Data Analyst Interview Questions 2026: SQL, Excel, Power BI, Python and Real-World Scenarios

Preparing for a Data Analyst interview in 2026? Knowing Excel, SQL, Power BI and Python is a good start, but employers also want to understand how you solve problems, interpret data and communicate business insights.

This guide covers essential Data Analyst interview questions with clear answers, practical examples and scenario-based explanations to help freshers, students and career switchers prepare for technical and business interviews.

What Do Interviewers Look for in a Data Analyst?

Data Analyst interviews commonly assess a combination of technical and analytical skills. The exact requirements vary by company and role.

Key areas include:

SQL: Retrieving, joining, filtering and aggregating data.
Excel: Cleaning data, using formulas and creating summaries.
Power BI: Data modeling, DAX calculations and dashboard development.
Python: Data manipulation and exploratory analysis, where required.
Statistics: Understanding distributions, averages, variability and relationships.
Business thinking: Translating business questions into measurable analysis.
Communication: Explaining findings, assumptions and recommendations clearly.

Interviewers may assess these skills through technical questions, practical assignments, project discussions or business case studies.

1. General Data Analyst Interview Questions

Q1. What is data analytics?

Data analytics is the process of examining data to identify patterns, understand performance, answer questions and support decision-making.

For example, a retail company might analyze sales data to identify its best-performing products, understand regional differences and investigate declining revenue.

The purpose is not simply to produce reports. It is to generate reliable insights that help answer business questions.

Q2. What are the main stages of data analysis?

A typical workflow includes:

Define the problem: Understand the business question and the expected outcome.
Collect data: Identify relevant sources and obtain the necessary information.
Clean the data: Address missing values, duplicates, inconsistencies and invalid records.
Analyze the data: Explore trends, patterns and relationships.
Present findings: Communicate the results through reports, visualizations or presentations.
Recommend action: Suggest evidence-based next steps and determine how their impact will be evaluated.

Q3. What is the difference between a Data Analyst and a Data Scientist?

A Data Analyst typically focuses on understanding existing data, explaining business performance, creating reports and answering operational questions.

A Data Scientist may work more extensively with statistical modeling, experimentation, predictive analytics and machine learning.

There is overlap between the roles, and actual responsibilities depend on the organization.

Q4. How would you explain a technical finding to a non-technical stakeholder?

I would begin with the business question, explain the most important finding in plain language, provide supporting evidence and recommend a practical next step.

For example:

“Our conversion rate declined this month, even though website traffic increased. The decline is concentrated in mobile visitors, so we should investigate the mobile checkout experience before changing the overall marketing strategy.”

This communicates the finding and a possible investigation without claiming an unverified cause.

2. SQL Interview Questions and Answers

SQL is essential for many Data Analyst roles because it allows analysts to retrieve and investigate data stored in relational databases.

Consider a table called Orders containing order IDs, customer IDs, order dates and order amounts.

Q5. What is the difference between WHERE and HAVING?

WHERE filters individual rows before grouping and aggregation. HAVING filters groups after aggregation.

For example, to find customers whose total order value exceeds ₹1,000:

SELECT

customer_id,

SUM(order_amount) AS total_spent

FROM Orders

GROUP BY customer_id

HAVING SUM(order_amount) > 1000;

The query groups orders by customer, calculates the total amount and returns customers whose total exceeds ₹1,000.

Q6. What is the difference between INNER JOIN and LEFT JOIN?

An INNER JOIN returns records with matching values in both tables.

A LEFT JOIN retains every record from the left table and adds matching information from the right table. When no match exists, the right-side columns contain NULL.

For example, a company might use a LEFT JOIN to identify customers who have never placed an order.

Q7. How do you identify duplicate records in SQL?

You can group records by the field or combination of fields that should be unique and use HAVING COUNT(*) > 1.

SELECT

email,

COUNT(*) AS record_count

FROM Customers

GROUP BY email

HAVING COUNT(*) > 1;

This identifies email addresses that occur more than once.

Before deleting records, determine whether they are genuine duplicates. Multiple orders from one customer, for example, are not necessarily duplicate transactions.

Q8. What is a window function in SQL?

A window function performs calculations across related rows while retaining individual rows in the result.

For example, rank employees by salary within each department:

SELECT

employee_name,

department,

salary,

DENSE_RANK() OVER (

PARTITION BY department

ORDER BY salary DESC

) AS salary_rank

FROM Employees;

PARTITION BY separates employees into departments, while ORDER BY ranks salaries within each department.

Window functions are useful for ranking, running totals, period comparisons and identifying top-performing records.

Q9. How do you find the second-highest salary?

One approach uses DENSE_RANK():

WITH RankedSalaries AS (

SELECT

salary,

DENSE_RANK() OVER (

ORDER BY salary DESC

) AS salary_rank

FROM Employees

)

SELECT DISTINCT salary

FROM RankedSalaries

WHERE salary_rank = 2;

This returns the second-highest distinct salary, including cases where multiple employees share the highest salary.

Q10. What is the difference between UNION and UNION ALL?

UNION combines query results and removes duplicate rows.

UNION ALL combines query results while retaining duplicates.

UNION ALL is often appropriate when duplicate rows are meaningful or when you want to avoid the additional work associated with duplicate elimination.

Q11. What is a Common Table Expression (CTE)?

A CTE is a named temporary result defined using WITH. It helps organize complex SQL queries into more readable steps.

For example, you could first calculate each customer's total spending in a CTE and then use the result to identify high-value customers.

CTEs are particularly useful for breaking down multi-step analytical logic.

3. Excel Interview Questions and Answers

Excel is widely used for data preparation, reconciliation, reporting and exploratory analysis.

Q12. What is the difference between VLOOKUP and XLOOKUP?

VLOOKUP searches for a value in the first column of a specified range and returns a value from a column to its right.

XLOOKUP supports searches in either direction, uses exact matching by default and allows you to specify what should be returned when no match is found.

For example:

=XLOOKUP(A2, CustomerIDs, CustomerNames, "Not Found")

This searches for the customer ID in cell A2 and returns the corresponding name.

XLOOKUP is available in supported modern Excel versions. For older versions, INDEX and MATCH may be useful alternatives.

Q13. What is a PivotTable, and why is it useful?

A PivotTable summarizes a dataset by categories and calculates measures such as sums, counts and averages.

For example, a Data Analyst could use a PivotTable to summarize sales by region, compare revenue by month or identify products with the highest sales.

PivotTables are particularly helpful when exploring large spreadsheets without writing complex formulas for every summary.

Q14. How do you handle missing values in Excel?

First, identify where values are missing and determine why they are missing.

Depending on the situation, you might:

Recover the information from a reliable source.
Leave the values missing if they cannot be verified.
Exclude affected records when justified.
Use a suitable statistical replacement when the analysis supports it.
Document how missing values were handled.

Avoid replacing every blank with zero automatically. A missing value may mean unknown, not applicable or not recorded, while zero represents an actual numerical value.

Q15. What are COUNT, COUNTA and COUNTIF used for?

COUNT counts cells containing numbers.
COUNTA counts non-empty cells.
COUNTIF counts cells that meet a specified condition.

For example:

=COUNTIF(B2:B100,"Ahmedabad")

This counts the cells in the selected range that contain Ahmedabad.

Q16. How would you identify duplicate records in Excel?

You can use Conditional Formatting to highlight duplicates, the Remove Duplicates feature to remove records, or formulas to identify repeated values.

However, determine what makes a record unique before removing anything. Two transactions belonging to the same customer may be legitimate even if the customer ID appears twice.

4. Power BI Interview Questions and Answers

Power BI enables analysts to transform data, build data models and present interactive reports.

Q17. What is the difference between a measure and a calculated column?

A calculated column evaluates an expression for individual rows and stores the resulting values in the data model.

A measure calculates an aggregation or other expression in response to the current filter context, typically when a report visual requests a result.

Example of a calculated column:

Profit = Sales[Revenue] - Sales[Cost]

Example of a measure:

Total Revenue = SUM(Sales[Revenue])

Measures are generally preferred for dynamic aggregations that respond to slicers and report filters.

Q18. What is the difference between Power Query and DAX?

Power Query is primarily used to connect to data sources and clean, transform and prepare data before loading it into the model.

DAX is used to define calculations within the data model, including measures and calculated columns.

For example, you might use Power Query to standardize date formats and remove unwanted rows, then use DAX to calculate total revenue or year-over-year growth.

Q19. What is a star schema?

A star schema is a data modeling design in which a central fact table connects to dimension tables.

A sales model might include:

A FactSales table containing transactions, quantities, amounts and dimension keys.
A DimProduct table containing product names and categories.
A DimDate table containing dates, months, quarters and years.
A DimCustomer table containing customer attributes.

This structure supports clearer relationships and more consistent reporting when designed appropriately.

Q20. What is the difference between a slicer and a filter in Power BI?

Both restrict the data displayed in a report.

A slicer is a visual control that allows report users to interactively select values, such as a year or region.

Filters can be applied at different scopes, including a particular visual, a report page or the entire report.

Q21. How would you improve the performance of a slow Power BI report?

Start by identifying the source of the delay.

Possible improvements include:

Removing unnecessary columns and rows.
Improving the data model and relationships.
Simplifying expensive DAX calculations.
Reducing excessive visuals and unnecessary interactions.
Reviewing Power Query transformations.
Using Performance Analyzer to investigate slow visuals.

The correct solution depends on whether the bottleneck is data loading, query execution, model design, calculations or report rendering.

5. Python Interview Questions and Answers

Python is useful for automating repetitive tasks, manipulating datasets and performing exploratory analysis.

Q22. What is Pandas?

Pandas is a Python library for working with structured data through objects such as DataFrames and Series.

For example:

import pandas as pd

df = pd.read_csv("sales.csv")

print(df.head())

print(df.info())

print(df.isnull().sum())

This code loads a CSV file, displays the first few rows, summarizes column types and counts missing values.

Q23. How do you remove duplicate rows using Pandas?

df = df.drop_duplicates()

To remove duplicates based on a particular column:

df = df.drop_duplicates(subset=["customer_id"])

The second example keeps one record per customer ID, so use it only when that is the intended business rule. In a transaction dataset, multiple rows per customer are usually expected.

Q24. How do you calculate total sales by region using Pandas?

sales_by_region = (

df.groupby("region", as_index=False)["sales"]

.sum()

.sort_values("sales", ascending=False)

)

print(sales_by_region)

This groups records by region, sums the sales and sorts the result from highest to lowest.

During an interview, explain the purpose of groupby(), the aggregation and the sorting operation rather than simply reproducing the code.

6. Statistics Interview Questions

Statistics helps Data Analysts interpret data carefully and avoid misleading conclusions.

Q25. What is the difference between mean, median and mode?

The mean is the arithmetic average, the median is the middle value after sorting the observations, and the mode is the most frequently occurring value.

The median can be more representative when a dataset contains extreme values.

For example, if most employees earn between ₹30,000 and ₹50,000 per month but one executive earns ₹5,00,000, the mean salary may be considerably higher than the salary of a typical employee.

Q26. What is the difference between correlation and causation?

Correlation indicates an association between variables. Causation means that a change in one factor produces a change in another.

For example, ice cream sales and swimming-related incidents might both increase during hot weather. This does not mean ice cream sales cause the incidents. Temperature and seasonal behavior may influence both.

A Data Analyst should investigate other explanations before making causal claims.

Q27. What is an outlier, and how would you handle it?

An outlier is an observation that differs substantially from the rest of the dataset.

You might investigate outliers using:

Summary statistics and visualizations.
The interquartile range method.
Domain knowledge and source-system checks.
Comparisons against historical or expected values.

Do not automatically delete outliers. They might represent data errors, unusual but genuine transactions or important business events.

Q28. What is hypothesis testing?

Hypothesis testing is a statistical method used to evaluate evidence about a claim or difference.

For example, a company might test whether a redesigned website changes its conversion rate.

The analyst defines the hypotheses, chooses an appropriate test, considers the assumptions and sample size, and interprets the results using a significance threshold or confidence interval.

Statistical significance does not automatically imply business importance or prove causation.

7. Real-World Business Case Interview Questions

Business case questions test how you investigate problems and translate analysis into recommendations.

There is often no single correct answer. Interviewers want to understand your reasoning, assumptions and approach.

Q29. A company's sales have declined by 20%. How would you investigate?

I would start by clarifying what the decline means and which period is being compared.

Then I would follow a structured approach.

Step 1: Validate the data

Check for missing records, reporting delays, changes in definitions and inconsistencies between periods.

Step 2: Break down the decline

Analyze revenue by product, region, sales channel, customer segment and time period.

Step 3: Investigate the underlying metrics

Revenue can be expressed as:

Revenue = Number of Orders × Average Order Value

I would determine whether the decline is associated with fewer orders, lower average order value or both.

Step 4: Test possible explanations

Investigate factors such as pricing, product availability, marketing activity, customer behavior and conversion rates.

Step 5: Present findings and recommendations

Explain what the evidence supports, identify unresolved questions and recommend the next actions.

The key is not to guess the reason for the decline. It is to develop and test plausible explanations.

Q30. Website traffic increased, but sales did not. What would you analyze?

I would examine the conversion funnel to understand what happens after visitors arrive.

Areas to investigate include:

Traffic sources and campaign quality.
Landing-page engagement.
Add-to-cart rates.
Checkout abandonment.
Conversion rates by device.
Payment failures.
Average order value.

I would also verify that tracking definitions and reporting have remained consistent.

The objective is to identify where performance changed and determine which factors are supported by the data.

Q31. How would you investigate increasing customer churn?

First, I would define churn and establish the period being analyzed.

Next, I would segment customers by tenure, usage, purchase frequency, product type and other relevant characteristics.

I would look for patterns associated with customers leaving, such as reduced engagement, service complaints or renewal issues.

I would then validate the findings and recommend appropriate retention experiments or interventions. Observed associations alone would not prove why customers left.

Q32. A dashboard shows an unexpected increase in revenue. What would you check?

I would first validate the result rather than immediately presenting it as positive business performance.

Checks would include:

Whether the reporting period is correct.
Whether transactions are duplicated.
Whether refunds or cancellations are included consistently.
Whether a join has multiplied transaction records.
Whether the revenue calculation has changed.
Whether the increase is concentrated in a specific product or customer segment.

This demonstrates an important principle: validate the numbers before interpreting them.

8. How to Prepare for a Data Analyst Interview

A structured preparation plan can help you identify weaknesses and practice consistently.

Week 1: Excel and SQL

Review Excel formulas, PivotTables, SQL filtering, joins, aggregations and subqueries.

Practice solving problems without immediately checking the answer.

Week 2: Advanced SQL and statistics

Practice window functions, CTEs, date-based analysis, descriptive statistics, outlier detection and correlation versus causation.

Explain the reasoning behind each solution.

Week 3: Power BI and Python

Build a dashboard using a clean data model and meaningful KPIs. Practice basic Pandas operations if Python is relevant to your target roles.

Week 4: Projects and mock interviews

Choose a business problem, analyze the data and present your findings.

Practice explaining your project in two minutes, then prepare for follow-up questions about your assumptions, calculations and recommendations.

Adjust this plan according to your starting level and the requirements of the jobs you are targeting.

9. Common Mistakes to Avoid

Avoid these frequent interview mistakes:

Memorizing answers without understanding the concepts.
Writing SQL without checking whether joins produce duplicate records.
Building dashboards without defining the metrics correctly.
Treating missing values as zero without justification.
Claiming that correlation proves causation.
Presenting assumptions as verified findings.
Listing projects on a resume without being able to explain the work.
Using AI-generated code without validating its output.

Employers need evidence that you can work carefully with data and explain your decisions.

10. Frequently Asked Questions

Can a fresher get a Data Analyst job without work experience?

Yes. Freshers can demonstrate their capabilities through practical projects, internships, case studies and a portfolio that explains their analytical process. Hiring requirements vary by employer.

Is SQL necessary for Data Analyst interviews?

SQL is a core skill for many Data Analyst roles because it enables candidates to retrieve, combine and aggregate information from databases.

Is Python compulsory for every Data Analyst job?

No. Requirements vary by role. Some positions focus primarily on SQL, Excel and BI tools, while others expect Python for automation and more extensive analysis.

How many projects should I include in my portfolio?

Start with two to four well-documented projects that demonstrate different analytical skills. Prioritize clear problem statements, sound analysis and actionable findings over project quantity.

How can I practice Data Analyst interview questions?

Practice SQL exercises, Excel assignments, dashboard walkthroughs, statistics questions and business scenarios. Explain your reasoning aloud and validate your results before presenting conclusions.

What is the best way to stand out in a Data Analyst interview?

Demonstrate how you approached a real problem, validated the data, selected appropriate metrics and communicated your findings. Practical reasoning is more convincing than memorized definitions alone.

Conclusion

Preparing for a Data Analyst interview requires more than learning software. You need to understand how to retrieve reliable data, analyze it correctly, interpret the results and communicate what they mean for the business.

Build your fundamentals in SQL, Excel, Power BI and statistics. Learn Python where it supports your target role. Most importantly, practice solving business problems and explaining the decisions behind your analysis.

The goal is not just to answer interview questions. It is to demonstrate that you can think and work like a Data Analyst.

Start Building Practical Data Analytics Skills with DataWithCS

At DataWithCS, explore learning opportunities in Data Analytics, practical projects and industry-relevant skills.

Whether you are a student, a fresher or a professional planning a career transition, focus on developing skills that you can demonstrate through your work.

Visit https://datawithcs.com/ to explore available programs and learning resources.

Ready to move beyond tutorials? Explore DataWithCS and enquire about the next available demo session.

Data Analyst Interview Questions 2026: SQL, Excel, Power BI, Python and Real-World Scenarios

What Do Interviewers Look for in a Data Analyst?

Data Analyst interviews commonly assess a combination of technical and analytical skills. The exact requirements vary by company and role.

Key areas include:

SQL: Retrieving, joining, filtering and aggregating data.
Excel: Cleaning data, using formulas and creating summaries.
Power BI: Data modeling, DAX calculations and dashboard development.
Python: Data manipulation and exploratory analysis, where required.
Statistics: Understanding distributions, averages, variability and relationships.
Business thinking: Translating business questions into measurable analysis.
Communication: Explaining findings, assumptions and recommendations clearly.

Interviewers may assess these skills through technical questions, practical assignments, project discussions or business case studies.

1. General Data Analyst Interview Questions

Q1. What is data analytics?

Data analytics is the process of examining data to identify patterns, understand performance, answer questions and support decision-making.

For example, a retail company might analyze sales data to identify its best-performing products, understand regional differences and investigate declining revenue.

The purpose is not simply to produce reports. It is to generate reliable insights that help answer business questions.

Q2. What are the main stages of data analysis?

A typical workflow includes:

Define the problem: Understand the business question and the expected outcome.
Collect data: Identify relevant sources and obtain the necessary information.
Clean the data: Address missing values, duplicates, inconsistencies and invalid records.
Analyze the data: Explore trends, patterns and relationships.
Present findings: Communicate the results through reports, visualizations or presentations.
Recommend action: Suggest evidence-based next steps and determine how their impact will be evaluated.

Q3. What is the difference between a Data Analyst and a Data Scientist?

A Data Analyst typically focuses on understanding existing data, explaining business performance, creating reports and answering operational questions.

A Data Scientist may work more extensively with statistical modeling, experimentation, predictive analytics and machine learning.

There is overlap between the roles, and actual responsibilities depend on the organization.

Q4. How would you explain a technical finding to a non-technical stakeholder?

I would begin with the business question, explain the most important finding in plain language, provide supporting evidence and recommend a practical next step.

For example:

This communicates the finding and a possible investigation without claiming an unverified cause.

2. SQL Interview Questions and Answers

SQL is essential for many Data Analyst roles because it allows analysts to retrieve and investigate data stored in relational databases.

Consider a table called Orders containing order IDs, customer IDs, order dates and order amounts.

Q5. What is the difference between WHERE and HAVING?

WHERE filters individual rows before grouping and aggregation. HAVING filters groups after aggregation.

For example, to find customers whose total order value exceeds ₹1,000:

SELECT

customer_id,

SUM(order_amount) AS total_spent

FROM Orders

GROUP BY customer_id

HAVING SUM(order_amount) > 1000;

The query groups orders by customer, calculates the total amount and returns customers whose total exceeds ₹1,000.

Q6. What is the difference between INNER JOIN and LEFT JOIN?

An INNER JOIN returns records with matching values in both tables.

A LEFT JOIN retains every record from the left table and adds matching information from the right table. When no match exists, the right-side columns contain NULL.

For example, a company might use a LEFT JOIN to identify customers who have never placed an order.

Q7. How do you identify duplicate records in SQL?

You can group records by the field or combination of fields that should be unique and use HAVING COUNT(*) > 1.

SELECT

email,

COUNT(*) AS record_count

FROM Customers

GROUP BY email

HAVING COUNT(*) > 1;

This identifies email addresses that occur more than once.

Before deleting records, determine whether they are genuine duplicates. Multiple orders from one customer, for example, are not necessarily duplicate transactions.

Q8. What is a window function in SQL?

A window function performs calculations across related rows while retaining individual rows in the result.

For example, rank employees by salary within each department:

SELECT

employee_name,

department,

salary,

DENSE_RANK() OVER (

PARTITION BY department

ORDER BY salary DESC

) AS salary_rank

FROM Employees;

PARTITION BY separates employees into departments, while ORDER BY ranks salaries within each department.

Window functions are useful for ranking, running totals, period comparisons and identifying top-performing records.

Q9. How do you find the second-highest salary?

One approach uses DENSE_RANK():

WITH RankedSalaries AS (

SELECT

salary,

DENSE_RANK() OVER (

ORDER BY salary DESC

) AS salary_rank

FROM Employees

)

SELECT DISTINCT salary

FROM RankedSalaries

WHERE salary_rank = 2;

This returns the second-highest distinct salary, including cases where multiple employees share the highest salary.

Q10. What is the difference between UNION and UNION ALL?

UNION combines query results and removes duplicate rows.

UNION ALL combines query results while retaining duplicates.

UNION ALL is often appropriate when duplicate rows are meaningful or when you want to avoid the additional work associated with duplicate elimination.

Q11. What is a Common Table Expression (CTE)?

A CTE is a named temporary result defined using WITH. It helps organize complex SQL queries into more readable steps.

For example, you could first calculate each customer's total spending in a CTE and then use the result to identify high-value customers.

CTEs are particularly useful for breaking down multi-step analytical logic.

3. Excel Interview Questions and Answers

Excel is widely used for data preparation, reconciliation, reporting and exploratory analysis.

Q12. What is the difference between VLOOKUP and XLOOKUP?

VLOOKUP searches for a value in the first column of a specified range and returns a value from a column to its right.

XLOOKUP supports searches in either direction, uses exact matching by default and allows you to specify what should be returned when no match is found.

For example:

=XLOOKUP(A2, CustomerIDs, CustomerNames, "Not Found")

This searches for the customer ID in cell A2 and returns the corresponding name.

XLOOKUP is available in supported modern Excel versions. For older versions, INDEX and MATCH may be useful alternatives.

Q13. What is a PivotTable, and why is it useful?

A PivotTable summarizes a dataset by categories and calculates measures such as sums, counts and averages.

For example, a Data Analyst could use a PivotTable to summarize sales by region, compare revenue by month or identify products with the highest sales.

PivotTables are particularly helpful when exploring large spreadsheets without writing complex formulas for every summary.

Q14. How do you handle missing values in Excel?

First, identify where values are missing and determine why they are missing.

Depending on the situation, you might:

Recover the information from a reliable source.
Leave the values missing if they cannot be verified.
Exclude affected records when justified.
Use a suitable statistical replacement when the analysis supports it.
Document how missing values were handled.

Avoid replacing every blank with zero automatically. A missing value may mean unknown, not applicable or not recorded, while zero represents an actual numerical value.

Q15. What are COUNT, COUNTA and COUNTIF used for?

COUNT counts cells containing numbers.
COUNTA counts non-empty cells.
COUNTIF counts cells that meet a specified condition.

For example:

=COUNTIF(B2:B100,"Ahmedabad")

This counts the cells in the selected range that contain Ahmedabad.

Q16. How would you identify duplicate records in Excel?

You can use Conditional Formatting to highlight duplicates, the Remove Duplicates feature to remove records, or formulas to identify repeated values.

However, determine what makes a record unique before removing anything. Two transactions belonging to the same customer may be legitimate even if the customer ID appears twice.

4. Power BI Interview Questions and Answers

Power BI enables analysts to transform data, build data models and present interactive reports.

Q17. What is the difference between a measure and a calculated column?

A calculated column evaluates an expression for individual rows and stores the resulting values in the data model.

A measure calculates an aggregation or other expression in response to the current filter context, typically when a report visual requests a result.

Example of a calculated column:

Profit = Sales[Revenue] - Sales[Cost]

Example of a measure:

Total Revenue = SUM(Sales[Revenue])

Measures are generally preferred for dynamic aggregations that respond to slicers and report filters.

Q18. What is the difference between Power Query and DAX?

Power Query is primarily used to connect to data sources and clean, transform and prepare data before loading it into the model.

DAX is used to define calculations within the data model, including measures and calculated columns.

For example, you might use Power Query to standardize date formats and remove unwanted rows, then use DAX to calculate total revenue or year-over-year growth.

Q19. What is a star schema?

A star schema is a data modeling design in which a central fact table connects to dimension tables.

A sales model might include:

A FactSales table containing transactions, quantities, amounts and dimension keys.
A DimProduct table containing product names and categories.
A DimDate table containing dates, months, quarters and years.
A DimCustomer table containing customer attributes.

This structure supports clearer relationships and more consistent reporting when designed appropriately.

Q20. What is the difference between a slicer and a filter in Power BI?

Both restrict the data displayed in a report.

A slicer is a visual control that allows report users to interactively select values, such as a year or region.

Filters can be applied at different scopes, including a particular visual, a report page or the entire report.

Q21. How would you improve the performance of a slow Power BI report?

Start by identifying the source of the delay.

Possible improvements include:

Removing unnecessary columns and rows.
Improving the data model and relationships.
Simplifying expensive DAX calculations.
Reducing excessive visuals and unnecessary interactions.
Reviewing Power Query transformations.
Using Performance Analyzer to investigate slow visuals.

The correct solution depends on whether the bottleneck is data loading, query execution, model design, calculations or report rendering.

5. Python Interview Questions and Answers

Python is useful for automating repetitive tasks, manipulating datasets and performing exploratory analysis.

Q22. What is Pandas?

Pandas is a Python library for working with structured data through objects such as DataFrames and Series.

For example:

import pandas as pd

df = pd.read_csv("sales.csv")

print(df.head())

print(df.info())

print(df.isnull().sum())

This code loads a CSV file, displays the first few rows, summarizes column types and counts missing values.

Q23. How do you remove duplicate rows using Pandas?

df = df.drop_duplicates()

To remove duplicates based on a particular column:

df = df.drop_duplicates(subset=["customer_id"])

The second example keeps one record per customer ID, so use it only when that is the intended business rule. In a transaction dataset, multiple rows per customer are usually expected.

Q24. How do you calculate total sales by region using Pandas?

sales_by_region = (

df.groupby("region", as_index=False)["sales"]

.sum()

.sort_values("sales", ascending=False)

)

print(sales_by_region)

This groups records by region, sums the sales and sorts the result from highest to lowest.

During an interview, explain the purpose of groupby(), the aggregation and the sorting operation rather than simply reproducing the code.

6. Statistics Interview Questions

Statistics helps Data Analysts interpret data carefully and avoid misleading conclusions.

Q25. What is the difference between mean, median and mode?

The mean is the arithmetic average, the median is the middle value after sorting the observations, and the mode is the most frequently occurring value.

The median can be more representative when a dataset contains extreme values.

For example, if most employees earn between ₹30,000 and ₹50,000 per month but one executive earns ₹5,00,000, the mean salary may be considerably higher than the salary of a typical employee.

Q26. What is the difference between correlation and causation?

Correlation indicates an association between variables. Causation means that a change in one factor produces a change in another.

A Data Analyst should investigate other explanations before making causal claims.

Q27. What is an outlier, and how would you handle it?

An outlier is an observation that differs substantially from the rest of the dataset.

You might investigate outliers using:

Summary statistics and visualizations.
The interquartile range method.
Domain knowledge and source-system checks.
Comparisons against historical or expected values.

Do not automatically delete outliers. They might represent data errors, unusual but genuine transactions or important business events.

Q28. What is hypothesis testing?

Hypothesis testing is a statistical method used to evaluate evidence about a claim or difference.

For example, a company might test whether a redesigned website changes its conversion rate.

The analyst defines the hypotheses, chooses an appropriate test, considers the assumptions and sample size, and interprets the results using a significance threshold or confidence interval.

Statistical significance does not automatically imply business importance or prove causation.

7. Real-World Business Case Interview Questions

Business case questions test how you investigate problems and translate analysis into recommendations.

There is often no single correct answer. Interviewers want to understand your reasoning, assumptions and approach.

Q29. A company's sales have declined by 20%. How would you investigate?

I would start by clarifying what the decline means and which period is being compared.

Then I would follow a structured approach.

Step 1: Validate the data

Check for missing records, reporting delays, changes in definitions and inconsistencies between periods.

Step 2: Break down the decline

Analyze revenue by product, region, sales channel, customer segment and time period.

Step 3: Investigate the underlying metrics

Revenue can be expressed as:

Revenue = Number of Orders × Average Order Value

I would determine whether the decline is associated with fewer orders, lower average order value or both.

Step 4: Test possible explanations

Investigate factors such as pricing, product availability, marketing activity, customer behavior and conversion rates.

Step 5: Present findings and recommendations

Explain what the evidence supports, identify unresolved questions and recommend the next actions.

The key is not to guess the reason for the decline. It is to develop and test plausible explanations.

Q30. Website traffic increased, but sales did not. What would you analyze?

I would examine the conversion funnel to understand what happens after visitors arrive.

Areas to investigate include:

Traffic sources and campaign quality.
Landing-page engagement.
Add-to-cart rates.
Checkout abandonment.
Conversion rates by device.
Payment failures.
Average order value.

I would also verify that tracking definitions and reporting have remained consistent.

The objective is to identify where performance changed and determine which factors are supported by the data.

Q31. How would you investigate increasing customer churn?

First, I would define churn and establish the period being analyzed.

Next, I would segment customers by tenure, usage, purchase frequency, product type and other relevant characteristics.

I would look for patterns associated with customers leaving, such as reduced engagement, service complaints or renewal issues.

I would then validate the findings and recommend appropriate retention experiments or interventions. Observed associations alone would not prove why customers left.

Q32. A dashboard shows an unexpected increase in revenue. What would you check?

I would first validate the result rather than immediately presenting it as positive business performance.

Checks would include:

Whether the reporting period is correct.
Whether transactions are duplicated.
Whether refunds or cancellations are included consistently.
Whether a join has multiplied transaction records.
Whether the revenue calculation has changed.
Whether the increase is concentrated in a specific product or customer segment.

This demonstrates an important principle: validate the numbers before interpreting them.

8. How to Prepare for a Data Analyst Interview

A structured preparation plan can help you identify weaknesses and practice consistently.

Week 1: Excel and SQL

Review Excel formulas, PivotTables, SQL filtering, joins, aggregations and subqueries.

Practice solving problems without immediately checking the answer.

Week 2: Advanced SQL and statistics

Practice window functions, CTEs, date-based analysis, descriptive statistics, outlier detection and correlation versus causation.

Explain the reasoning behind each solution.

Week 3: Power BI and Python

Build a dashboard using a clean data model and meaningful KPIs. Practice basic Pandas operations if Python is relevant to your target roles.

Week 4: Projects and mock interviews

Choose a business problem, analyze the data and present your findings.

Practice explaining your project in two minutes, then prepare for follow-up questions about your assumptions, calculations and recommendations.

Adjust this plan according to your starting level and the requirements of the jobs you are targeting.

9. Common Mistakes to Avoid

Avoid these frequent interview mistakes:

Memorizing answers without understanding the concepts.
Writing SQL without checking whether joins produce duplicate records.
Building dashboards without defining the metrics correctly.
Treating missing values as zero without justification.
Claiming that correlation proves causation.
Presenting assumptions as verified findings.
Listing projects on a resume without being able to explain the work.
Using AI-generated code without validating its output.

Employers need evidence that you can work carefully with data and explain your decisions.

10. Frequently Asked Questions

Can a fresher get a Data Analyst job without work experience?

Yes. Freshers can demonstrate their capabilities through practical projects, internships, case studies and a portfolio that explains their analytical process. Hiring requirements vary by employer.

Is SQL necessary for Data Analyst interviews?

SQL is a core skill for many Data Analyst roles because it enables candidates to retrieve, combine and aggregate information from databases.

Is Python compulsory for every Data Analyst job?

No. Requirements vary by role. Some positions focus primarily on SQL, Excel and BI tools, while others expect Python for automation and more extensive analysis.

How many projects should I include in my portfolio?

Start with two to four well-documented projects that demonstrate different analytical skills. Prioritize clear problem statements, sound analysis and actionable findings over project quantity.

How can I practice Data Analyst interview questions?

Practice SQL exercises, Excel assignments, dashboard walkthroughs, statistics questions and business scenarios. Explain your reasoning aloud and validate your results before presenting conclusions.

What is the best way to stand out in a Data Analyst interview?

Conclusion

The goal is not just to answer interview questions. It is to demonstrate that you can think and work like a Data Analyst.

Start Building Practical Data Analytics Skills with DataWithCS

At DataWithCS, explore learning opportunities in Data Analytics, practical projects and industry-relevant skills.

Whether you are a student, a fresher or a professional planning a career transition, focus on developing skills that you can demonstrate through your work.

Visit https://datawithcs.com/ to explore available programs and learning resources.

Ready to move beyond tutorials? Explore DataWithCS and enquire about the next available demo session.