PROJECT-1

INSTRUCTIONS

• You will be working through the entire business intelligence workflow: connecting and shaping the source
data, building a relational model, adding calculated columns and measures, and designing an
interactive report. I have attached all of the files you need to get started:
• Universal_Data.png (Universal Data logo)
• Universal_Data_CSV_Files.zip (zipped folder containing the 8 CSV files you'll need to build your report)
• In case you get stuck, I've also included a completed report (“Universal_Data_Report_Complete") for you to
reference along the way.
DATA ENGINEERING (SHAPING)
1) Update your Power BI options and settings as follows:
• Deselect the "Autodetect new relationships after data is loaded" option in the Data Load tab
• Make sure that Locale for import is set to "English (United States)" in the Regional Settings tab

2) Connect to the UniversalData_Customers csv file

• Name the table "Customers", and make sure that headers have been promoted
• Confirm that data types are accurate (Note: "customer_id" should be whole numbers, and both "customer_acct_num" and
"customer_postal_code" should be text)
• Add a new column named "full_name" to merge the the "first_name" and "last_name" columns, separated by a space
• Create a new column named "birth_year" to extract the year from the "birthdate" column, and format as text
• Create a conditional column named "has_children" which equals "N" if "total_children" = 0, otherwise "Y“
DATA ENGINEERING (SHAPING)
• Deselect the "Autodetect new relationships after data is loaded" option in the Data Load tab
•
• Steps
• Open Power BI desktop – file – Options and settings – Options – current files- data load – relationships – Disable auto detect new
relationships after data is loaded
•
1 Table name has been changed by double clicking on the table name.
2 data types has been changed
3 create duplicate columns of first name and last name and merged the 2 columns with space to
create renamed column as full name
4 duplicate column for the birth date and – transform- time- year remaned column as birth year
5 Added conditional column
3) Connect to the UniversalData_Products csv file
• Name the table "Products" and make sure that headers have been promoted
• Confirm that data types are accurate (Note: "product_id" should be whole numbers, "product_sku" should be text), "product_retail_price"
and "product_cost" should be decimal numbers)
• Use the statistics tools to return the number of distinct product brands, followed by distinct product names
• Spot check: You should see 111 brands and 1,560 product names
• Add a calculated column named "discount_price", equal to 90% of the original retail price
• Format as a fixed decimal number, and then use the rounding tool to round to 2 digits
• Select "product_brand" and use the Group By option to calculate the average retail price by brand, and name the new column "Avg
Retail Price"
• Spot check: You should see an average retail price of $2.18 for Washington products, and $2.21 for Green Ribbon
• Delete the last applied step to return the table to its pre-grouped state
• Replace "null" values with zeros in both the "recyclable" and "low-fat" columns

4) Connect to the UniversalData_Stores csv file

• Name the table "Stores" and make sure that headers have been promoted
• Confirm that data types are accurate (Note: "store_id" and "region_id" should be whole numbers)
• Add a calculated column named "full_address", by merging "store_city", "store_state", and "store_country", separated by a comma and
space (hint: use a custom separator)
• Add a calculated column named "area_code", by extracting the characters before the dash ("-") in the "store_phone" field
 Connected to the product table and all the data types are checked
Statistical tool- transform - statistic- count distinct values

Select "product_brand" and use the Group By option to calculate the average retail price by brand, and name the new column "Avg Retail
Price
• Add a calculated column named "discount_price", equal to 90% of the original retail price
• Format as a fixed decimal number, and then use the rounding tool to round to 2 digits
Ans: Custom column added and renamed as discount_Price
[product_retail_price]*90/100 used Rounding option for 2 Decimal

Replace "null" values with zeros in both the "recyclable" and "low-fat" columns
4) Connect to the UniversalData_Stores csv file
• Name the table "Stores" and make sure that headers have been promoted
• Confirm that data types are accurate (Note: "store_id" and "region_id" should be whole numbers)
• Add a calculated column named "full_address", by merging "store_city", "store_state", and "store_country", separated by a comma and
space (hint: use a custom separator)

Created Duplicate columns store city store state store country and merged all the columns with , as separator
5) Connect to the UniversalData_Regions csv file
• Name the table "Regions" and make sure that headers have been promoted
• Confirm that data types are accurate (Note: "region_id" should be whole numbers)

6) Connect to the UniversalData_Calendar csv file

• Name the table "Calendar" and make sure that headers have been promoted
• Use the date tools in the query editor to add the following columns:
• Start of Week (starting Sunday
• Name of Day
• Start of Month
• Name of Month
• Quarter of Year
• Year

7) Connect to the UniversalData_Returns csv file

• Name the table "Return_Data" and make sure that headers have been promoted
• Confirm that data types are accurate (all ID columns and quantity should be whole numbers)
 5 Connected to universal data- region and remanded column as region 1st column promoted as headers

6) Connected to the UniversalData_Calendar csv file

• ReNamed the table "Calendar" and promoted headers
• used date option in the transform
• Start of Week (starting Sunday
• Name of Day
• Start of Month
• Name of Month
• Quarter of Year
• Year

7) Connected to the UniversalData_Returns csv file

• ReNamed the table "Return_Data" and hearders promoted
• Confirmed data type is correct for all the columns
8) Add a new folder on your desktop (or in your documents) named "UniversalData_Transactions", containing both
the UniversalData_Transactions_1997 and UniversalData_Transactions_1998 csv files
• Connect to the folder path, and choose "Edit" (vs. Combine and Edit)
• Click the "Content" column header (double arrow icon) to combine the files, then remove the "Source.Name" column
• Name the table "Transaction_Data", and confirm that headers have been promoted
• Confirm that data types are accurate (all ID columns and quantity should be whole numbers)
• Spot check: You should see data from 1/1/1997 through 12/30/1998 in the "transaction_date" column
•

9) With the exception of the two data tables, disable "Include in Report Refresh", then Close & Apply
• Confirm that all 7 tables are now accessible within both the RELATIONSHIPS view and the DATA view

10) Save your .pbix file (i.e. " UniversalData_Report")

8) Created single csv file for the UniversalData_Transactions_1997 and UniversalData_Transactions_1998 in the laptop in the
new folder, Pulled the data to the power bi desktop and remanded file as Transaction_Data.

9) With the exception of the two data tables, disable "Include in Report Refresh", then Close & Apply
Right click on the Transaction_Data in queries and uncheck in include in report refresh

10 Saved file as UniversalData_Report

DATA MODELLING

1) In the RELATIONSHIPS view, arrange your tables with the lookup tables above the data tables
• Connect Transaction_Data to Customers, Products, and Stores using valid primary/foreign keys
• Connect Transaction_Data to Calendar using both date fields, with an inactive "stock_date" relationship
• Connect Return_Data to Products, Calendar, and Stores using valid primary/foreign keys
• Connect Stores to Regions as a "snowflake" schema

All the above task as been done

2) Confirm the following:

• All relationships follow one-to-many cardinality, with primary keys (1) on the lookup side and foreign keys (*) on the data side
• Filters are all one-way (no two-way filters)
• Filter context flows "downstream" from lookup tables to data tables
• Data tables are connected via shared lookup tables (not directly to each other)
DATA MODELLING
•

3) Hide all foreign keys in both data tables from Report View, as well as "region_id" from the Stores table
4) In the DATA view, complete the following:
• Update all date fields (across all tables) to the “m/d/yyyy" format using the formatting tools in the Modeling tab
• Update "product_retail_price", "product_cost", and "discount_price" to Currency ($ English) format
• In the Customers table, categorize "customer_city" as City, "customer_postal_code" as Postal Code, and "customer_country"
as Country/Region
• In the Stores table, categorize "store_city" as City, "store_state" as State or Province, "store_country" as Country/Region, and
"full_address" as Address

In the date View made all the above changes and saved pbx File
DAX MEASURES
1) In the DATA view, add the following calculated columns:
• In the Calendar table, or Sundays (otherwise add a column named "Weekend"
• Equals "Y" for Saturdays "N")

DAX = Weekend = IF('Calendar'[Day Name] == "Sunday" || 'Calendar'[Day Name] == "Saturday", "Y", "N")

• In the Calendar table, add a column named "End of Month"

• Returns the last date of the current month for each row

DAX= End of Month = EOMONTH('Calendar'[date],0)

•
• In the Customers table, add a column named "Current Age"
• Calculates current customer ages using the "birthdate" column and the TODAY() function

Current Age = DATEDIFF(Customers[birthdate],TODAY(),YEAR)

• In the Customers table, add a column named "Priority"
• Equals "High" for customers who own homes and have Golden membership cards (otherwise "Standard")
Priority = IF(Customers[homeowner] == "Y" && Customers[member_card] == "Golden", "High", "Standard")
DAX MEASURES
•
• In the Customers table, add a column named "Short Country"
• Returns the first three characters of the customer country, and converts to all uppercase

Short_Country = UPPER( LEFT(Customers[customer_country], 3) )

• In the Customers table, add a column named "House Number"
• Extracts all characters/numbers before the first space in the "customer_address" column (hint: use SEARCH)

House Number = left(Customers[customer_address], SEARCH(" ", Customers[customer_address]))

• In the Products table, add a column named "Price Tier"
• Equals "High" if the retail price is >$3, "Mid" if the retail price is >$1, and "Low" otherwise

Price_Tier = IF(Products[product_retail_price] < 1, "Low", IF(Products[product_retail_price] < 3, "Mid", "High"))

• In the Stores table, add a column named "Years_Since_Remodel"
• Calculates the number of years between the current date (TODAY()) and the last remodel date

Years_Since_Remodel = DATEDIFF(Stores[last_remodel_date],TODAY(),YEAR)
2) In the REPORT view, add the following measures (Assign to tables as you see fit, and use a matrix to match the "spot
check" values)

• Create new measures named "Quantity Sold" and "Quantity Returned" to calculate the sum of quantity from each data table
• Spot check: You should see total Quantity Sold = 833,489 and total Quantity Returned = 8,289

Quantity Returned = SUM(Return_Data[quantity])

• Create new measures named "Total Transactions" and "Total Returns" to calculate the count of rows from each data table
• Spot check: You should see 269,720 transactions and 7,087 returns
Total trasnsactions = COUNT('Transactional Data'[customer_id])
• Create a new measure named "Return Rate" to calculate the ratio of quantity returned to quantity sold (format as %)
• Spot check: You should see an overall return rate of 0.99%

Return Rate = [Quantity Returned]/[Total Quantity Sold] ( Check Spot is not matching )
• Create a new measure named "Weekend Transactions" to calculate transactions on weekends
• Spot check: You should see 76,608 total weekend transactions

Weekend Transactions = CALCULATE(COUNTROWS('Transactional Data'),'Calendar'[Weekend] == "Y")(Not matching with

check spot)
2)
• Create a new measure named "% Weekend Transactions" to calculate weekend transactions as a percentage of total transactions
(format as %)
• Spot check: You should see 28.4% weekend transactions

Unable to find the solution for this.

• Create new measures named "All Transactions" and "All Returns" to calculate grand total transactions and returns (regardless of
filter context)
• Spot check: You should see 269,720 transactions and 7,087 returns across all rows (test with product_brand on rows)
All transactions = COUNT('Transactional Data'[customer_id])
All returns = COUNT((Return_Data[product_id]))
• Create a new measure to calcul ate "Total Revenue" based on transaction quantity and product retail price, and format as
$ (hint: you'll need an iterator)
• Spot check: You should see a total revenue of $1,764,546

Calculated column first Revenue = 'Transactional Data'[quantity]*'Transactional Data'[Retail Price]

Total Revenue = SUM('Transactional Data'[Revenue])
• Create a new measure to calculate "Total Cost" based on transaction quantity and product cost, and format as $ (hint: you'll need
an iterator)
• Spot check: You should see a total cost of $711,728

get the cost column from product to transactional data cost = RELATED(Products[product_cost])
Order Cost = 'Transactional Data'[quantity]*'Transactional Data'[cost]
Total cost = SUM('Transactional Data'[Order Cost])

• Create a new measure named "Total Profit" to calculate total revenue minus total cost, and format as $
• Spot check: You should see a total profit of $1,052,819
Total Profit = [Total Revenue]-[Total cost]
• Create a new measure to calculate "Profit Margin" by dividing total profit by total revenue calculate total revenue (format as %)
• Spot check: You should see an overall profit margin of 59.67%

Profit Margin = [Total Profit]/[Total Revenue] (Check spot not matched)

• Create a new measure to calculate "Total Cost" based on transaction quantity and product cost, and format as $ (hint: you'll need
an iterator)
• Spot check: You should see a total cost of $711,728
• Create a new measure named "Total Profit" to calculate total revenue minus total cost, and format as $
• Spot check: You should see a total profit of $1,052,819
• Create a new measure to calculate "Profit Margin" by dividing total profit by total revenue calculate total revenue (format as %)
• Spot check: You should see an overall profit margin of 59.67%
• Create a new measure named "Unique Products" to calculate the number of unique product names in the Products table
• Spot check: You should see 1,560 unique products
• $Create a new measure named "YTD Revenue" to calculate year-to-date total revenue, and format as
• Spot check: Create a matrix with "Start of Month" on rows; you should see $872,924 in YTD Revenue in September 1998
• Create a new measure named "60-Day Revenue" to calculate a running revenue total over a 60-day period, and format as $
• Spot check: Create a matrix with "date" on rows; you should see $97,570 in 60-Day Revenue on 4/14/1997
• Create new measures named "Last Month Transactions", "Last Month Revenue", "Last Month Profit", and "Last Month
Returns"
• Spot check: Create a matrix with "Start of Month" on rows to confirm accuracy
• Create a new measure named "Revenue Target" based on a 5% lift over the previous month revenue, and format as $
• Spot check: You should see a Revenue Target of $99,223 in March 1998
•
• Create a new measure named "Unique Products" to calculate the number of unique product names in the Products table
• Spot check: You should see 1,560 unique products
Unique Products = DISTINCTCOUNT(Products[product_name])

• $Create a new measure named "YTD Revenue" to calculate year-to-date total revenue, and format as
• Spot check: Create a matrix with "Start of Month" on rows; you should see $872,924 in YTD Revenue in September 1998
YTD Revenue = CALCULATE(SUM('Transactional Data'[Revenue]),DATESYTD('Calendar'[date].[Year]))

• Create a new measure named "60-Day Revenue" to calculate a running revenue total over a 60-day period, and format as $
• Spot check: Create a matrix with "date" on rows; you should see $97,570 in 60-Day Revenue on 4/14/1997
Last 60 Days Revenue = VAR LAST_DATE= LASTDATE('Transactional Data'[stock_date])
RETURN
CALCULATE([Total Revenue],FILTER(ALL('Transactional Data'[stock_date]),[stock_date]> LAST_DATE -60 && 'Transactional
Data'[stock_date]<LAST_DATE))

•
• Create new measures named "Last Month Transactions", "Last Month Revenue", "Last Month Profit", and "Last Month
Returns"
• Spot check: Create a matrix with "Start of Month" on rows to confirm accuracy
Last month transactions = CALCULATE([Total trasnsactions], PREVIOUSMONTH('Calendar'[date]))
Last month revenue = CALCULATE([Total Revenue], PREVIOUSMONTH('Calendar'[date]))
Last Month Profit = CALCULATE([Total Profit], PREVIOUSMONTH('Calendar'[date]))
• Create a new measure named "Revenue Target" based on a 5% lift over the previous month revenue, and format as $
• Spot check: You should see a Revenue Target of $99,223 in March 1998
Revenue Target = CALCULATE([Last month revenue]*5/100+[Last month revenue])
 BUILDING THE REPORT
1) Rename the tab "Topline Performance" and insert the universal data logo

2) Insert a Matrix visual to show Total Transactions, Total Profit, Profit Margin, and Return Rate by Product_Brand (on rows)
• Add conditional formatting to show data bars on the Total Transactions column, and color scales on Profit Margin (White to Green)
and Return Rate (White to Red)
• Add a visual level Top N filter to only show the top 30 product brands, then sort descending by Total Transactions

3) Add a KPI Card to show Total Transactions, with Start of Month as the trend axis and Last Month Transactions as the target
goal
• Update the title to "Current Month Transactions", and format as you see fit
• Create two more copies: one for Total Profit (vs. Last month Profit) and one for Total Returns (vs. Last Month Returns)
• Make sure to update titles, and change the Returns chart to color coding to "Low is Good"

4) Add a Map visual to show Total Transactions by store city

• Add a slicer for store country
• Under the "selection controls" menu in the formatting pane, activate the "Show Select All" option
• Pro Tip: Change the orientation in the "General" formatting menu to horizontal and resize to create a vertical stack
(rather than a list)
 BUILDING THE REPORT
1) Rename the tab "Topline Performance" and insert the universal data logo

2) Insert a Matrix visual to show Total Transactions, Total Profit, Profit Margin, and Return Rate by Product_Brand (on rows)
• Add conditional formatting to show data bars on the Total Transactions column, and color scales on Profit Margin (White to Green)
and Return Rate (White to Red)
• Add a visual level Top N filter to only show the top 30 product brands, then sort descending by Total Transactions

•
5) Next to the map, add a Treemap visual to break down Total Transactions by store country
• Pull in store_state and store_city beneath store_country in the "Group" field to enable drill-up and drill-down functionality

6) Beneath the map, add a Column Chart to show Total Revenue by week, and format as you see fit
• Add a report level filter to only show data for 1998
• Update the title to "Weekly Revenue Trending"

7) In the lower right, add a Gauge Chart to show Total Revenue against Revenue Target (as either "target value" or "maximum
value")
• Add a visual level Top N filter to show the latest Start of Month
• Remove data labels, and update the title to "Revenue vs. Target"

8) Select the Matrix and activate the Edit interactions option to prevent the Treemap from filtering
9) Select "USA" in the country slicer, and drill down to select "Portland" in the Treemap
• Add a new bookmark named "Portland 1000 Sales"
• Add a new report page, named "Notes"
• Insert a text box and write something along the lines of "Portland hits 1,000 sales in December"
• Add a button (your choice) and use the "Action" properties to link it to the bookmark you created
• Test the bookmark by CTRL-clicking the button
• Find 2-3 additional insights from the Topline Performance tab and add new bookmarks and notes linking back

10) Get creative! Practice creating new visuals, pages, or bookmarks to continue exploring the data!