Crack Power BI Interviews: Comprehensive Q&A from MNC Panels

Q1: How do you ensure that your Power-BI reports are both user-friendly and provide
deep insights?
Ans: I focus on creating a hierarchical structure in my reports. The top layer provides a high-
level overview with key metrics and visuals. Tooltips, drill-through, and interactive filters
allow users to delve deeper into specific areas of interest. I also ensure consistent color
coding and use clear labels to make the report intuitive

Q2: Describe a situation where you had to integrate unconventional data sources into
Power-BI.
Ans: I once had to integrate data from a third-party API with our internal databases in
Power-BI. I used Power Query to call the API, transform the JSON response into a structured
format, and then combined it with our SQL Server data to create a comprehensive report.

Q3: In SQL, how do you handle situations where you need to perform calculations on large
datasets without affecting database performance?
Ans: I use techniques like creating indexed views or temporary tables to store intermediate
results. This way, I can break down complex calculations into manageable steps and avoid
redundant computations. Additionally, I ensure that the database schema is optimized with
appropriate indexing to speed up query execution.
✅SCENARIO: HOW TO ADD A DATE COLUMN WITH DATE DATA TYPE ALONG WITH
ANOTHER COLUMN HAVING NUMBER DATA TYPE IN POWER QUERY EDITOR OF POWER
BI?
Ans: 1. Change Data type of the date column from date to whole number.
2. Add that changed data type column with Numeric value / column having numeric
value.
3. At the end change the data type from whole number to Date data type.
4. got the result.
Data Link :
Date + 30 days (1).pbix - Google Drive
File Link :
https://drive.google.com/file/d/16sSyXwFTPo6cZ4WWrHh9aw45fyEErJL6/view

SCENARIO: HOW TO USE NESTED CALCULATE DAX FUNCTION IN POWER BI ❓

Total_Sales_Amt_YTD_PY_Nested = CALCULATE ( CALCULATE ( CALCULATE ( SUM (
Orders[Sales] ) ), DATESYTD ( Orders[Order Date] ) ), PARALLELPERIOD ( Orders[Order Date],
-1, YEAR ) )

Pbix File link:

Calculate VS Nested Calculate.pbix - Google Drive (Click on Link)

QUESTION:- CALCULATE THE RUNNING TOTAL OF SALES OVER TIME, WHERE THE
RUNNING TOTAL AT ANY POINT REPRESENTS THE SUM OF SALES FROM THE START OF THE
YEAR TO THE CURRENT YERA ❓
Running Total = SUMX ( FILTER ( ALL ( SalesDetail ), SalesDetail[Order Date] Lessthen= MAX (
SalesDetail[Order Date] ) ), SalesDetail[Sales] )
Data Link :
SalesDetail.xlsx - Google Sheets
Pbix File Link :
Running sales .pbix - Google Drive

There are 2 slicers and 5 visuals in a Page, when I click on 1st slicers other visuals are
filtering and for the 2nd slicers they are not filtering. Why?

We have some possible reasons: -

1.Relationships in the Data Model: Ensure that there is a valid relationship between the field
used in the second slicer and the fields in the visuals you expect to be filtered. In Power BI,
relationships between tables play a crucial role in filtering data across different visuals.
2. Slicer Configuration: Check the configuration of the second slicer to make sure it is set up
correctly. Ensure that the slicer is bound to the appropriate field and that the field in the
slicer is part of the relationship established in the data model.
3. Cross-Filter Direction: Verify the cross-filter direction on the relationship between the
tables. Depending on the nature of your data, you may need to set the cross-filter direction
to "Both" or "Single" to control how filtering occurs between tables.
4. Filters in Visual Interactions: Inspect the visual interactions settings for each visual on the
page. Click on a visual, go to the Format pane, and check the "Edit interactions" option.
Ensure that the second slicer is set to affect the visuals you want it to influence. Incorrect
settings here can lead to visuals not responding to slicer interactions.
5. Data Cardinality: Consider the cardinality of the relationship between the tables. If there
is a many-to many relationships, it might require additional considerations or adjustments.
Ensure that the data model's cardinality is appropriate for the type of relationship you want
to establish.

What Will Happen If Many To Many Relationship Exist Between 2 Tables In Power BI In
Data Modeling
In Power BI data modeling, many-to-many relationships can present several challenges that
can impact the accuracy and performance of your reports and dashboards. Here are some
key issues to consider:
Filtering and Slicing Issues: Filtering and slicing data may not behave as expected due to the
potential for multiple matches. This can lead to unexpected results when visualizing data or
creating measures.
Cross Filtering: In Power BI, relationships automatically enable cross-filtering between
tables. With many-to-many relationships, cross-filtering can become more complex and may
not provide the desired outcomes.
Ambiguity in Aggregations: Aggregations (such as sum, average, etc.) can become
ambiguous when there are multiple related records. Power BI may not be able to determine
which records to aggregate.
Performance Impact: Many-to-many relationships can have a performance impact,
particularly with large datasets, as Power BI has to navigate through potentially complex
relationships to fetch the data.
DAX Measures and Calculations: When working with many-to-many relationships, you may
need to create more complex DAX measures and calculations to handle the ambiguity and
provide accurate results.

How you Handle MANY TO MANY Relationship In Power BI

In Power BI, resolving a many-to-many relationship typically involves creating a bridge table,
also known as a junction table or a link table. This bridge table helps break down the many-
to-many relationship into two one-to-many relationships, making it easier for Power BI to
handle.
Here are the general steps to resolve a many-to-many relationship in Power BI:
1. Identify the Tables Involved: Identify the two tables with the many-to-many
relationship. Let's call them Table A and Table B.
2. Create a Bridge Table: Create a new table, often referred to as a bridge table, that
contains unique combinations of keys from both Table A and Table B. This table acts
as a bridge between the two tables.

Suppose we have a report called Report 1, which we published in Workspace 1. Report 1

has two visualizations: Visualization 1 and Visualization 2. Similarly, suppose we have a
report called Report 2, which we published in Workspace 2. Report 2 also has two
visualizations: Visualization 1 and Visualization 2. Now, we have clicked on the pin icon of
Visualization 1 from Report 1, which is available in Workspace 1, and created Dashboard 1
by adding Visualization 1 from Report 1 to it. Can we now pin Visualization 1 from Report
2, which is available in Workspace 2, into Dashboard 1?
So, the answer to this question is No, we can't pin Visualization 1 of Report 2, which is
available in Workspace 2, into Dashboard 1.
Why? Because if we publish Report 1 into Workspace 1, its scope is limited to Workspace 1.
This means we can pin the visualization from Report 1 into Dashboard 1 because Dashboard
1 is within Workspace 1. Therefore, we can pin Visualization 1 of Report 1 only into
Dashboard 1.
Similarly, if we publish Report 2 into Workspace 2, its scope is limited to Workspace 2. This
means we can pin the visualization from Report 2 into Dashboard 2, which is created within
Workspace 2. Therefore, we can pin Visualization 1 of Report 2 only into Dashboard 2.
That's why we can't pin Visualization 1 of Report 2 into Dashboard 1, because Dashboard 1
is in Workspace 1, and Visualization 1 of Report 2 is in Workspace 2.

What kind of reports you used to build? Are you an individual contributor?

What are the data sources you have been used in your projects?

Have you used Direct Query connectivity mode? When can we go for Import and
Direct Query connectivity modes?

What is Query folding?

When you pull the data from SQL Server with Direct Query connectivity mode, what kind of
power query data transformations you can perform?

What is Star schema and what are the best practices to be followed when it comes to
cardinality and cross filter direction?
At which scenario will you use USERELATIONSHIP DAX function?

What are the DAX functions you have used in the reports?

Can you list the non scalar DAX functions?

What are table DAX functions?

For which scenario you have used SUMMARIZE AND SUMMARIZECOLUMNS?

Can you explain the use case for Left join and Left Anti join?

We can use VLOOKUP function to get the table A column to table B, similarly how we
can do it in Power Query level?

Do you know at which scenario we can use ISONORAFTER DAX function?

Is it required to have relationship when using RELATED DAX function?

What is LOOKUPVALUE DAX function?

Did you implement RLS?

Can we create Power BI report from Power BI Service?

What will happen to schedule refresh when it was failed multiple times?

There are two columns like Month Name and Sales Amount plotted in Sales Chart and how
will you sort Month’s month wise when it is alphabetical order?

What are Actions available for Card visual?

Every month revenue is generating for the product chocolates. end user wants to
see month on month running total. Can you write a measure?

What are the best practices you will follow for optimizing the DAX measures?

Question 1
Can we use User name and user principle name in static RLS?

If it has to do with credentials, it has to do with dynamic row level security, so it's not
used for static row level security. You should already know about static row level
security and dynamic row level security. It's important to know that username and
user principal name are both Dax functions that return the credentials of the logged-in
user.

Question 2
Does Power BI maintain / Store historical data and how?

The answer is no, Power BI does not maintain or store historical data. However,
we can use Power BI to do predictive analysis, which requires historical data.

Question 3
Is it possible to create drill down and drill up on 3 fact tables?

For this question, you need to know that fact tables usually aren't linked to each other. Fact
tables are the tables where transactional data is stored. This means that in real-time
projects, we usually have separate fact tables that aren't linked to each other. If they aren't
linked, you can't make a drill down or drill up on those fact tables.
Question 4
If there are constraint applied on table, does Primary key and Foreign key will affect in
Power BI?

In Power BI, when we load tables, we make a model on top of them.

We find the primary keys, the foreign keys between two tables, and
then we set up a one-to-many relationship. If a constant is applied
to the primary key or the primary column of the table, it will
definitely have an effect on the model. The primary key foreign key
relationship may break, which will cause the one-to-many
relationship to break as

Question 5
If we are taking data from Data Base, table has 6k rows and table got truncated has only
1k rows. Does Power BI will have the 6k rows from the table?

If we connect Power BI to any table in a database, it will pick up the exact number of
records that are in that table. For example, if the table has 1,000 rows, Power BI will pick
up those rows. If it has 6,000 rows, Power BI will pick up those rows. It depends on the
current number of rows in the table in the database. Whatever is there, it will be loaded
in Power BI.

Question 6
How will you identify type of license looking at workspace?

Those who have worked with Power BI Service will know that by looking at the
name of the workspace, you can tell what kind of license it has. For example, let
me show you through a screenshot. This is an app workspace, and these are all the
workspaces that are here. But if I talk about this MSX insights PRD, you can see
that there is one icon over here that is called a Diamond icon.

Question 7
What is the size of work space?

Space for the workspace is again determined by the type of license it has. For example, a
Premium license has 1GB of space, a Pro license has 400GB, and a Premium per user
license has 100GB of space.
Question 8
What is the unit testing in Power bi?

Basically, unit testing means checking the report at the unit level, which is the lowest
level you can test any report you've made. This means checking the report row by row
to see if it has the right data and matches the data coming from the source. For
example, if the data is coming from a SQL Server database, you have to check the row
b of every table that is being pulled into Power BI.

Question 9
Do you do the report validation, how do you do that?

While report validation is basically checking that the features you've added to Power
BI are functioning properly, you can also check that the KPIs you've created, the
calculations you've made, and the measures you've made or calculated are correct.
To do this, you need to use a sample size of data to make sure. For example, if you
have a running total measure, you can check each record individually.

Question 10
How will you cross verify the data which is coming from data source to report?

We create a model in Power BI using one-to- many relationships, and then we show the
data in a table from multiple tables. To check if the data being shown is actually in the
database, we have to go to the database, write a select statement using joints, and then
take and fetch the query results. Once we have the query results, we export the data
from the database in the same way we did in Power BI.

Question 11
Did you done any automation?

Basically, the PowerBI service has two features: subscriptions and alerts. You can
tell the interviewer about these two features and whether you have used them or
not. You can also talk about power automate for PowerBI. With this visual, you can
also talk about some automations that people usually do, so you can also give an
example of power automate in the automation part of this question.
Question 12
What is dual mode?

You know what import mode, direct query mode, and live connection are. But what is dual
mode? Well, dual mode is when one table acts as an import mode for one table and as a
direct query mode for another table at the same time. That table is in dual mode because
it handles both modes at the same time.

Question 13
Difference between Calculated Column and Measures and how they are helpful while
creating report?

Calculated columns works on row context, which means it performs the calculation row by
row, while Measures operate on filter context, which means they offer the result on the
flyer, which means whatever filters are used, the result will be generated.

The Calculated column takes up space in the model, but the Measures do not.
The calculated column is physically present, which means you can establish relationships
with this column in the data model. However, since measures do not exist, we cannot create
any type of relationship with them.

Question 14
How will you improve your power bi reports performance?
Our first step is to verify that all columns in the query-loaded tables have the correct data
types. Incorrectly assigning columns is because their data type creates a lot of space in the
model.
Second, remove unneeded columns from tables. Performance will improve if you eliminate
columns that are not used in any report visuals.
We can reduce the number of rows that you're pulling from the Source alright with the
assistance of aggregation or by reducing the number of rows from the Source itself.
Unselect the auto-generate date and time calendar in file and settings. Unless unchecked,
power bi generates a date table for each year, quarter, and month, which takes up model
space and reduces performance.

Question 15
How many types of relations are there we can perform while creating data model in PBI?

One-to-one, many-to-one, and many-to-many relationships

One-to-one connections mean both columns have unique values. Due to redundant data,
this rare cardinality class typically indicates a faulty model design.
Builders should choose a one-to-many or many-to-one link that matches the column order.
Consider how to link Product- Sales using each database's ProductID field. The Product table
has one-to-many cardinality since the ProductID column contains unique values. Reversing
Sales to Product yields many-to-one cardinality.

A many-to-many relationship suggests redundant data in both columns. This cardinality is

rare. It helps with complex model creation. It may relate many-to-many or high-grain facts.
Product category data stores sales target information, whereas product dimension data
stores product data.

Question 16
What are the types of filtering you have used in your report?
So we can speak about visual level filters, page level filters, report level filters, drill through
filters, and book markings if you have used book marks in any sort of filtering, so all of these
alternatives you can discuss in filtering.

Question 17
If you're calculating total sales and you want to apply filter function inside the calculate
column how it will work ?

So, the filter function is never used directly in any Dax code;
it is always nested within a calculate function. What does the filter function do?
It creates a subset of the table. As you all know, the second argument to the calculate
function is the filter expression, so the filter function is used there. It will help the Dax code
run faster if it is used within or next to the calculate functions.

Question 18
How many dashboards you have created till date and what are the objective of the
dashboards?
So, depending on your experience, you should talk about at least five to six reports that you
wrote from start to finish.
If you can, try to tell the reviewer about the goals of all of those reports as well. This will
make an effect on them.

Question 19
How will you decide that which are the metrics will be in dashboard to show these things
to stakeholders without getting any requirement from stakeholders?
You need to know about the domain on which your report is based.
To do this, you can Google it, learn about it, or watch some YouTube videos.
Then, you can try to figure out what kinds of general KPI’s are used or terms that are
specifically used in that domain.
With those KPI’s and terms, you can go ahead and show the demo.

Question 20
What are bidirectional relationship, have you used them in your report?
Bidirectional relationship is when you set the direction in both directions, right cross filter
Direction single or both.
If you set it to both, you have a bidirectional relationship.
We shouldn't use bidirectional filters very often because they slow down the report, but if
you need to use them for some calculations, you can use cross filter Dax as a workaround.

Question 21
Difference between Star schema and snowflake schema?
In a star schema, the fact table is in the middle, and dimension tables are arranged around it
in the shape of a star.
Dimension tables are linked to the fact table using
one-to- many relationship.
That's it for the star schema's structure. It's safer to use when 80% of the data is in the fact
table.
In the snowflake design, when your dimension tables are very large, it means that your
master data is also very large.
In this case, we standardize the data by splitting
the dimension tables into multiple dimension tables.
This way, we have less duplicate data and the model runs faster.

Question 22
How many types of Securities are there in power Bi service?
With Power BI's row-level security (RLS), you may restrict data access. You may set row-level
filters inside roles to restrict data access. Power BI lets users view semantic models in a
workspace. RLS restricts data viewing to Viewers. Not for Contributors, Members, or
Admins.
Static RLS: This form of RLS relies on static rules that remain constant throughout time.
Static RLS may limit access to a department- specific report to department members. The
regulations will stay unchanged in this situation.
Dynamic RLS: This RLS uses changing rules. Say you have a project report and a team
working on it. You may need to adjust the rules to restrict report access to existing team
members when they change. Rules alter over time in this scenario.

Question 23
How many types or roles are available in power bi service and what was your role?

It's mostly talking about the Power BI admin role, the member role, the contributor role, or
the viewer role.
As a Power BI developer, I usually had access at the member level, which meant I could
make reports, publish them, make changes to them, and even add users.
As a power BI developer, I also used to give users viewer-level access.
Question 24
Connecting different data sources with power bi especially Ms SQL (they are using
microsoft's ecosystem)as database and SharePoint as well?
This question shows that the interviewer wants to know how to combine data from
different sources into a single Power BI report.
For example, if you're using both a SQL Server
database and SharePoint as a source, remember that if you imported the data from the
database, you can also include other data sources, such as SharePoint.
If the database connection was direct query
connectivity, however, the SharePoint connection would not have been possible.

Question 25
If an Excel file contain multiple sheets how will you import all of them in power bi?

I think this question is asking how to pull all of the data from multiple CSV files at once.
To answer that, you can say that you will put all of the files in a separate folder and then pull
that folder into Power Query.
Once you have that folder in Power Query, you
can click on "Combine Files" and all the data from all of the CSV files will be uploaded
immediately into Power Query.

Question 26
Help me to take you through the process you follow and create dashboard from scratch
how you start?

In real life, any project has one business analyst who works directly with the client and
gathers the requirements.
Once he understands the requirements, he puts them on a BRD and shares it with you as a
power bi developer.
The business analyst also makes you meet with him.
You will analyze those items using the BRD, then fetch the tables in power query, conduct all
the necessary transformations, load the tables, and model them in the data model tab.

And do the visualization. If any calculations are needed using Dax, you will create measures
and calculated columns in the BRD. Finally, using the standard visuals in the standard pane,
you will visualize the report and give the client a demo after a few days.
Question 1
How to handle missing values?
In Power Query Editor, use the "Data Profiling"tool to find null numbers and how often they
happen.
Simply use the "Replace Values" function whenthere are only a few empty values in a
column.
When you need to do complicated calculations or work with null values in multiple fields or
tables, you should use DAX formulas.
When there are a lot of empty cells in a column,use the "Fill Down" or "Fill Up" choices.
If you want to change null numbers based on a certain situation, use the "Conditional
Column"option.

Question 2
Explain the difference between Calculated Column and Measure?

Let's talk about a table called "customer." It has columns for age, name, and gender. In
Power BI, we use Dax functions to find the average salary and then use a filter to make sure
that the gender of the employees is female. Since this is being calculated at the table level,
it is called a "measure" and we will be using it.
Calculated Measure Examples:-
Avg Salary = Calculate(Average(‘Customer’[Salary]), ‘Customer’[Gender] = “Female”)

You can create a calculated column on a row level. For example, if you have a column called
"age," you could make another column called "age bins" so that anyone between 0 and 20
would have "0 to 20" 21 to 40 41 to 60. This way, every value will have a corresponding
value in the "age bins" column. In this way, you're creating a new column from an old one,
which is why it's called a "calculated column."
Calculated Column Example:
Age Bins = SWITCH ( TRUE (), 'Customer'[Age] >= 0 && 'Customer'[Age] <= 20, "0 to 20",
'Customer'[Age] > 20
&& 'Customer'[Age] <= 40, "21 to 40", 'Customer'[Age] >
40 && 'Customer'[Age] <= 60, "41 to 60", 'Customer'[Age]
> 60, "61 and above", BLANK () )

Question 3
What is DAX and give an example?
In easy words Dax is Data Analysis Expressions; we've used them in Excel and are now using
them in Power BI.
We create Dax how? First, click the three dots on the data tab. Then, start working on Dax
methods. expression functions for data analysis There are intelligence functions and figure
sum filter functions. Power BI has more than 250 to 300 Dax tools.

Question 4
Explain RLS (Row Level Security) and how do you implement RLS
So, let's say you're working with clients from India, Hong Kong, China, Pakistan, Europe, and
other places. You've made a general dashboard that different people have to use. This is an
example of ROW level security. Users from India, Germany, and Hong Kong: If you only want
Indian users to see data related to India and only want German users to see data related to
Germany, you can use the RLS function. RLS is easy to understand.

Question 5
How do you disable a graph that is changing dynamically?
For example, you have a dashboard with several slicers, such as year, month, and city. There
are also several graphs. You want one of the graphs to stay the same no matter which slicer
you choose. For example, if you choose year, that graph should not change. For good
results, use Power BI's "Edit Interactions" feature. Now, go to the "Format" menu and click
on "Open" . If you click on "edit interactions," you'll see a sign that lets you turn off that
graph. If you click on that, you'll see "slicer," and it's very easy to change interactions.

Question 6
Explain the process of publishing a report?
You can answer the basic concepts behind power Bi desktop and power Bi service until that
once your dashboard is ready, when we go and click on publish, it basically asks for a
workspace where we need to publish this and we select the workspace, and then in that
workspace, our report, data set, and semantic model are all published there, so if you want
to access that from power bi service, go to app.power.com. Go to your workplace, and you
will be able to view the report.

Question 7
How to perform Join operations?
You can append and merge data in the Power Query Editor. If you want to upend two sets of
data that have the same columns, you can just choose the "upend" option. If you want to
join two tables, say "customers" and "orders," you can use the "merge" options as well.
These options include "left join," "inner join," and "right join."

Question 8
How to create relationships in Power BI?
Power BI has features that let you do things like joins, merges, and more. Another tab is
"Relationship," which lets you link two different data sources that share a column. There is
an option for "cardinality" when you drag and drop. For example, if you have two tables
called "customer ID" and "customer ID," and you want to move this to "customer ID," you
will see this choice. Different variations of this are possible. You can do that with what you
have. "One- on-one" only means This is where you can map one customer ID to another
customer ID.

Here is a simple example of a one-to-many relationship: each customer in the customers

table is unique, but each customer may have more than one order in the orders table. Thus,
there is a one-to-many cardinality. It's easy to make connections. Power BI makes some
relationships for you by default, but you can also make your own if you don't see any.

Question 9
Good Practices in Power BI?
The first good thing to do What I'm going to talk about is very easy. Let's say you're using
multiple data sources and multiple queries in the data tab. Let's say you're creating
measures on table one, table three, and table five by combining table one and table three
and making a new measure. Always start with a blank query and then add your measures.
All of your measures should be in the new blank query so you can maintain track of all of
them.

The second good practice is to limit the number of visuals on your report's dashboard. No
one likes ugly dashboards with 10 to 12 graphs in one dash; limit the visuals to only show
the important things and try to include some of Power BI's features, like tool tips or drill
throughs, and add them on how to do tool tips and drill throughs.
The third good practice is to always use certified visualizations. To open the visualization
marketplace, go to the visualization tab and click on the three dots. There are many
visualizations to choose from, but please only use the certified ones. Those have been
approved by Microsoft Power BI. The approved ones can be seen as blue buttons.

Question 10
How many types or roles are available in power bi service and what was your role?
It's mostly talking about the Power BI admin role, the member role, the contributor role, or
the viewer role. As a Power BI developer, I usually had access at the member level, which
meant I could make reports, publish them, make changes to them, and even add users.
As a power BI developer, I also used to give users viewer-level access.