's next?
Lesson 7: Create Key Performance Indicators.
Create Key Performance Indicators
Premium
In this lesson, you create Key Performance Indicators (KPIs). KPIs are used to gauge performance of a value
defined by a Base measure, against a Target value also defined by a measure, or by an absolute value. In reporting
client applications, KPIs can provide business professionals a quick and easy way to understand a summary of
business success or to identify trends. To learn more, see KPIs
Prerequisites
in this lesson, you should have completed the previous lesson: Lesson 6: Create measures.

To create an InternetCurrentQuarterSalesPerformance KPI
1. In the model designer, click the FactInternetSales table.
2. In the measure grid, click an empty cell.
3. In the formula bar, above the table, type the following formula:
InternetCurrentQuarterSalesPerformance :=IF([InternetPreviousQuarterSalesProportionToQTD]<>0,
([InternetCurrentQuarterSales]-
[InternetPreviousQuarterSalesProportionToQTD])/[InternetPreviousQuarterSalesProportionToQTD],BLANK())
This measure serves as the Base measure for the KPI.

4. In the measure grid, right-click InternetCurrentQuarterSalesPerformance > Create KPI.
5. In the Key Performance Indicator (KPI) dialog box, in Target select Absolute Value, and then type 1.1.
6. In the left (low) slider field, type 1, and then in the right (high) slider field, type 1.07.
7. In Select Icon Style, select the diamond (red), triangle (yellow), circle (green) icon type.
TIP
Notice the expandable Descriptions label below the available icon styles. Use descriptions for the various KPI
elements to make them more identifiable in client applications.
8. Click OK to complete the KPI.

In the measure grid, notice the icon next to the InternetCurrentQuarterSalesPerformance measure. This
icon indicates that this measure serves as a Base value for a KPI.
To create an InternetCurrentQuarterMarginPerformance KPI
1. In the measure grid for the FactInternetSales table, click an empty cell.
InternetCurrentQuarterMarginPerformance :=IF([InternetPreviousQuarterMarginProportionToQTD]<>0,
([InternetCurrentQuarterMargin]-
[InternetPreviousQuarterMarginProportionToQTD])/[InternetPreviousQuarterMarginProportionToQTD],BLANK())
3. Right-click InternetCurrentQuarterMarginPerformance > Create KPI.

5. In the left (low) slider field, slide until the field displays 0.8, and then slide the right (high) slider field, until
the field displays 1.03.
6. In Select Icon Style, select the diamond (red), triangle (yellow), circle (green) icon type, and then click OK.
What's next?
Lesson 8: Create perspectives.
Create perspectives
Premium
In this lesson, you create an Internet Sales perspective. A perspective defines a viewable subset of a model that
provides focused, business-specific, or application-specific viewpoints. When a user connects to a model by using
a perspective, they see only those model objects (tables, columns, measures, hierarchies, and KPIs) as fields
defined in that perspective. To learn more, see Perspectives.
The Internet Sales perspective you create in this lesson excludes the DimCustomer table object. When you create a
perspective that excludes certain objects from view, that object still exists in the model. However, it is not visible in
a reporting client field list. Calculated columns and measures either included in a perspective or not can still
calculate from object data that is excluded.
The purpose of this lesson is to describe how to create perspectives and become familiar with the tabular model
authoring tools. If you later expand this model to include additional tables, you can create additional perspectives
to define different viewpoints of the model, for example, Inventory and Sales.
Estimated time to complete this lesson: Five minutes
Prerequisites
in this lesson, you should have completed the previous lesson: Lesson 7: Create Key Performance Indicators.
Create perspectives
To create an Internet Sales perspective
1. Click the Model menu > Perspectives > Create and Manage.
2. In the Perspectives dialog box, click New Perspective.
3. Double-click the New Perspective column heading, and then rename Internet Sales.
4. Select the all the tables except DimCustomer.
In a later lesson, you use the Analyze in Excel feature to test this perspective. The Excel PivotTable Fields List
includes each table except the DimCustomer table.
What's next?
Lesson 9: Create hierarchies.
Create hierarchies
Premium
In this lesson, you create hierarchies. Hierarchies are groups of columns arranged in levels. For example, a
Geography hierarchy might have sublevels for Country, State, County, and City. Hierarchies can appear separate
from other columns in a reporting client application field list, making them easier for client users to navigate and
include in a report. To learn more, see Hierarchies
To create hierarchies, use the model designer in Diagram View. Creating and managing hierarchies is not
supported in Data View.
Prerequisites
in this lesson, you should have completed the previous lesson: Lesson 8: Create perspectives.
Create hierarchies
To create a Category hierarchy in the DimProduct table
1. In the model designer (diagram view), right-click the DimProduct table > Create Hierarchy. A new
hierarchy appears at the bottom of the table window. Rename the hierarchy Category.
2. Click and drag the ProductCategoryName column to the new Category hierarchy.
3. In the Category hierarchy, right-click ProductCategoryName > Rename, and then type Category.
NOTE
Renaming a column in a hierarchy does not rename that column in the table. A column in a hierarchy is just a
representation of the column in the table.
4. Click and drag the ProductSubcategoryName column to the Category hierarchy. Rename it
Subcategory.
5. Right-click the ModelName column > Add to hierarchy, and then select Category. Rename it Model.
6. Finally, add EnglishProductName to the Category hierarchy. Rename it Product.
To create hierarchies in the DimDate table
1. In the DimDate table, create a hierarchy named Calendar.
2. Add the following columns in-order:
CalendarYear
CalendarSemester
CalendarQuarter
MonthCalendar
DayNumberOfMonth
3. In the DimDate table, create a Fiscal hierarchy. Include the following columns in-order:
FiscalYear
FiscalSemester
FiscalQuarter
MonthCalendar
DayNumberOfMonth
4. Finally, in the DimDate table, create a ProductionCalendar hierarchy. Include the following columns in-
order:
CalendarYear
WeekNumberOfYear
DayNumberOfWeek
What's next?
Lesson 10: Create partitions.
Create partitions
Premium
In this lesson, you create partitions to divide the FactInternetSales table into smaller logical parts that can be
processed (refreshed) independent of other partitions. By default, every table you include in your model has one
partition, which includes all the table's columns and rows. For the FactInternetSales table, we want to divide the
data by year; one partition for each of the table's five years. Each partition can then be processed independently. To
learn more, see Partitions.
Prerequisites
in this lesson, you should have completed the previous lesson: Lesson 9: Create Hierarchies.
Create partitions
To create partitions in the FactInternetSales table
1. In Tabular Model Explorer, expand Tables, and then right-click FactInternetSales > Partitions.
2. In Partition Manager, click Copy, and then change the name to FactInternetSales2010.
Because you want the partition to include only those rows within a certain period, for the year 2010, you
must modify the query expression.
3. Click Design to open Query Editor, and then click the FactInternetSales2010 query.
4. In preview, click the down arrow in the OrderDate column heading, and then click Date/Time Filters >
Between.
5. In the Filter Rows dialog box, in Show rows where: OrderDate, leave is after or equal to, and then in the
date field, enter 1/1/2010. Leave the And operator selected, then select is before, then in the date field,
enter 1/1/2011, and then click OK.
Notice in Query Editor, in APPLIED STEPS, you see another step named Filtered Rows. This filter is to
select only order dates from 2010.
6. Click Import.
In Partition Manager, notice the query expression now has an additional Filtered Rows clause.
This statement specifies this partition should include only the data in those rows where the OrderDate is in
the 2010 calendar year as specified in the filtered rows clause.
To create a partition for the 2011 year
1. In the partitions list, click the FactInternetSales2010 partition you created, and then click Copy. Change
the partition name to FactInternetSales2011.
You do not need to use Query Editor to create a new filtered rows clause. Because you created a copy of the
query for 2010, all you need to do is make a slight change in the query for 2011.
2. In Query Expression, in-order for this partition to include only those rows for the 2011 year, replace the
years in the Filtered Rows clause with 2011 and 2012, respectively, like:
let
Source = #"SQL/localhost;AdventureWorksDW2014",
dbo_FactInternetSales = Source{[Schema="dbo",Item="FactInternetSales"]}[Data],
#"Removed Columns" = Table.RemoveColumns(dbo_FactInternetSales,{"OrderDateKey", "DueDateKey",
"ShipDateKey"}),
#"Filtered Rows" = Table.SelectRows(#"Removed Columns", each [OrderDate] >= #datetime(2011, 1, 1,
0, 0, 0) and [OrderDate] < #datetime(2012, 1, 1, 0, 0, 0))
in
#"Filtered Rows"
To create partitions for 2012, 2013, and 2014.

Follow the previous steps, creating partitions for 2012, 2013, and 2014, changing the years in the Filtered
Rows clause to include only rows for that year.
Delete the FactInternetSales partition
Now that you have partitions for each year, you can delete the FactInternetSales partition; preventing overlap
when choosing Process all when processing partitions.
To delete the FactInternetSales partition
Click the FactInternetSales partition, and then click Delete.
Process partitions
In Partition Manager, notice the Last Processed column for each of the new partitions you created shows these
partitions have never been processed. When you create partitions, you should run a Process Partitions or Process
Table operation to refresh the data in those partitions.
To process the FactInternetSales partitions
1. Click OK to close Partition Manager.
2. Click the FactInternetSales table, then click the Model menu > Process > Process Partitions.
3. In the Process Partitions dialog box, verify Mode is set to Process Default.
4. Select the checkbox in the Process column for each of the five partitions you created, and then click OK.
If you're prompted for Impersonation credentials, enter the Windows user name and password you
specified in Lesson 2.
The Data Processing dialog box appears and displays process details for each partition. Notice that a
different number of rows for each partition are transferred. Each partition includes only those rows for the
year specified in the WHERE clause in the SQL Statement. When processing is finished, go ahead and close
the Data Processing dialog box.
What's next?
Go to the next lesson: Lesson 11: Create Roles.
Create roles
Premium
In this lesson, you create roles. Roles provide model database object and data security by limiting access to only
those users that are role members. Each role is defined with a single permission: None, Read, Read and Process,
Process, or Administrator. Roles can be defined during model authoring by using Role Manager. After a model
has been deployed, you can manage roles by using SQL Server Management Studio (SSMS ). To learn more, see
Roles.
NOTE
Creating roles is not necessary to complete this tutorial. By default, the account you are currently logged in with has
Administrator privileges on the model. However, for other users in your organization to browse by using a reporting client,
you must create at least one role with Read permissions and add those users as members.
You create three roles:

Sales Manager - This role can include users in your organization for which you want to have Read
permission to all model objects and data.
Sales Analyst US - This role can include users in your organization for which you want only to be able to
browse data related to sales in the United States. For this role, you use a DAX formula to define a Row
Filter, which restricts members to browse data only for the United States.
Administrator - This role can include users for which you want to have Administrator permission, which
allows unlimited access and permissions to perform administrative tasks on the model database.
Because Windows user and group accounts in your organization are unique, you can add accounts from your
particular organization to members. However, for this tutorial, you can also leave the members blank. You test the
effect of each role later in Lesson 12: Analyze in Excel.
Prerequisites
in this lesson, you should have completed the previous lesson: Lesson 10: Create partitions.
Create roles
To create a Sales Manager user role
1. In Tabular Model Explorer, right-click Roles > Roles.
2. In Role Manager, click New.
3. Click the new role, and then in the Name column, rename the role to Sales Manager.
4. In the Permissions column, click the dropdown list, and then select the Read permission.
5. Optional: Click the Members tab, and then click Add. In the Select Users or Groups dialog box, enter the
Windows users or groups from your organization you want to include in the role.
To create a Sales Analyst US user role
2. Rename the role to Sales Analyst US.
3. Give this role Read permission.
4. Click the Row Filters tab, and then for the DimGeography table only, in the DAX Filter column, type the
following formula:
=DimGeography[CountryRegionCode] = "US"
A Row Filter formula must resolve to a Boolean (TRUE/FALSE ) value. With this formula, you are specifying
that only rows with the Country Region Code value of "US" are visible to the user.
To create an Administrator user role
1. Click New.
2. Rename the role to Administrator.
3. Give this role Administrator permission.
What's next?
Lesson 12: Analyze in Excel.
Analyze in Excel
Premium
In this lesson, you use the Analyze in Excel feature to open Microsoft Excel, automatically create a connection to
the model workspace, and automatically add a PivotTable to the worksheet. The Analyze in Excel feature is meant
to provide a quick and easy way to test the efficacy of your model design prior to deploying your model. You do
not perform any data analysis in this lesson. The purpose of this lesson is to familiarize you, the model author, with
the tools you can use to test your model design.
To complete this lesson, Excel must be installed on the same computer as Visual Studio.
Prerequisites
in this lesson, you should have completed the previous lesson: Lesson 11: Create roles.
Browse using the Default and Internet Sales perspectives

In these first tasks, you browse your model by using both the default perspective, which includes all model objects,
and also by using the Internet Sales perspective you created earlier. The Internet Sales perspective excludes the
Customer table object.
To browse by using the Default perspective
1. Click the Model menu > Analyze in Excel.
2. In the Analyze in Excel dialog box, click OK.
Excel opens with a new workbook. A data source connection is created using the current user account and
the Default perspective is used to define viewable fields. A PivotTable is automatically added to the
worksheet.
3. In Excel, in the PivotTable Field List, notice the DimDate and FactInternetSales measure groups appear.
The DimCustomer, DimDate, DimGeography, DimProduct, DimProductCategory,
DimProductSubcategory, and FactInternetSales tables with their respective columns also appear.
4. Close Excel without saving the workbook.
To browse by using the Internet Sales perspective
1. Click the Model menu, and then click Analyze in Excel.
2. In the Analyze in Excel dialog box, leave Current Windows User selected, then in the Perspective drop-
down listbox, select Internet Sales, and then click OK.
3. In Excel, in PivotTable Fields, notice the DimCustomer table is excluded from the field list.
Browse by using roles

Roles are an important part of any tabular model. Without at least one role to which users are added as members,
users cannot access and analyze data using your model. The Analyze in Excel feature provides a way for you to
test the roles you have defined.
To browse by using the Sales Manager user role
1. In SSDT, click the Model menu, and then click Analyze in Excel.
2. In Specify the user name or role to use to connect to the model, select Role, and then in the drop-
down listbox, select Sales Manager, and then click OK.
Excel opens with a new workbook. A PivotTable is automatically created. The Pivot Table Field List includes
all the data fields available in your new model.
What's next?
Go to the next lesson: Lesson 13: Deploy.
Deploy
Premium
In this lesson, you configure deployment properties; specifying a server to deploy to and a name for the model.
You then deploy the model to the server. After your model is deployed, users can connect to it by using a reporting
client application. To learn more, see Deploy to Azure Analysis Services and Tabular model solution deployment.
Prerequisites
in this lesson, you should have completed the previous lesson: Lesson 12: Analyze in Excel.
IMPORTANT
If deploying to Azure Analysis Services, you must have Administrator permissions on the serever.
IMPORTANT
If you installed the AdventureWorksDW sample database on an on-premises SQL Server, and you're deploying your model
to an Azure Analysis Services server, an On-premises data gateway is required.
Deploy the model

To configure deployment properties
1. In Solution Explorer, right-click the AW Internet Sales project, and then click Properties.
2. In the AW Internet Sales Property Pages dialog box, under Deployment Server, in the Server
property, enter the full server name. If connecting to Azure Analysis Services, server name must inlcude the
full URL.
3. In the Database property, type Adventure Works Internet Sales.
4. In the Model Name property, type Adventure Works Internet Sales Model.
5. Verify your selections and then click OK.
To deploy the Adventure Works Internet Sales
1. In Solution Explorer, right-click the AW Internet Sales project > Build.
2. Right-click the AW Internet Sales project > Deploy.
When deploying to Azure Analysis Services, you may be prompted to enter your account. Enter your
organizational account and password, for example [email protected]. This account must be in
Admins on the server.
The Deploy dialog box appears and displays the deployment status of the metadata and each table included
in the model.
3. When deployment successfully completes, go ahead and click Close.
This lesson describes the most common and easiest method to deploy a tabular model from SSDT. Advanced
deployment options such as the Deployment Wizard or automating with XMLA and AMO provide greater
flexibility, consistency, and scheduled deployments. To learn more, see Tabular model solution deployment.
Conclusion
Congratulations! You're finished authoring and deploying your first Analysis Services Tabular model. This tutorial
has helped guide you through completing the most common tasks in creating a tabular model. Now that your
Adventure Works Internet Sales model is deployed, you can use SQL Server Management Studio to manage the
model; create process scripts and a backup plan. Users can also now connect to the model using a reporting client
application such as Microsoft Excel or Power BI.
What's next?
Connect with Power BI Desktop
Supplemental Lesson - Dynamic security
Supplemental Lesson - Detail rows
Supplemental Lesson - Ragged hierarchies
Supplemental lesson - Detail Rows
Premium
In this supplemental lesson, you use the DAX Editor to define a custom Detail Rows Expression. A Detail Rows
Expression is a property on a measure, providing end-users more information about the aggregated results of a
measure.
Prerequisites
This supplemental lesson article is part of a tabular modeling tutorial. Before performing the tasks in this
supplemental lesson, you should have completed all previous lessons or have a completed Adventure Works
Internet Sales sample model project.
What's the issue?

Let's look at the details of the InternetTotalSales measure, before adding a Detail Rows Expression.
1. In SSDT, click the Model menu > Analyze in Excel to open Excel and create a blank PivotTable.
2. In PivotTable Fields, add the InternetTotalSales measure from the FactInternetSales table to Values,
CalendarYear from the DimDate table to Columns, and EnglishCountryRegionName to Rows. The
PivotTable now gives an aggregated results from the InternetTotalSales measure by regions and year.
3. In the PivotTable, double-click an aggregated value for a year and a region name. Here we double-clicked
the value for Australia and the year 2014. A new sheet opens containing data, but not useful data.
What we want to see here is a table containing columns and rows of data that contribute to the aggregated result
of the InternetTotalSales measure. To do that, we can add a Detail Rows Expression as a property of the measure.
Add a Detail Rows Expression

To create a Detail Rows Expression
1. In the FactInternetSales table's measure grid, click the InternetTotalSales measure.
2. In Properties > Detail Rows Expression, click the editor button to open the DAX Editor.
3. In DAX Editor, enter the following expression:
SELECTCOLUMNS(
FactInternetSales,
"Sales Order Number", FactInternetSales[SalesOrderNumber],
"Customer First Name", RELATED(DimCustomer[FirstName]),
"Customer Last Name", RELATED(DimCustomer[LastName]),
"City", RELATED(DimGeography[City]),
"Order Date", FactInternetSales[OrderDate],
"Internet Total Sales", [InternetTotalSales]
)
This expression specifies names, columns, and measure results from the FactInternetSales table and related
tables are returned when a user double-clicks an aggregated result in a PivotTable or report.
4. Back in Excel, delete the sheet created in Step 3, then double-click an aggregated value. This time, with a
Detail Rows Expression property defined for the measure, a new sheet opens containing a lot more useful
data.
5. Redeploy your model.
See also
SELECTCOLUMNS Function (DAX)
Supplemental lesson - Dynamic security
Supplemental lesson - Ragged hierarchies
Supplemental lesson - Dynamic security
Premium
In this supplemental lesson, you create an additional role that implements dynamic security. Dynamic security
provides row-level security based on the user name or login id of the user currently logged on.
To implement dynamic security, you add a table to your model containing the user names of those users that can
connect to the model and browse model objects and data. The model you create using this tutorial is in the context
of Adventure Works; however, to complete this lesson, you must add a table containing users from your own
domain. You do not need the passwords for the user names that are added. To create an EmployeeSecurity table,
with a small sample of users from your own domain, you use the Paste feature, pasting employee data from an
Excel spreadsheet. In a real-world scenario, the table containing user names would typically be a table from an
actual database as a data source; for example, a real DimEmployee table.
To implement dynamic security, you use two DAX functions: USERNAME Function (DAX) and LOOKUPVALUE
Function (DAX). These functions, applied in a row filter formula, are defined in a new role. By using the
LOOKUPVALUE function, the formula specifies a value from the EmployeeSecurity table. The formula then passes
that value to the USERNAME function, which specifies the user name of the user logged on belongs to this role.
The user can then browse only data specified by the role's row filters. In this scenario, you specify that sales
employees can only browse Internet sales data for the sales territories in which they are a member.
Those tasks that are unique to this Adventure Works tabular model scenario, but would not necessarily apply to a
real-world scenario are identified as such. Each task includes additional information describing the purpose of the
task.
Prerequisites
This supplemental lesson article is part of a tabular modeling tutorial, which should be completed in order. Before
performing the tasks in this supplemental lesson, you should have completed all previous lessons.
Add the DimSalesTerritory table to the AW Internet Sales Tabular

Model Project
To implement dynamic security for this Adventure Works scenario, you must add two additional tables to your
model. The first table you add is DimSalesTerritory (as Sales Territory) from the same AdventureWorksDW
database. You later apply a row filter to the SalesTerritory table that defines the particular data the logged on user
can browse.
To add the DimSalesTerritory table
1. In Tabular Model Explorer > Data Sources, right-click your connection, and then click Import New
Tables.
If the Impersonation Credentials dialog box appears, type the impersonation credentials you used in Lesson
2: Add Data.
2. In Navigator, select the DimSalesTerritory table, and then click OK.
3. In Query Editor, click the DimSalesTerritory query, and then remove SalesTerritoryAlternateKey
column.
4. Click Import.
The new table is added to the model workspace. Objects and data from the source DimSalesTerritory table
are then imported into your AW Internet Sales Tabular Model.
5. After the table has been imported successfully, click Close.
Add a table with user name data

The DimEmployee table in the AdventureWorksDW sample database contains users from the AdventureWorks
domain. Those user names do not exist in your own environment. You must create a table in your model that
contains a small sample (at least three) of actual users from your organization. You then add these users as
members to the new role. You do not need the passwords for the sample user names, but you do need actual
Windows user names from your own domain.
To add an EmployeeSecurity table
1. Open Microsoft Excel, creating a worksheet.
2. Copy the following table, including the header row, and then paste it into the worksheet.
|EmployeeId|SalesTerritoryId|FirstName|LastName|LoginId|
|---------------|----------------------|--------------|-------------|------------|
|1|2|<user first name>|<user last name>|\<domain\username>|
3. Replace the first name, last name, and domain\username with the names and login ids of three users in
your organization. Put the same user on the first two rows, for EmployeeId 1, showing this user belongs to
more than one sales territory. Leave the EmployeeId and SalesTerritoryId fields as they are.
4. Save the worksheet as SampleEmployee.
5. In the worksheet, select all the cells with employee data, including the headers, then right-click the selected
data, and then click Copy.
6. In SSDT, click the Edit menu, and then click Paste.
If Paste is grayed out, click any column in any table in the model designer window, and try again.
7. In the Paste Preview dialog box, in Table Name, type EmployeeSecurity.
8. In Data to be pasted, verify the data includes all the user data and headers from the SampleEmployee
worksheet.
9. Verify Use first row as column headers is checked, and then click Ok.
A new table named EmployeeSecurity with employee data copied from the SampleEmployee worksheet is
created.
Create relationships between FactInternetSales, DimGeography, and

DimSalesTerritory table
The FactInternetSales, DimGeography, and DimSalesTerritory table all contain a common column,
SalesTerritoryId. The SalesTerritoryId column in the DimSalesTerritory table contains values with a different Id for
each sales territory.
To create relationships between the FactInternetSales, DimGeography, and the DimSalesTerritory table
1. In Diagram View, in the DimGeography table, click, and hold on the SalesTerritoryId column, then drag
the cursor to the SalesTerritoryId column in the DimSalesTerritory table, and then release.
2. In the FactInternetSales table, click, and hold on the SalesTerritoryId column, then drag the cursor to the
SalesTerritoryId column in the DimSalesTerritory table, and then release.
Notice the Active property for this relationship is False, meaning it's inactive. The FactInternetSales table
already has another active relationship.
Hide the EmployeeSecurity Table from client applications

In this task, you hide the EmployeeSecurity table, keeping it from appearing in a client application's field list. Keep
in mind that hiding a table does not secure it. Users can still query EmployeeSecurity table data if they know how.
To secure the EmployeeSecurity table data, preventing users from being able to query any of its data, you apply a
filter in a later task.
To hide the EmployeeSecurity table from client applications
In the model designer, in Diagram View, right-click the Employee table heading, and then click Hide from
Client Tools.
Create a Sales Employees by Territory user role

In this task, you create a user role. This role includes a row filter defining which rows of the DimSalesTerritory
table are visible to users. The filter is then applied in the one-to-many relationship direction to all other tables
related to DimSalesTerritory. You also apply a filter that secures the entire EmployeeSecurity table from being
queryable by any user that is a member of the role.
NOTE
The Sales Employees by Territory role you create in this lesson restricts members to browse (or query) only sales data for the
sales territory to which they belong. If you add a user as a member to the Sales Employees by Territory role that also exists
as a member in a role created in Lesson 11: Create Roles, you get a combination of permissions. When a user is a member of
multiple roles, the permissions, and row filters defined for each role are cumulative. That is, the user has the greater
permissions determined by the combination of roles.
To create a Sales Employees by Territory user role

1. In SSDT, click the Model menu, and then click Roles.
A new role with the None permission is added to the list.
3. Click the new role, and then in the Name column, rename the role to Sales Employees by Territory.
5. Click the Members tab, and then click Add.
6. In the Select User or Group dialog box, in Enter the object named to select, type the first sample user
name you used when creating the EmployeeSecurity table. Click Check Names to verify the user name is
valid, and then click Ok.
Repeat this step, adding the other sample user names you used when creating the EmployeeSecurity table.
7. Click the Row Filters tab.
8. For the EmployeeSecurity table, in the DAX Filter column, type the following formula:
=FALSE()
This formula specifies that all columns resolve to the false Boolean condition. No columns for the
EmployeeSecurity table can be queried by a member of the Sales Employees by Territory user role.
9. For the DimSalesTerritory table, type the following formula:
='DimSalesTerritory'[SalesTerritoryKey]=LOOKUPVALUE('EmployeeSecurity'[SalesTerritoryId],
'EmployeeSecurity'[LoginId], USERNAME(),
'EmployeeSecurity'[SalesTerritoryId], 'DimSalesTerritory'[SalesTerritoryKey])
In this formula, the LOOKUPVALUE function returns all values for the
DimEmployeeSecurity[SalesTerritoryId] column, where the EmployeeSecurity[LoginId] is the same as the
current logged on Windows user name, and EmployeeSecurity[SalesTerritoryId] is the same as the
DimSalesTerritory[SalesTerritoryId].
The set of sales territory IDs returned by LOOKUPVALUE is then used to restrict the rows shown in the
DimSalesTerritory table. Only rows where the SalesTerritoryID for the row is in the set of IDs returned by
the LOOKUPVALUE function are displayed.
10. In Role Manager, click Ok.
Test the Sales Employees by Territory User Role

In this task, you use the Analyze in Excel feature in SSDT to test the efficacy of the Sales Employees by Territory
user role. You specify one of the user names you added to the EmployeeSecurity table and as a member of the
role. This user name is then used as the effective user name in the connection created between Excel and the
model.
To test the Sales Employees by Territory user role
2. In the Analyze in Excel dialog box, in Specify the user name or role to use to connect to the model,
select Other Windows User, and then click Browse.
3. In the Select User or Group dialog box, in Enter the object name to select, type a user name you
included in the EmployeeSecurity table, and then click Check Names.
4. Click Ok to close the Select User or Group dialog box, and then click Ok to close the Analyze in Excel
dialog box.
Excel opens with a new workbook. A PivotTable is automatically created. The PivotTable Fields list includes
most of the data fields available in your new model.
Notice the EmployeeSecurity table is not visible in the PivotTable Fields list. You hid this table from client
tools in a previous task.
5. In the Fields list, in ∑ Internet Sales (measures), select the InternetTotalSales measure. The measure is
entered into the Values fields.
6. Select the SalesTerritoryId column from the DimSalesTerritory table. The column is entered into the
Row Labels fields.
Notice Internet sales figures appear only for the one region to which the effective user name you used
belongs. If you select another column, like City from the DimGeography table as Row Label field, only cities
in the sales territory to which the effective user belongs are displayed.
This user cannot browse or query any Internet sales data for territories other than the one they belong to.
This restriction is because the row filter defined for the DimSalesTerritory table, in the Sales Employees by
Territory user role, secures data for all data related to other sales territories.
See Also
USERNAME Function (DAX)
LOOKUPVALUE Function (DAX)
CUSTOMDATA Function (DAX)
Supplemental lesson - Ragged hierarchies
Premium
In this supplemental lesson, you resolve a common problem when pivoting on hierarchies that contain blank
values (members) at different levels. For example, an organization where a high-level manager has both
departmental managers and non-managers as direct reports. Or, geographic hierarchies composed of Country-
Region-City, where some cities lack a parent State or Province, such as Washington D.C., Vatican City. When a
hierarchy has blank members, it often descends to different, or ragged, levels.
Tabular models at the 1400 compatibility level have an additional Hide Members property for hierarchies. The
Default setting assumes there are no blank members at any level. The Hide blank members setting excludes
blank members from the hierarchy when added to a PivotTable or report.
Prerequisites
This supplemental lesson article is part of a tabular modeling tutorial. Before performing the tasks in this
supplemental lesson, you should have completed all previous lessons or have a completed Adventure Works
Internet Sales sample model project.
If you've created the AW Internet Sales project as part of the tutorial, your model does not yet contain any data or
hierarchies that are ragged. To complete this supplemental lesson, you first have to create the problem by adding
some additional tables, create relationships, calculated columns, a measure, and a new Organization hierarchy.
That part takes about 15 minutes. Then, you get to solve it in just a few minutes.
Add tables and objects

To add new tables to your model
1. In Tabular Model Explorer, expand Data Sources, then right-click your connection > Import New Tables.
2. In Navigator, select DimEmployee and FactResellerSales, and then click OK.
3. In Query Editor, click Import
4. Create the following relationships:
FILTER
TABLE 1 COLUMN DIRECTION TABLE 2 COLUMN ACTIVE
FactResellerSale OrderDateKey Default DimDate Date Yes

s
FactResellerSale DueDate Default DimDate Date No

s
FactResellerSale ShipDateKey Default DimDate Date No

s
FactResellerSale ProductKey Default DimProduct ProductKey Yes

s
FactResellerSale EmployeeKey To Both Tables DimEmployee EmployeeKey Yes

s
5. In the DimEmployee table, create the following calculated columns:

Path
=PATH([EmployeeKey],[ParentEmployeeKey])
FullName
=[FirstName] & " " & [MiddleName] & " " & [LastName]
Level1
=LOOKUPVALUE(DimEmployee[FullName],DimEmployee[EmployeeKey],PATHITEM([Path],1,1))
Level2
Level3
Level4
Level5
6. In the DimEmployee table, create a hierarchy named Organization. Add the following columns in-order:
Level1, Level2, Level3, Level4, Level5.
7. In the FactResellerSales table, create the following measure:
ResellerTotalSales:=SUM([SalesAmount])
8. Use Analyze in Excel to open Excel and automatically create a PivotTable.

9. In PivotTable Fields, add the Organization hierarchy from the DimEmployee table to Rows, and the
ResellerTotalSales measure from the FactResellerSales table to Values.
As you can see in the PivotTable, the hierarchy displays rows that are ragged. There are many rows where
blank members are shown.
To fix the ragged hierarchy by setting the Hide members property

1. In Tabular Model Explorer, expand Tables > DimEmployee > Hierarchies > Organization.
2. In Properties > Hide Members, select Hide blank members.
3. Back in Excel, refresh the PivotTable.

Now that looks a whole lot better!
See Also
Lesson 9: Create hierarchies
Supplemental Lesson - Dynamic security
Supplemental Lesson - Detail rows
Adventure Works Internet Sales tutorial (1200)
Premium
This tutorial provides lessons on how to create an Analysis Services tabular model at the 1200 compatibility level
by using SQL Server Data Tools (SSDT), and deploy your model to an Analysis Services server on-premises or in
Azure.
If you're using SQL Server 2017 or Azure Analysis Services, and you want to create your model at the 1400
compatibility level, use the Tabular modeling (1400 compatibility level). This updated version uses the modern Get
Data feature to connect and import source data, uses the M language to configure partitions, and includes
additional supplemental lessons.
IMPORTANT
You should create your tabular models at the latest compatibility level supported by your server. Later compatibility level
models provide improved performance, additional features, and will upgrade to future compatibility levels more seamlessly.
What you learn

How to create a new tabular model project in SSDT.
How to import data from a SQL Server relational database into a tabular model project.
How to create and manage relationships between tables in the model.
How to create and manage calculations, measures, and Key Performance Indicators that help users analyze
model data.
How to create and manage perspectives and hierarchies that help users more easily browse model data by
providing business and application-specific viewpoints.
How to create partitions dividing table data into smaller logical parts, that can be processed independent
from other partitions.
How to secure model objects and data by creating roles with user members.
How to deploy a tabular model to an Analysis Services server on-premises or in Azure.
Scenario
This tutorial is based on Adventure Works Cycles, a fictitious company. Adventure Works is a large, multinational
manufacturing company that produces bicycles, parts, and accessories for commercial markets in North America,
Europe, and Asia. With headquarters in Bothell, Washington, the company employs 500 workers. Additionally,
Adventure Works employs several regional sales teams throughout its market base.
To better support the data analysis needs of sales and marketing teams and of senior management, you are tasked
with creating a tabular model for users to analyze Internet sales data in the AdventureWorksDW sample database.
In order to complete the tutorial, and the Adventure Works Internet Sales tabular model, you must complete a
number of lessons. In each lesson is a number of tasks; completing each task in order is necessary for completing
the lesson. While in a particular lesson there may be several tasks that accomplish a similar outcome, but how you
complete each task is slightly different. This is to show that there is often more than one way to complete a
particular task, and to challenge you by using skills you've learned in previous tasks.
The purpose of the lessons is to guide you through authoring a basic tabular model running in In-Memory mode
by using many of the features included in SSDT. Because each lesson builds upon the previous lesson, you should
complete the lessons in order. Once you've completed all of the lessons, you have authored and deployed the
Adventure Works Internet Sales sample tabular model on an Analysis Services server.
This tutorial does not provide lessons or information about managing a deployed tabular model database by using
SQL Server Management Studio, or using a reporting client application to connect to a deployed model to browse
model data.
Prerequisites
In order to complete this tutorial, you need the following prerequisites:
The latest version of SSDT.
The latest version of SQL Server Management Studio. Get the latest version.
A client application such as Power BI Desktop or Excel.
A SQL Server instance with the Adventure Works DW sample database. This sample database includes the
data necessary to complete this tutorial. Get the latest version.
An Azure Analysis Services or SQL Server 2016 or later Analysis Services instance to deploy your model
to. Sign up for a free Azure Analysis Services trial.
Lessons
This tutorial includes the following lessons:
Lesson 1: Create a New Tabular Model Project 10 minutes
Lesson 2: Add Data 20 minutes
Lesson 3: Mark as Date Table 3 minutes
Lesson 4: Create Relationships 10 minutes
Lesson 5: Create Calculated Columns 15 minutes
Lesson 6: Create Measures 30 minutes
Lesson 7: Create Key Performance Indicators 15 minutes
Lesson 8: Create Perspectives 5 minutes
Lesson 9: Create Hierarchies 20 minutes
Lesson 10: Create Partitions 15 minutes
Lesson 11: Create Roles 15 minutes

Lesson 12: Analyze in Excel 20 minutes
Lesson 13: Deploy 5 minutes
Supplemental lessons
This tutorial also includes Supplemental Lessons. Topics in this section are not required to complete the tutorial,
but can be helpful in better understanding advanced tabular model authoring features.
Implement Dynamic Security by Using Row Filters 30 minutes
Configure Reporting Properties for Power View Reports 30 minutes
Next steps
To begin the tutorial, continue to the first lesson: Lesson 1: Create a New Tabular Model Project.
Lesson 1: Create a New Tabular Model Project
Premium
In this lesson, you will create a new, blank tabular model project in SQL Server Data Tools (SSDT). Once your new
project is created, you can begin adding data by using the Table Import Wizard. This lesson also gives you a brief
introduction to the tabular model authoring environment in SSDT.
Prerequisites
This topic is the first lesson in a tabular model authoring tutorial. To complete this lesson, you must have the
AdventureWorksDW sample database installed on a SQL Server instance. To learn more, see Tabular Modeling
(Adventure Works Tutorial).
Create a new tabular model project

To create a new tabular model project
1. In SSDT, on the File menu, click New > Project.
2. In the New Project dialog box, expand Installed > Business Intelligence > Analysis Services, and then
click Analysis Services Tabular Project.
3. In Name, type AW Internet Sales, and then specify a location for the project files.
By default, Solution Name will be the same as the project name; however, you can type a different solution
name.
4. Click OK.
5. In the Tabular model designer dialog box, select Integrated workspace.
The workspace will host a tabular model database with the same name as the project during model
authoring. Integrated workspace means SSDT will use a built-in instance, eliminating the need to install a
separate Analysis Services server instance just for model authoring. To learn more, see Workspace
Database.
6. In Compatibility level, verify SQL Server 2016 (1200) is selected, and then click OK.
If you don't see SQL Server 2016 RTM (1200) in the Compatibility level listbox, you're not using the latest
version of SQL Server Data Tools. To get the latest version, see Install SQL Server Data tools.
If you're using the latest version of SSDT, you can also choose SQL Server 2017 (1400). However, to
complete lesson 13: Deploy, you'll need a SQL Server 2017 or Azure server to deploy to.
Selecting an earlier compatibility level is only recommended if you intend on deploying your completed
tabular model to a different Analysis Services instance running an earlier version of SQL Server. Integrated
workspace is not supported for earlier compatibility levels. To learn more see, Compatibility level.
Understanding the SSDT tabular model authoring environment

Now that you've created a new tabular model project, let's take a moment to explore the tabular model authoring
environment in SSDT.
After your project is created, it opens in SSDT. On the right side, in Tabular Model Explorer, you'll see a tree
view of the objects in your model. Since you haven't yet imported data, the folders will be empty. You can right-
click an object folder to perform actions, similar to the menu bar. As you step through this tutorial, you'll use the
Tabular Model Explorer to navigate different objects in your model project.
Click the Solution Explorer tab. Here, you'll see your Model.bim file. If you don't see the designer window to the
left (the empty window with the Model.bim tab), in Solution Explorer, under AW Internet Sales Project,
double-click the Model.bim file. The Model.bim file contains all of the metadata for your model project.
Let's look at the model properties. Click Model.bim. In the Properties window, you'll see the model properties,
most important of which is the DirectQuery Mode property. This property specifies whether or not the model is
deployed in In-Memory mode (Off) or DirectQuery mode (On). For this tutorial, you will author and deploy your
model in In-Memory mode.
When you create a new model, certain model properties are set automatically according to the Data Modeling
settings that can be specified in the Tools > Options dialog box. Data Backup, Workspace Retention, and
Workspace Server properties specify how and where the workspace database (your model authoring database) is
backed up, retained in-memory, and built. You can change these settings later if necessary, but for now, just leave
these properties as they are.
In Solution Explorer, right-click AW Internet Sales (project), and then click Properties. The AW Internet
Sales Property Pages dialog box appears. These are the advanced project properties. You will set some of these
properties later when you are ready to deploy your model.
When you installed SSDT, several new menu items were added to the Visual Studio environment. Let's look at
those specific to authoring tabular models. Click on the Model menu. From here, you can launch the Table Import
Wizard, view and edit existing connections, refresh workspace data, browse your model in Excel with the Analyze
in Excel feature, create perspectives and roles, select the model view, and set calculation options.
Click on the Table menu. Here, you can create and manage relationships between tables, create and manage,
specify date table settings, create partitions, and edit table properties.
Click on the Column menu. Here, you can add and delete columns in a table, freeze columns, and specify sort
order. You can also use the AutoSum feature to create a standard aggregation measure for a selected column.
Other toolbar buttons provide quick access to frequently used features and commands.
Explore some of the dialogs and locations for various features specific to authoring tabular models. While some
items will not yet be active, you can get a good idea of the tabular model authoring environment.
Additional resources
To learn more about the different types of tabular model projects, see Tabular Model Projects. To learn more about
the tabular model authoring environment, see Tabular Model Designer.
What's next?
Go to the next lesson: Lesson 2: Add data.
Lesson 2: Add Data
Premium
In this lesson, you'll use the Table Import Wizard in SSDT to connect to the AdventureWorksDW SQL sample
database, select data, preview and filter the data, and then import the data into your model workspace.
By using the Table Import Wizard, you can import data from a variety of relational sources: Access, SQL, Oracle,
Sybase, Informix, DB2, Teradata, and more. The steps for importing data from each of these relational sources are
very similar to what is described below. Data can also be selected using a stored procedure. To learn more about
importing data and the different types of data sources you can import from, see Data Sources.
Prerequisites
This topic is part of a tabular modeling tutorial, which should be completed in order. Before performing the tasks
in this lesson, you should have completed the previous lesson: Lesson 1: Create a New Tabular Model Project.
Create a connection
To create a connection to a the AdventureWorksDW2014 database
1. In Tabular Model Explorer, right-click Data Sources > Import from Data Source.
This launches the Table Import Wizard, which guides you through setting up a connection to a data source.
If you don't see Tabular Model Explorer, double click Model.bim in Solution Explorer to open the model
in the designer.
Note: If you're creating your model at the 1400 compatibility level, you'll see the new Get Data experience
instead of the Table Import Wizard. The dialogs will appear a little different from the steps below, but you'll
still be able to follow along.
2. In the Table Import Wizard, under Relational Databases, click Microsoft SQL Server > Next.
3. In the Connect to a Microsoft SQL Server Database page, in Friendly Connection Name, type
Adventure Works DB from SQL.
4. In Server name, type the name of the server where you installed the AdventureWorksDW database.
5. In the Database name field, select AdventureWorksDW, and then click Next.
6. In the Impersonation Information page, you need to specify the credentials Analysis Services will use to
connect to the data source when importing and processing data. Verify Specific Windows user name
and password is selected, and then in User Name and Password, enter your Windows logon credentials,
and then click Next.
NOTE
Using a Windows user account and password provides the most secure method of connecting to a data source. For
more information, see Impersonation.
7. In the Choose How to Import the Data page, verify Select from a list of tables and views to choose
the data to import is selected. You want to select from a list of tables and views, so click Next to display a
list of all the source tables in the source database.
8. In the Select Tables and Views page, select the check box for the following tables: DimCustomer,
DimDate, DimGeography, DimProduct, DimProductCategory, DimProductSubcategory, and
FactInternetSales.
DO NOT click Finish.
Filter the table data

The DimCustomer table that you're importing from the sample database contains a subset of the data from the
original SQL Server Adventure Works database. You will filter out some more of the columns from the
DimCustomer table that aren't necessary when imported into your model. When possible, you'll want to filter out
data that won't be used in order to save in-memory space used by the model.
To filter the table data prior to importing
1. Select the row for the DimCustomer table, and then click Preview & Filter. The Preview Selected Table
window opens with all the columns in the DimCustomer source table displayed.
2. Clear the checkbox at the top of the following columns: SpanishEducation, FrenchEducation,
SpanishOccupation, FrenchOccupation.
Since the values for these columns are not relevant to Internet sales analysis, there is no need to import
these columns. Eliminating unnecessary columns will make your model smaller and more efficient.
3. Verify that all other columns are checked, and then click OK.
Notice the words Applied filters are now displayed in the Filter Details column in the DimCustomer
row; if you click on that link you'll see a text description of the filters you just applied.
4. Filter the remaining tables by clearing the checkboxes for the following columns in each table:
DimDate
COLUMN
DateKey
SpanishDayNameOfWeek
FrenchDayNameOfWeek
SpanishMonthName
FrenchMonthName
DimGeography
COLUMN
SpanishCountryRegionName
FrenchCountryRegionName
IpAddressLocator
DimProduct
COLUMN
SpanishProductName
FrenchProductName
FrenchDescription
ChineseDescription
ArabicDescription
HebrewDescription
ThaiDescription
GermanDescription
JapaneseDescription
TurkishDescription
DimProductCategory
COLUMN
SpanishProductCategoryName
FrenchProductCategoryName
DimProductSubcategory
COLUMN
SpanishProductSubcategoryName
FrenchProductSubcategoryName
FactInternetSales
COLUMN
OrderDateKey
DueDateKey
ShipDateKey
Import the selected tables and column data

Now that you've previewed and filtered out unnecessary data, you can import the rest of the data you do want.
The wizard imports the table data along with any relationships between tables. New tables and columns are
created in the model and data that you filtered out will not be imported.
To import the selected tables and column data
1. Review your selections. If everything looks okay, click Finish.
While importing the data, the wizard displays how many rows have been fetched. When all the data has
been imported, a message indicating success is displayed.
TIP
To see the relationships that were automatically created between the imported tables, on the Data preparation row,
click Details.
2. Click Close.
The wizard closes and the model designer now shows your imported tables.
Save your model project

It's important to frequently save your model project.
To save the model project
Click File > Save All.
What's next?
Go to the next lesson: Lesson 3: Mark as Date Table.
Lesson 3: Mark as Date Table
Premium
In Lesson 2: Add Data, you imported a dimension table named DimDate. While in your model this table is named
DimDate, it can also be known as a Date table, in that it contains date and time data.
Whenever you use DAX time-intelligence functions in calculations, as you'll do when you create measures a little
later, you must specify date table properties, which include a Date table and a unique identifier Date column in
that table.
In this lesson, you'll mark the DimDate table as the Date table and the Date column (in the Date table) as the Date
column (unique identifier).
Before we mark the date table and date column, we need to do a little housekeeping to make our model easier to
understand. You'll notice in the DimDate table a column named FullDateAlternateKey. It contains one row for
every day in each calendar year included in the table. We'll be using this column a lot in measure formulas and in
reports. But, FullDateAlternateKey is not really a good identifier for this column. We'll rename it to Date, making it
easier to identify and include in formulas. Whenever possible, it's a good idea to rename objects like tables and
columns to make them easier to identify in client reporting applications like Power BI and Excel.
Prerequisites
in this lesson, you should have completed the previous lesson: Lesson 2: Add data.
To rename the FullDateAlternateKey column
2. Double click the header for the FullDateAlternateKey column, and then rename it to Date.
To set Mark as Date Table
1. Select the Date column, and then in the Properties window, under Data Type, make sure Date is selected.
2. Click the Table menu, then click Date, and then click Mark as Date Table.
3. In the Mark as Date Table dialog box, in the Date listbox, select the Date column as the unique identifier.
It will usually be selected by default. Click OK.
What's next?
Go to the next lesson: Lesson 4: Create Relationships.
Lesson 4: Create Relationships
Premium
In this lesson, you will verify the relationships that were created automatically when you imported data and add
new relationships between different tables. A relationship is a connection between two tables that establishes how
the data in those tables should be correlated. For example, the DimProduct table and the DimProductSubcategory
table have a relationship based on the fact that each product belongs to a subcategory. To learn more, see
Relationships.
Prerequisites
in this lesson, you should have completed the previous lesson: Lesson 3: Mark as Date Table.
Review existing relationships and add new relationships

When you imported data by using the Table Import Wizard, you got seven tables from the AdventureWorksDW
database. Generally, when you import data from a relational source, existing relationships are automatically
imported together with the data. However, before you proceed with authoring your model you should verify those
relationships between tables were created properly. For this tutorial, you will also add three new relationships.
To review existing relationships
1. Click the Model menu > Model View > Diagram View.
The model designer now appears in Diagram View, a graphical format displaying all of the tables you
imported with lines between them. The lines between tables indicate the relationships that were
automatically created when you imported the data.
Use the minimap controls in the lower-right corner of the model designer to adjust the view to include as
many of the tables as possible. You can also click, and drag tables to different locations, bringing tables
closer together, or putting them in a particular order. Moving tables does not affect the relationships already
between the tables. To view all of the columns in a particular table, click, and drag on a table edge to expand
or make it smaller.
2. Click the solid line between the DimCustomer table and the DimGeography table. The solid line between
these two tables show this relationship is active, that is, it is used by default when calculating DAX formulas.
Notice the GeographyKey column in the DimCustomer table and the GeographyKey column in the
DimGeography table now both each appear within a box. This show these are the columns used in the
relationship. The relationship's properties now also appear in the Properties window.
TIP
In addition to using the model designer in diagram view, you can also use the Manage Relationships dialog box to
show the relationships between all tables in a table format. Right-click Relationships in Tabular Model Explorer, and
then click Manage Relationships. The Manage Relationships dialog box show the relationships that were
automatically created when you imported data.
3. Use the model designer in diagram view, or the Manage Relationships dialog box, to verify the following
relationships were created when each of the tables were imported from the AdventureWorksDW database:
ACTIVE TABLE RELATED LOOKUP TABLE
Yes DimCustomer [GeographyKey] DimGeography [GeographyKey]
Yes DimProduct DimProductSubcategory

[ProductSubcategoryKey] [ProductSubcategoryKey]
Yes DimProductSubcategory DimProductCategory

[ProductCategoryKey] [ProductCategoryKey]
Yes FactInternetSales [CustomerKey] DimCustomer [CustomerKey]
Yes FactInternetSales [ProductKey] DimProduct [ProductKey]
If any of the relationships in the table above are missing, verify that your model includes the following
tables: DimCustomer, DimDate, DimGeography, DimProduct, DimProductCategory,
DimProductSubcategory, and FactInternetSales. If tables from the same data source connection are
imported at separate times, any relationships between those tables will not be created and must be created
manually.
Take a closer look
In Diagram View, you'll notice an arrow, an asterisk, and a number on the lines that show the relationship between
tables.
The arrow shows the filter direction, the asterisk shows this table is the many side in the relationship's cardinality,
and the 1 shows this table is the one side of the relationship. If you need to edit a relationship; for example, change
the relationship's filter direction or cardinality, double-click the relationship line in Diagram View to open the Edit
Relationship dialog.
Most likely, you will never need to edit a relationship. These features are meant for advanced data modeling and
are outside the scope of this tutorial. To learn more, see Bi-directional cross filters for tabular models in SQL
Server 2016 Analysis Services.
In some cases, you may need to create additional relationships between tables in your model to support certain
business logic. For this tutorial, you need to create three additional relationships between the FactInternetSales
table and the DimDate table.
To add new relationships between tables
1. In the model designer, in the FactInternetSales table, click, and hold on the OrderDate column, then drag
the cursor to the Date column in the DimDate table, and then release.
A solid line appears showing you have created an active relationship between the OrderDate column in the
Internet Sales table and the Date column in the Date table.
NOTE
When creating relationships, the cardinality and filter direction between the primary table and the related lookup
table is automatically selected.
2. In the FactInternetSales table, click, and hold on the DueDate column, then drag the cursor to the Date
column in the DimDate table, and then release.
A dotted line appears showing you have created an inactive relationship between the DueDate column in
the FactInternetSales table and the Date column in the DimDate table. You can have multiple
relationships between tables, but only one relationship can be active at a time.
3. Finally, create one more relationship; in the FactInternetSales table, click, and hold on the ShipDate
column, then drag the cursor to the Date column in the DimDate table, and then release.
What's next?
Go to the next lesson: Lesson 5: Create Calculated Columns.
Lesson 5: Create Calculated Columns
Premium
In this lesson, you will create new data in your model by adding calculated columns. A calculated column is based
on data that already exists in the model. To learn more, see Calculated Columns.
You will create five new calculated columns in three different tables. The steps are slightly different for each task.
This is to show you there are several ways to create new columns, rename them, and place them in various
locations in a table.
Prerequisites
in this lesson, you should have completed the previous lesson: Lesson 4: Create Relationships.

Create a MonthCalendar calculated column in the DimDate table
1. Click the Model menu > Model View > Data View.
Calculated columns can only be created by using the model designer in Data View.
2. In the model designer, click the DimDate table (tab).
3. Right-click the CalendarQuarter column header, and then click Insert Column.
A new column named Calculated Column 1 is inserted to the left of the Calendar Quarter column.
4. In the formula bar above the table, type the following formula. AutoComplete helps you type the fully
qualified names of columns and tables, and lists the functions that are available.
=RIGHT(" " & FORMAT([MonthNumberOfYear],"#0"), 2) & " - " & [EnglishMonthName]
Values are then populated for all the rows in the calculated column. If you scroll down through the table,
you will see that rows can have different values for this column, based on the data that is in each row.
5. Rename this column to MonthCalendar.
The MonthCalendar calculated column provides a sortable name for Month.

Create a DayOfWeek calculated column in the DimDate table
1. With the DimDate table still active, click on the Column menu, and then click Add Column.
=RIGHT(" " & FORMAT([DayNumberOfWeek],"#0"), 2) & " - " & [EnglishDayNameOfWeek]
When you've finished building the formula, press ENTER. The new column is added to the far right of the
table.
3. Rename the column to DayOfWeek.
4. Click on the column heading, and then drag the column between the EnglishDayNameOfWeek column
and the DayNumberOfMonth column.
TIP
Moving columns in your table makes it easier to navigate.
The DayOfWeek calculated column provides a sortable name for the day of week.
Create a ProductSubcategoryName calculated column in the DimProduct table
1. In the DimProduct table, scroll to the far right of the table. Notice the right-most column is named Add
Column (italicized), click the column heading.
2. In the formula bar, type the following formula.
=RELATED('DimProductSubcategory'[EnglishProductSubcategoryName])
3. Rename the column to ProductSubcategoryName.

The ProductSubcategoryName calculated column is used to create a hierarchy in the DimProduct table which
includes data from the EnglishProductSubcategoryName column in the DimProductSubcategory table.
Hierarchies cannot span more than one table. You will create hierarchies later in Lesson 9.
Create a ProductCategoryName calculated column in the DimProduct table
1. With the DimProduct table still active, click the Column menu, and then click Add Column.
=RELATED('DimProductCategory'[EnglishProductCategoryName])
3. Rename the column to ProductCategoryName.

The ProductCategoryName calculated column is used to create a hierarchy in the DimProduct table which
includes data from the EnglishProductCategoryName column in the DimProductCategory table. Hierarchies
cannot span more than one table.
Create a Margin calculated column in the FactInternetSales table
1. In the model designer, select the FactInternetSales table.
2. Add a new column.
=[SalesAmount]-[TotalProductCost]
4. Rename the column to Margin.

5. Drag the column between the SalesAmount column and the TaxAmt column.
The Margin calculated column is used to analyze profit margins for each sale.
What's next?
Go to the next lesson: Lesson 6: Create Measures.
Lesson 6: Create Measures
Premium
In this lesson, you will create measures to be included in your model. Similar to the calculated columns you
created in the previous lesson, a measure is a calculation created by using a DAX formula. However, unlike
calculated columns, measures are evaluated based on a user selected filter; for example, a particular column or
slicer added to the Row Labels field in a PivotTable. A value for each cell in the filter is then calculated by the
applied measure. Measures are powerful, flexible calculations that you will want to include in almost all tabular
models to perform dynamic calculations on numerical data. To learn more, see Measures.
To create measures, you will use the Measure Grid. By default, each table has an empty measure grid; however, you
typically will not create measures for every table. The measure grid appears below a table in the model designer
when in Data View. To hide or show the measure grid for a table, click the Table menu, and then click Show
Measure Grid.
You can create a measure by clicking on an empty cell in the measure grid, and then typing a DAX formula in the
formula bar. When you click ENTER to complete the formula, the measure will then appear in the cell. You can also
create measures using a standard aggregation function by clicking on a column, and then clicking on the AutoSum
button (∑) on the toolbar. Measures created using the AutoSum feature will appear in the measure grid cell
directly beneath the column, but can be moved.
In this lesson, you will create measures by both entering a DAX formula in the formula bar and by using the
AutoSum feature.
Prerequisites
in this lesson, you should have completed the previous lesson: Lesson 5: Create Calculated Columns.
Create measures
To create a DaysCurrentQuarterToDate measure in the DimDate table
2. In the measure grid, click the top-left empty cell.
DaysCurrentQuarterToDate:=COUNTROWS( DATESQTD( 'DimDate'[Date]))
Notice the top-left cell now contains a measure name, DaysCurrentQuarterToDate, followed by the
result, 92.
Unlike calculated columns, with measure formulas you can type the measure name, followed by a comma,
followed by the formula expression.
To create a DaysInCurrentQuarter measure in the DimDate table
1. With the DimDate table still active in the model designer, in the measure grid, click the empty cell below
the measure you just created.
DaysInCurrentQuarter:=COUNTROWS( DATESBETWEEN( 'DimDate'[Date], STARTOFQUARTER(

LASTDATE('DimDate'[Date])), ENDOFQUARTER('DimDate'[Date])))
When creating a comparison ratio between one incomplete period and the previous period; the formula
must take into account the proportion of the period that has elapsed, and compare it to the same
proportion in the previous period. In this case, [DaysCurrentQuarterToDate]/[DaysInCurrentQuarter] gives
the proportion elapsed in the current period.
To create an InternetDistinctCountSalesOrder measure in the FactInternetSales table
1. Click the FactInternetSales table.
2. Click on the SalesOrderNumber column heading.
3. On the toolbar, click the down-arrow next to the AutoSum (∑) button, and then select DistinctCount.
The AutoSum feature automatically creates a measure for the selected column using the DistinctCount
standard aggregation formula.
4. In the measure grid, click the new measure, and then in the Properties window, in Measure Name,
rename the measure to InternetDistinctCountSalesOrder.
To create additional measures in the FactInternetSales table
1. By using the AutoSum feature, create and name the following measures:
MEASURE NAME COLUMN AUTOSUM (∑) FORMULA
InternetOrderLinesCount SalesOrderLineNumber Count =COUNTA([SalesOrderLine

Number])
MEASURE NAME COLUMN AUTOSUM (∑) FORMULA
InternetTotalUnits OrderQuantity Sum =SUM([OrderQuantity])
InternetTotalDiscountAmo DiscountAmount Sum =SUM([DiscountAmount])

unt
InternetTotalProductCost TotalProductCost Sum =SUM([TotalProductCost])
InternetTotalSales SalesAmount Sum =SUM([SalesAmount])
InternetTotalMargin Margin Sum =SUM([Margin])
InternetTotalTaxAmt TaxAmt Sum =SUM([TaxAmt])
InternetTotalFreight Freight Sum =SUM([Freight])
2. By clicking on an empty cell in the measure grid, and by using the formula bar, create and name the
following measures in order:
InternetPreviousQuarterMargin:=CALCULATE([InternetTotalMargin],PREVIOUSQUARTER('DimDate'[Date]))
InternetCurrentQuarterMargin:=TOTALQTD([InternetTotalMargin],'DimDate'[Date])
InternetPreviousQuarterMarginProportionToQTD:=[InternetPreviousQuarterMargin]*
InternetPreviousQuarterSales:=CALCULATE([InternetTotalSales],PREVIOUSQUARTER('DimDate'[Date]))
InternetCurrentQuarterSales:=TOTALQTD([InternetTotalSales],'DimDate'[Date])
InternetPreviousQuarterSalesProportionToQTD:=[InternetPreviousQuarterSales]*
Measures created for the FactInternetSales table can be used to analyze critical financial data such as sales, costs,
and profit margin for items defined by the user selected filter.
What's next?
Go to the next lesson: Lesson 7: Create Key Performance Indicators.
Lesson 7: Create Key Performance Indicators
Premium
In this lesson, you will create Key Performance Indicators (KPIs). KPIs are used to gauge performance of a value,
defined by a Base measure, against a Target value, also defined by a measure or by an absolute value. In reporting
client applications, KPIs can provide business professionals a quick and easy way to understand a summary of
business success or to identify trends. To learn more, see KPIs.
Prerequisites
in this lesson, you should have completed the previous lesson: Lesson 6: Create Measures.

To create an InternetCurrentQuarterSalesPerformance KPI
1. In the model designer, click the FactInternetSales table (tab).
2. In the measure grid, click an empty cell.
InternetCurrentQuarterSalesPerformance
:=IFERROR([InternetCurrentQuarterSales]/[InternetPreviousQuarterSalesProportionToQTD],BLANK())
This measure will serve as the Base measure for the KPI.
4. Right-click InternetCurrentQuarterSalesPerformance > Create KPI.
6. In the left (low) slider field, type 1, and then in the right (high) slider field, type 1.07.
7. In Select Icon Style, select the diamond (red), triangle (yellow), circle (green) icon type.
TIP
Notice the expandable Descriptions label below the available icon styles. Use this to enter descriptions for the
various KPI elements to make them more identifiable in client applications.
8. Click OK to complete the KPI.

In the measure grid, notice the icon next to the InternetCurrentQuarterSalesPerformance measure. This
icon indicates that this measure serves as a Base value for a KPI.
To create an InternetCurrentQuarterMarginPerformance KPI
1. In the measure grid for the FactInternetSales table, click an empty cell.
InternetCurrentQuarterMarginPerformance :=IF([InternetPreviousQuarterMarginProportionToQTD]<>0,
([InternetCurrentQuarterMargin]-
[InternetPreviousQuarterMarginProportionToQTD])/[InternetPreviousQuarterMarginProportionToQTD],BLANK())
3. Right-click InternetCurrentQuarterMarginPerformance > Create KPI.

5. In Define Status Thresholds, slide the left (low) slider field until the field displays 0.8, and then slide the
right (high) slider field, until the field displays 1.03.
6. In Select Icon Style, select the diamond (red), triangle (yellow), circle (green) icon type, and then click OK.
What's next?
Go to the next lesson: Lesson 8: Create Perspectives.
Lesson 8: Create Perspectives
Premium
In this lesson, you will create an Internet Sales perspective. A perspective defines a viewable subset of a model that
provides focused, business-specific, or application-specific viewpoints. When a user connects to a model by using
a perspective, they see only those model objects (tables, columns, measures, hierarchies, and KPIs) as fields
defined in that perspective.
The Internet Sales perspective you create in this lesson will exclude the DimCustomer table object. When you
create a perspective that excludes certain objects from view, that object still exists in the model; however, it is not
visible in a reporting client field list. Calculated columns and measures either included in a perspective or not can
still calculate from object data that is excluded.
The purpose of this lesson is to describe how to create perspectives and become familiar with the tabular model
authoring tools. If you later expand this model to include additional tables, you can create additional perspectives
to define different viewpoints of the model, for example, Inventory and Sales. To learn more, see Perspectives.
Prerequisites
in this lesson, you should have completed the previous lesson: Lesson 7: Create Key Performance Indicators.
Create perspectives
To create an Internet Sales perspective
1. Click the Model menu > Perspectives > Create and Manage.
2. In the Perspectives dialog box, click New Perspective.
3. Double-click the New Perspective column heading, and then rename Internet Sales.
4. Select the all of the tables except DimCustomer.
In a later lesson, you will use the Analyze in Excel feature to test this perspective. The Excel PivotTable Fields
List will include each table except the DimCustomer table.
What's next?
Go to the next lesson: Lesson 9: Create Hierarchies.
Lesson 9: Create Hierarchies
Premium
In this lesson, you will create hierarchies. Hierarchies are groups of columns arranged in levels; for example, a
Geography hierarchy might have sub-levels for Country, State, County, and City. Hierarchies can appear separate
from other columns in a reporting client application field list, making them easier for client users to navigate and
include in a report. To learn more, see Hierarchies.
To create hierarchies, you'll use the model designer in Diagram View. Creating and managing hierarchies is not
supported in Data View.
Prerequisites
in this lesson, you should have completed the previous lesson: Lesson 8: Create Perspectives.
Create hierarchies
To create a Category hierarchy in the DimProduct table
1. In the model designer (diagram view), right-click the DimProduct table > Create Hierarchy. A new
hierarchy appears at the bottom of the table window. Rename the hierarchy Category.
2. Click and drag the ProductCategoryName column to the new Category hierarchy.
3. In the Category hierarchy, right-click the ProductCategoryName > Rename, and then type Category.
NOTE
Renaming a column in a hierarchy does not rename that column in the table. A column in a hierarchy is just a
representation of the column in the table.
4. Click and drag the ProductSubcategoryName column to the Category hierarchy. Rename it
Subcategory.
5. Right-click the ModelName column > Add to hierarchy, and then select Category. Do the same for
EnglishProductName. Rename these columns in the hierarchy Model and Product.
To create hierarchies in the DimDate table
1. In the DimDate table, create a new hierarchy named Calendar.
2. Add the following columns in-order:
CalendarYear
CalendarSemester
CalendarQuarter
MonthCalendar
DayNumberOfMonth
3. In the DimDate table, create a Fiscal hierarchy. Include the following columns:
FiscalYear
FiscalSemester
FiscalQuarter
MonthCalendar
DayNumberOfMonth
4. Finally, in the DimDate table, create a ProductionCalendar hierarchy. Include the following columns:
CalendarYear
WeekNumberOfYear
DayNumberOfWeek
What's next?
Go to the next lesson: Lesson 10: Create Partitions.
Lesson 10: Create Partitions
Premium
In this lesson, you will create partitions to divide the FactInternetSales table into smaller logical parts that can be
processed (refreshed) independent of other partitions. By default, every table you include in your model has one
partition which includes all of the table's columns and rows. For the FactInternetSales table, we want to divide the
data by year; one partition for each of the table's five years. Each partition can then be processed independently. To
learn more, see Partitions.
Prerequisites
in this lesson, you should have completed the previous lesson: Lesson 9: Create Hierarchies.
Create partitions
To create partitions in the FactInternetSales table
1. In Tabular Model Explorer, expand Tables, right-click FactInternetSales > Partitions.
2. In the Partition Manager dialog box, click Copy.
3. In Partition Name, change the name to FactInternetSales2010.
TIP
Notice the column names in the Table Preview window display those columns included in the model table (checked)
with the column names from the source. This is because the Table Preview window displays columns from the source
table, not from the model table.
4. Select the SQL button just above the right side of the preview window to open the SQL Statement editor.
Because you want the partition to include only those rows within a certain period, you must include a
WHERE clause. You can only create a WHERE clause by using a SQL Statement.
5. In the SQL Statement field, replace the existing statement by copying and pasting the following statement:
SELECT
[dbo].[FactInternetSales].[ProductKey],
[dbo].[FactInternetSales].[CustomerKey],
[dbo].[FactInternetSales].[PromotionKey],
[dbo].[FactInternetSales].[CurrencyKey],
[dbo].[FactInternetSales].[SalesTerritoryKey],
[dbo].[FactInternetSales].[SalesOrderNumber],
[dbo].[FactInternetSales].[SalesOrderLineNumber],
[dbo].[FactInternetSales].[RevisionNumber],
[dbo].[FactInternetSales].[OrderQuantity],
[dbo].[FactInternetSales].[UnitPrice],
[dbo].[FactInternetSales].[ExtendedAmount],
[dbo].[FactInternetSales].[UnitPriceDiscountPct],
[dbo].[FactInternetSales].[DiscountAmount],
[dbo].[FactInternetSales].[ProductStandardCost],
[dbo].[FactInternetSales].[TotalProductCost],
[dbo].[FactInternetSales].[SalesAmount],
[dbo].[FactInternetSales].[TaxAmt],
[dbo].[FactInternetSales].[Freight],
[dbo].[FactInternetSales].[CarrierTrackingNumber],
[dbo].[FactInternetSales].[CustomerPONumber],
[dbo].[FactInternetSales].[OrderDate],
[dbo].[FactInternetSales].[DueDate],
[dbo].[FactInternetSales].[ShipDate]
FROM [dbo].[FactInternetSales]
WHERE (([OrderDate] >= N'2010-01-01 00:00:00') AND ([OrderDate] < N'2011-01-01 00:00:00'))
This statement specifies the partition should include all of the data in those rows where the OrderDate is
for the 2010 calendar year as specified in the WHERE clause.
6. Click Validate.
1. In the partitions list, click the FactInternetSales2010 partition you just created, and then click Copy.
2. In Partition Name, type FactInternetSales2011.
3. In the SQL Statement, in-order for the partition to include only those rows for the 2011 year, replace the
WHERE clause with the following:

Follow the steps above, using the following WHERE clause.


Delete the FactInternetSales partition
Now that you have partitions for each year, you can delete the FactInternetSales partition. This prevents overlap
when choosing Process all when processing partitions.
To delete the FactInternetSales partition
Click the FactInternetSales partition, and then click Delete.
Process partitions
In Partition Manager, notice the Last Processed column for each of the new partitions you just created shows
these partitions have never been processed. When you create new partitions, you should run a Process Partitions
or Process Table operation to refresh the data in those partitions.
To process the FactInternetSales partitions
1. Click OK to close the Partition Manager dialog box.
2. Click the FactInternetSales table, then click the Model menu > Process > Process Partitions.
3. In the Process Partitions dialog box, verify Mode is set to Process Default.
4. Select the checkbox in the Process column for each of the five partitions you created, and then click OK.
If you're prompted for Impersonation credentials, enter the Windows user name and password you
specified in Lesson 2.
The Data Processing dialog box appears and displays process details for each partition. Notice that a
different number of rows for each partition are transferred. This is because each partition includes only
those rows for the year specified in the WHERE clause in the SQL Statement. When processing is finished,
go ahead and close the Data Processing dialog box.
What's next?
Go to the next lesson: Lesson 11: Create Roles.
Lesson 11: Create Roles
Premium
In this lesson, you will create roles. Roles provide model database object and data security by limiting access to
only those Windows users which are role members. Each role is defined with a single permission: None, Read,
Read and Process, Process, or Administrator. Roles can be defined during model authoring by using Role
Manager. After a model has been deployed, you can manage roles by using SQL Server Management Studio. To
learn more, see Roles.
NOTE
Creating roles is not necessary to complete this tutorial. By default, the account you are currently logged in with will have
Administrator privileges on the model. However, to allow other users in your organization to browse the model by using a
reporting client, you must create at least one role with Read permissions and add those users as members.
You will create three roles:

Sales Manager - This role can include users in your organization for which you want to have Read
permission to all model objects and data.
Sales Analyst US - This role can include users in your organization for which you want only to be able to
browse data related to sales in the United States. For this role, you will use a DAX formula to define a Row
Filter, which restricts members to browse data only for the United States.
Administrator - This role can include users for which you want to have Administrator permission, which
allows unlimited access and permissions to perform administrative tasks on the model database.
Because Windows user and group accounts in your organization are unique, you can add accounts from your
particular organization to members. However, for this tutorial, you can also leave the members blank. You will still
be able to test the effect of each role later in Lesson 12: Analyze in Excel.
Prerequisites
in this lesson, you should have completed the previous lesson: Lesson 10: Create Partitions.
Create roles
To create a Sales Manager user role
1. In Tabular Model Explorer, right-click Roles > Roles.
3. Click on the new role, and then in the Name column, rename the role to Sales Manager.
To create a Sales Analyst US user role
2. Rename the role to Sales Analyst US.
3. Give this role Read permission.
4. Click on the Row Filters tab, and then for the DimGeography table only, in the DAX Filter column, type the
following formula:
=DimGeography[CountryRegionCode] = "US"
A Row Filter formula must resolve to a Boolean (TRUE/FALSE ) value. With this formula, you are specifying
that only rows with the Country Region Code value of "US" be visible to the user.
5. Optional: Click on the Members tab, and then click Add. In the Select Users or Groups dialog box, enter
the Windows users or groups from your organization you want to include in the role.
To create an Administrator user role
1. Click New.
2. Rename the role to Administrator.
3. Give this role Administrator permission.
4. Optional: Click on the Members tab, and then click Add. In the Select Users or Groups dialog box, enter
the Windows users or groups from your organization you want to include in the role.
What's next?
Go to the next lesson: Lesson 12: Analyze in Excel.
Lesson 12: Analyze in Excel
Premium
In this lesson, you will use the Analyze in Excel feature in SSDT to open Microsoft Excel, automatically create a
data source connection to the model workspace, and automatically add a PivotTable to the worksheet. The Analyze
in Excel feature is meant to provide a quick and easy way to test the efficacy of your model design prior to
deploying your model. You will not perform any data analysis in this lesson. The purpose of this lesson is to
familiarize you, the model author, with the tools you can use to test your model design. Unlike using the Analyze in
Excel feature, which is meant for model authors, end-users will use client reporting applications like Excel or Power
BI to connect to and browse deployed model data.
In order to complete this lesson, Excel must be installed on the same computer as SSDT. To learn more, see
Analyze in Excel.
Prerequisites
in this lesson, you should have completed the previous lesson: Lesson 11: Create Roles.
Browse using the Default and Internet Sales perspectives

In these first tasks, you will browse your model by using both the default perspective, which includes all model
objects, and also by using the Internet Sales perspective you earlier. The Internet Sales perspective excludes the
Customer table object.
To browse by using the Default perspective
1. Click the Model menu > Analyze in Excel.
2. In the Analyze in Excel dialog box, click OK.
Excel will open with a new workbook. A data source connection is created using the current user account
and the Default perspective is used to define viewable fields. A PivotTable is automatically added to the
worksheet.
3. In Excel, in the PivotTable Field List, notice the DimDate and FactInternetSales measure groups appear,
as well as the DimCustomer, DimDate, DimGeography, DimProduct, DimProductCategory,
DimProductSubcategory, and FactInternetSales tables with all of their respective columns appear.
To browse by using the Internet Sales perspective
1. Click the Model menu, and then click Analyze in Excel.
2. In the Analyze in Excel dialog box, leave Current Windows User selected, then in the Perspective drop-
down listbox, select Internet Sales, and then click OK.
3. In Excel, in PivotTable Fields, notice the DimCustomer table is excluded from the field list.
Browse by using roles

Roles are an integral part of any tabular model. Without at least one role to which users are added as members,
users will not be able to access and analyze data using your model. The Analyze in Excel feature provides a way for
you to test the roles you have defined.
To browse by using the Sales Manager user role
select Role, and then in the drop-down listbox, select Sales Manager, and then click OK.
Excel will open with a new workbook. A PivotTable is automatically created. The Pivot Table Field List
includes all of the data fields available in your new model.
What's next?
Go to the next lesson: Lesson 13: Deploy.
Lesson 13: Deploy
Premium
In this lesson, you will configure deployment properties; specifying an on-premises or Azure server instance, and a
name for the model. You'll then deploy the model to that instance. After your model is deployed, users can connect
to it by using a reporting client application. To learn more about deploying, see Tabular model solution deployment
and Deploy to Azure Analysis Services.
Prerequisites
in this lesson, you should have completed the previous lesson: Lesson 12: Analyze in Excel.
Deploy the model

To configure deployment properties
1. In Solution Explorer, right-click the AW Internet Sales project, and then click Properties.
2. In the AW Internet Sales Property Pages dialog box, under Deployment Server, in the Server
property, type the name of an Azure Analysis Services server or an on-premises server instance running in
Tabular mode. This will be the server instance your model will be deployed to.
IMPORTANT
You must have Administrator permissions on the remote Analysis Services instance in-order to deploy to it.
3. In the Database property, type Adventure Works Internet Sales.

4. In the Model Name property, type Adventure Works Internet Sales Model.
5. Verify your selections and then click OK.
To deploy the Adventure Works Internet Sales tabular model
1. In Solution Explorer, right-click the AW Internet Sales project > Build.
2. Right-click the AW Internet Sales project > Deploy.
When deploying to Azure Analysis Services, you'll likely be prompted to enter your account. Enter your
organizational account and passsword, for example [email protected]. This account must be in
Admins on the server instance.
The Deploy dialog box appears and displays the deployment status of the metadata as well as each table
included in the model.
3. When deployment successfully completes, go ahead and click Close.
Conclusion
Congratulations! You're finished authoring and deploying your first Analysis Services Tabular model. This tutorial
has helped guide you through completing the most common tasks in creating a tabular model. Now that your
Adventure Works Internet Sales model is deployed, you can use SQL Server Management Studio to manage the
model; create process scripts and a backup plan. Users can also now connect to the model using a reporting client
application such as Microsoft Excel or Power BI.
See also
DirectQuery Mode
Configure default data modeling and deployment properties
Tabular Model Databases
What's next?
Supplemental Lesson - Implement Dynamic Security by Using Row Filters.
Supplemental Lesson - Configure Reporting Properties for Power View Reports.
Supplemental Lesson - Implement Dynamic Security
by Using Row Filters
Premium
In this supplemental lesson, you will create an additional role that implements dynamic security. Dynamic security
provides row-level security based on the user name or login id of the user currently logged on. To learn more, see
Roles.
To implement dynamic security, you must add a table to your model containing the Windows user names of those
users that can create a connection to the model as a data source and browse model objects and data. The model
you create using this tutorial is in the context of Adventure Works Corp.; however, in order to complete this lesson,
you must add a table containing users from your own domain. You will not need the passwords for the user names
that will be added. To create an EmployeeSecurity table, with a small sample of users from your own domain, you
will use the Paste feature, pasting employee data from an Excel spreadsheet. In a real-world scenario, the table
containing user names you add to a model would typically use a table from an actual database as a data source; for
example, a real DimEmployee table.
In order to implement dynamic security, you will use two new DAX functions: USERNAME Function (DAX) and
LOOKUPVALUE Function (DAX). These functions, applied in a row filter formula, are defined in a new role. Using
the LOOKUPVALUE function, the formula specifies a value from the EmployeeSecurity table and then passes that
value to the USERNAME function, which specifies the user name of the user logged on belongs to this role. The
user can then browse only data specified by the role's row filters. In this scenario, you will specify that sales
employees can only browse Internet sales data for the sales territories in which they are a member.
In order to complete this supplemental lesson, you will complete a series of tasks. Those tasks that are unique to
this Adventure Works tabular model scenario, but would not necessarily apply to a real-world scenario are
identified as such. Each task includes additional information describing the purpose of the task.
Prerequisites
This supplemental lesson topic is part of a tabular modeling tutorial, which should be completed in order. Before
Add the DimSalesTerritory table to the AW Internet Sales Tabular

Model Project
In order to implement dynamic security for this Adventure Works scenario, you must add two additional tables to
your model. The first table you will add is DimSalesTerritory (as Sales Territory) from the same
AdventureWorksDW database. You will later apply a row filter to the SalesTerritory table that defines the particular
data the logged on user can browse.
To add the DimSalesTerritory table
1. In SSDT, click the Model menu, and then click Existing Connections.
2. In the Existing Connections dialog box, verify the Adventure Works DB from SQL data source
connection is selected, and then click Open.
If the Impersonation Credentials dialog box appears, type the impersonation credentials you used in Lesson
2: Add Data.
3. On the Choose How to Import the Data page, leave Select from a list of tables and views to choose
the data to import selected, and then click Next.
4. On the Select Tables and Views page, select the DimSalesTerritory table.
5. Click Preview and Filter.
6. Deselect the SalesTerritoryAlternateKey column, and then click Ok.
7. On the Select Tables and Views page, click Finish.
The new table will be added to the model workspace. Objects and data from the source DimSalesTerritory
table are then imported into your AW Internet Sales Tabular Model.
8. After the table has been imported, click Close.
Add a table with user name data

Because the DimEmployee table in the AdventureWorksDW sample database contains users from the
AdventureWorks domain, and those user names do not exist in your own environment, you must create a table in
your model that contains a small sample (three) of actual users from your organization. You will then add these
users as members to the new role. You do not need the passwords for the sample user names, but you will need
actual Windows user names from your own domain.
To add an EmployeeSecurity table
1. Open Microsoft Excel, creating a new worksheet.
2. Copy the following table, including the header row, and then paste it into the worksheet.
|EmployeeId|SalesTerritoryId|FirstName|LastName|LoginId|
|---------------|----------------------|--------------|-------------|------------|
3. Replace the first name, last name, and domain\username with the names and login ids of three users in
your organization. Put the same user on the first two rows, for EmployeeId 1. This will show this user
belongs to more than one sales territory. Leave the EmployeeId and SalesTerritoryId fields as they are.
4. Save the worksheet as SampleEmployee.
5. In the worksheet, select all of the cells with employee data, including the headers, then right-click the
selected data, and then click Copy.
6. In SSDT, click the Edit menu, and then click Paste.
If Paste is grayed out, click any column in any table in the model designer window, and try again.
7. In the Paste Preview dialog box, in Table Name, type EmployeeSecurity.
8. In Data to be pasted, verify the data includes all of the user data and headers from the SampleEmployee
worksheet.
9. Verify Use first row as column headers is checked, and then click Ok.
A new table named EmployeeSecurity with employee data copied from the SampleEmployee worksheet is
created.
Create relationships between FactInternetSales, DimGeography, and
DimSalesTerritory table
The FactInternetSales, DimGeography, and DimSalesTerritory table all contain a common column,
SalesTerritoryId. The SalesTerritoryId column in the DimSalesTerritory table contains values with a different Id for
each sales territory.
To create relationships between the FactInternetSales, DimGeography, and the DimSalesTerritory table
1. In the model designer, in Diagram View, in the DimGeography table, click and hold on the
SalesTerritoryId column, then drag the cursor to the SalesTerritoryId column in the DimSalesTerritory
table, and then release.
2. In the FactInternetSales table, click and hold on the SalesTerritoryId column, then drag the cursor to the
SalesTerritoryId column in the DimSalesTerritory table, and then release.
Notice the Active property for this relationship is False, meaning it's inactive. This is because the
FactInternetSales table already has another active relationship that is used in measures.
Hide the EmployeeSecurity Table from client applications

In this task, you will hide the EmployeeSecurity table, keeping it from appearing in a client application's field list.
Keep in-mind that hiding a table does not secure it. Users can still query EmployeeSecurity table data if they know
how. In order to secure the EmployeeSecurity table data, preventing users from being able to query any of its data,
you will apply a filter in a later task.
To hide the EmployeeSecurity table from client applications
In the model designer, in Diagram View, right-click the Employee table heading, and then click Hide from
Client Tools.
Create a Sales Employees by Territory user role

In this task, you will create a new user role. This role will include a row filter defining which rows of the
DimSalesTerritory table are visible to users. The filter is then applied in the one-to-many relationship direction to
all other tables related to DimSalesTerritory. You will also apply a simple filter that secures the entire
EmployeeSecurity table from being queryable by any user that is a member of the role.
NOTE
The Sales Employees by Territory role you create in this lesson restricts members to browse (or query) only sales data for the
sales territory to which they belong. If you add a user as a member to the Sales Employees by Territory role that also exists
as a member in a role created in Lesson 11: Create Roles, you will get a combination of permissions. When a user is a
member of multiple roles, the permissions, and row filters defined for each role are cumulative. That is, the user will have the
greater permissions determined by the combination of roles.
To create a Sales Employees by Territory user role

1. In SSDT, click the Model menu, and then click Roles.
A new role with the None permission is added to the list.
3. Click on the new role, and then in the Name column, rename the role to Sales Employees by Territory.
5. Click on the Members tab, and then click Add.
6. In the Select User or Group dialog box, in Enter the object named to select, type the first sample user
name you used when creating the EmployeeSecurity table. Click Check Names to verify the user name is
valid, and then click Ok.
Repeat this step, adding the other sample user names you used when creating the EmployeeSecurity table.
7. Click on the Row Filters tab.
8. For the EmployeeSecurity table, in the DAX Filter column, type the following formula.
=FALSE()
This formula specifies that all columns resolve to the false Boolean condition; therefore, no columns for the
EmployeeSecurity table can be queried by a member of the Sales Employees by Territory user role.
9. For the DimSalesTerritory table, type the following formula.
='Sales Territory'[Sales Territory Id]=LOOKUPVALUE('Employee Security'[Sales Territory Id],

'Employee Security'[Login Id], USERNAME(),
'Employee Security'[Sales Territory Id],
'Sales Territory'[Sales Territory Id])
In this formula, the LOOKUPVALUE function returns all values for the
DimEmployeeSecurity[SalesTerritoryId] column, where the EmployeeSecurity[LoginId] is the same as the
current logged on Windows user name, and EmployeeSecurity[SalesTerritoryId] is the same as the
DimSalesTerritory[SalesTerritoryId].
The set of sales territory IDs returned by LOOKUPVALUE is then used to restrict the rows shown in the
DimSalesTerritory table. Only rows where the SalesTerritoryID for the row is in the set of IDs returned by
the LOOKUPVALUE function are displayed.
10. In Role Manager, click Ok.
Test the Sales Employees by Territory User Role

In this task, you will use the Analyze in Excel feature in SSDT to test the efficacy of the Sales Employees by
Territory user role. You will specify one of the user names you added to the EmployeeSecurity table and as a
member of the role. This user name will then be used as the effective user name in the connection created between
Excel and the model.
To test the Sales Employees by Territory user role
select Other Windows User, and then click Browse.
3. In the Select User or Group dialog box, in Enter the object name to select, type one of the user names
you included in the EmployeeSecurity table, and then click Check Names.
4. Click Ok to close the Select User or Group dialog box, and then click Ok to close the Analyze in Excel
dialog box.
Excel will open with a new workbook. A PivotTable is automatically created. The PivotTable Fields list
includes most of the data fields available in your new model.
Notice the EmployeeSecurity table is not visible in the PivotTable Fields list. This is because you chose to
hide this table from client tools in a previous task.
5. In the Fields list, in ∑ Internet Sales (measures), select the InternetTotalSales measure. The measure will
be entered into the Values fields.
6. Select the SalesTerritoryId column from the DimSalesTerritory table. The column will be entered into
the Row Labels fields.
Notice Internet sales figures appear only for the one region to which the effective user name you used
belongs. If you select another column; for example, City, from the DimGeography table as Row Label field,
only cities in the sales territory to which the effective user belongs are displayed.
This user cannot browse or query any Internet sales data for territories other than the one they belong
because the row filter defined for the Sales Territory table in the Sales Employees by Territory user role
effectively secures data for all data related to other sales territories.
See Also
USERNAME Function (DAX)
LOOKUPVALUE Function (DAX)
CUSTOMDATA Function (DAX)
Supplemental Lesson - Configure Reporting
Properties for Power View Reports
Premium
In this supplemental lesson, you will set reporting properties for the AW Internet Sales project. Reporting
properties make it easier for users to select and display model data in Power View. You will also set properties to
hide certain columns and tables, and create new data for use in charts.
Prerequisites
This supplemental lesson is part of a tabular modeling tutorial, which should be completed in order. Before
In order to complete this particular supplemental lesson, you must also have the following:
The AW Internet Sales project (completed through this tutorial) ready to be deployed or already deployed to an
Analysis Services server.
Model properties that affect reporting

When authoring a tabular model, there are certain properties that you can set on individual columns and tables to
enhance the user reporting experience in Power View. In addition, you can create additional model data to support
data visualization and other features specific to the reporting client. For the sample Adventure Works Internet
Sales Model, here are some of the changes you will make:
Add new data - Adding new data in a calculated column by using a DAX formula creates date information
in a format that is easier to display in charts.
Hide tables and columns that are not useful to the end user - The Hidden property controls whether
tables and table columns are displayed in the reporting client. Items with that are hidden are still part of the
model and remain available for queries and calculations.
Enable one-click tables - By default, no action occurs if an user clicks a table in the field list. To change
this behavior so that a click on the table adds the table to the report, you will set Default Field Set on each
column that you want to include in the table. This property is set on the table columns that end users will
most likely want to use.
Set grouping where needed - The Keep Unique Rows property determines if the values in the column
should be grouped by values in a different field, such as an identifier field. For columns that contain
duplicate values such as Customer Name (for example, multiple customers named John Smith), it is
important to group (keep unique rows) on the Row IDentifier field in order to provide your end users with
the correct results.
Set data types and data formats - By default, Power View applies rules based on column data type to
determine whether the field can be used as a measure. Because each data visualization in Power View also
has rules about where measures and non-measures can be placed, it is important to set the data type in the
model, or override the default, to achieve the behavior you want for your user.
Set the Sort by Column property - The Sort By Column property specifies if the values in the column
should be sorted by values in a different field. For example, on the Month Calendar column that contains
the month name, sort by the column Month Number.
Hide tables from client Tools

Because there is already a Product Category calculated column and Product Subcategory calculated column in the
Product table, it is not necessary to have the Product Category and Product Subcategory tables visible to client
applications.
To hide the Product Category and Product Subcategory tables
1. In the model designer, right-click on the Product Category table (tab), and then click Hide from Client
Tools.
2. Right-click on the Product Subcategory table (tab), and then click Hide from Client Tools.
Create new data for charts

Sometimes it may be necessary to create new data in your model by using DAX formulas. In this task, you will add
two new calculated columns to the Date table. These new columns will provide date fields in a format convenient
for use in charts.
To create new data for charts
1. In the Date table, scroll to the far right, and then click on Add Column.
2. Add two new calculated columns using the following formulas in the formula bar:
COLUMN NAME FORMULA
Year Quarter =[Calendar Year] & " Q" & [Calendar Quarter]
Year Month =[Calendar Year] & FORMAT([Month],"#00")
Default field set

The Default Field Set is a predefined list of columns and measures for a table that are automatically added to a
report canvas when the table is clicked on in the report field list. Essentially, you can specify the default columns,
measures, and field ordering users will want to see when this table is visualized in Power View reports. For the
Internet Sales model, you will define a default field set and order for the Customer, Geography, and Product
tables. Included are only those most common columns that users will want to see when analyzing Adventure
Works Internet Sales data by using Power View reports.
For detailed information about Default Field Set, see Configure Default Field Set for Power View Reports.
To set default field set for tables
1. In the model designer, click the Customer table (tab).
2. In the Properties window, under Reporting Properties, in the Default Field Set property, click Click to
edit to open the Default Field Set dialog box.
3. In the Default Field Set dialog box, in the Fields in the table list box, press Ctrl, and select the following
fields, and then click Add.
Birth Date, Customer Alternate ID, First Name, Last Name.
4. In the Default fields, in order window, use the Move Up and Move Down buttons to put the following
order:
Customer Alternate ID
First Name
Last Name
Birth Date.
5. Click Ok to close the Default Field Set dialog box for the Customer table.
6. Perform these same steps for the Geography table, selecting the following fields and putting them in this
order.
City, State Province Code, Country Region Code.
7. Finally, perform these same steps for the Product table, selecting the following fields and putting them in
this order.
Product Alternate ID, Product Name.
Table behavior
By using Table Behavior properties, you can change the default behavior for different visualization types and
grouping behavior for tables used in Power View reports. This allows better default placement of identifying
information such as names, images, or titles in tile, card, and chart layouts.
For detailed information about Table Behavior properties, see Configure Table Behavior Properties for Power View
Reports.
To set table behavior
2. In the Properties window, in the Table Behavior property, click Click to edit, to open the Table Behavior
dialog box.
3. In the Table Behavior dialog box, in the Row IDentifier dropdown list box, select the Customer ID
column.
4. In the Keep Unique Rows list box, select First Name and Last Name.
This property setting specifies these columns provide values that should be treated as unique even if they
are duplicates, for example, when two or more employees share the same name.
5. In the Default Label dropdown list box, select the Last Name column.
This property setting specifies this column provides a display name to represent row data.
6. Repeat these steps for the Geography table, selecting the Geography ID column as the Row IDentifier,
and the City column in the Keep Unique Rows list box. You do not need to set a Default Label for this
table.
7. Repeat these steps, for the Product table, selecting the Product ID column as the Row IDentifier, and the
Product Name column in the Keep Unique Rows list box. For Default Label, select Product Alternate
ID.
Reporting properties for columns

There are a number of basic column properties and specific reporting properties on columns you can set to
improve the model reporting experience. For example, it may not be necessary for users to see every column in
every table. Just as you hid the Product Category and Product Subcategory tables earlier, by using a column's
Hidden property, you can hide particular columns from a table that is otherwise shown. Other properties, such as
Data Format and Sort by Column, can also affect how column data can appear in reports. You will set some of
those on particular columns now. Other columns require no action, and are not shown below.
You will only set a few different column properties here, but there are many others. For more detailed information
about column reporting properties, see Column Properties.
To set properties for columns
2. Click on the Customer ID column to display the column properties in the Properties window.
3. In the Properties window, set the Hidden property to True. The Customer ID column then becomes
greyed out in the model designer.
4. Repeat these steps, setting the following column and reporting properties for each table specified. Leave all
other properties at their default settings.
Note: For all date columns, make sure Data Type is Date.
Customer
COLUMN PROPERT Y VALUE
Geography ID Hidden True
Birth Date Data Format Short Date
Date
NOTE
Because the Date table was selected as the models date table by using the Mark as Date Table setting, in Lesson 7:
Mark as Date Table, and the Date column in the Date table as the column to be used as the unique identifier, the
Row IDentifier property for the Date column will automatically be set to True, and cannot be changed. When using
time-intelligence functions in DAX formulas, you must specify a date table. In this model, you created a number of
measures using time-intelligence functions to calculate sales data for various periods such as previous and current
quarters, and also for use in KPIs. For more information about specifying a date table, see Specify Mark as Date Table
for use with Time Intelligence.
Date Data Format Short Date
Day Number of Week Hidden True
Day Name Sort By Column Day Number of Week
Day of Week Hidden True
Day of Month Hidden True
Day of Year Hidden True
Month Name Sort By Column Month

Month Hidden True
Month Calendar Hidden True
Fiscal Quarter Hidden True
Fiscal Year Hidden True
Fiscal Semester Hidden True
Geography
Geography ID Hidden True
Sales Territory ID Hidden True
Product
Product ID Hidden True
Product Alternate ID Default Label True
Product Subcategory ID Hidden True
Product Start Date Data Format Short Date
Product End Date Data Format Short Date
Internet Sales
Product ID Hidden True
Customer ID Hidden True
Promotion ID Hidden True
Currency ID Hidden True
Sales Territory ID Hidden True
Order Quantity Data Type Decimal Number
Data Format Decimal Number
Decimal Places 0
Order Date Data Format Short Date
Due Date Data Format Short Date
Ship Date Data Format Short Date
Redeploy the Adventure Works Internet Sales tabular model

Because you have changed the model, you must re-deploy it.
To redeploy the Adventure Works Internet Sales tabular model
In SSDT, click the Build menu, and then click Deploy Adventure Works Internet Sales Model.
The Deploy dialog box appears and displays the deployment status of the metadata as well as each table
included in the model.
Next steps
You can now use Excel to visualize data from the model. Ensure the Analysis Services and Reporting Services
accounts on the SharePoint site have read permissions to the Analysis Services instance where you deployed your
model.

Tutorial Analysis Service Tabular Model

Uploaded by

Copyright:

Available Formats

Tutorial Analysis Service Tabular Model

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Tutorial Analysis Service Tabular Model

Uploaded by

Copyright:

Available Formats

Adventure Works Internet Sales tutorial (1400)

5/9/2019 • 3 minutes to read • Edit Online

What you learn

LESSON ESTIMATED TIME TO COMPLETE

1 - Create a new tabular model project 10 minutes

2 - Get data 10 minutes

3 - Mark as Date Table 3 minutes

4 - Create relationships 10 minutes

5 - Create calculated columns 15 minutes

6 - Create measures 30 minutes

7 - Create Key Performance Indicators (KPI) 15 minutes

8 - Create perspectives 5 minutes

9 - Create hierarchies 20 minutes

10 - Create partitions 15 minutes

11 - Create roles 15 minutes

12 - Analyze in Excel 5 minutes

LESSON ESTIMATED TIME TO COMPLETE

Detail Rows 10 minutes

Dynamic security 30 minutes

Ragged hierarchies 20 minutes

Create a new tabular model project

Understanding the SSDT tabular model authoring environment

Estimated time to complete this lesson: 10 minutes

Filter the table data

Import the selected tables and column data

Save your model project

Review existing relationships and add new relationships

ACTIVE TABLE RELATED LOOKUP TABLE

Yes DimCustomer [GeographyKey] DimGeography [GeographyKey]

Yes DimProduct DimProductSubcategory

Yes DimProductSubcategory DimProductCategory

Yes FactInternetSales [CustomerKey] DimCustomer [CustomerKey]

Yes FactInternetSales [ProductKey] DimProduct [ProductKey]

Create calculated columns

=RIGHT(" " & FORMAT([MonthNumberOfYear],"#0"), 2) & " - " & [EnglishMonthName]

=RIGHT(" " & FORMAT([DayNumberOfWeek],"#0"), 2) & " - " & [EnglishDayNameOfWeek]

3. Rename the column to ProductSubcategoryName.

3. Rename the column to ProductCategoryName.

4. Rename the column to Margin.

DaysCurrentQuarterToDate:=COUNTROWS( DATESQTD( 'DimDate'[Date]))

DaysInCurrentQuarter:=COUNTROWS( DATESBETWEEN( 'DimDate'[Date], STARTOFQUARTER(

COLUMN MEASURE NAME AUTOSUM (∑) FORMULA

SalesOrderLineNumber InternetOrderLinesCount Count =COUNTA([SalesOrderLine

OrderQuantity InternetTotalUnits Sum =SUM([OrderQuantity])

DiscountAmount InternetTotalDiscountAmo Sum =SUM([DiscountAmount])

TotalProductCost InternetTotalProductCost Sum =SUM([TotalProductCost])

SalesAmount InternetTotalSales Sum =SUM([SalesAmount])

Margin InternetTotalMargin Sum =SUM([Margin])

TaxAmt InternetTotalTaxAmt Sum =SUM([TaxAmt])

Freight InternetTotalFreight Sum =SUM([Freight])

Create Key Performance Indicators

This measure serves as the Base measure for the KPI.

8. Click OK to complete the KPI.

3. Right-click InternetCurrentQuarterMarginPerformance > Create KPI.

To create partitions for 2012, 2013, and 2014.

You create three roles:

Browse using the Default and Internet Sales perspectives

4. Close Excel without saving the workbook.

Browse by using roles