An Accenture Data Analyst interview can test much more than basic SQL or spreadsheet skills. The supplied interview research indicates that candidates may be evaluated across SQL, Power BI and DAX, Python and Pandas, statistics, machine learning fundamentals, Excel, and stakeholder management. The process also places importance on explaining assumptions, solving practical problems, and communicating analytical findings clearly.
If you are preparing for an Accenture Data Analyst interview, these 30 questions cover the key technical and behavioral areas highlighted in the research.
SQL Interview Questions
SQL is a core part of the Data Analyst evaluation. The research emphasizes advanced querying, window functions, data quality, duplicates, and query performance.
1. What is the difference between OLTP and OLAP?
Answer:
OLTP systems are designed for frequent, short transactions such as purchases or banking transactions. They are generally normalized to maintain data integrity.
OLAP systems are designed for analytical queries, historical analysis, and aggregations. They are commonly optimized for read-heavy analytical workloads.
2. What is the difference between clustered and non-clustered indexes?
Answer:
A clustered index determines the physical ordering of rows in a table, so a table can generally have only one clustered index.
A non-clustered index is a separate structure containing indexed values and references to the underlying rows. Multiple non-clustered indexes can be created, but they add maintenance overhead during inserts, updates, and deletes.
3. What is the difference between EXISTS and IN in SQL?
Answer:
IN checks whether a value matches a list or the result of a subquery. EXISTS checks whether the subquery returns at least one matching row.
EXISTS can be advantageous for large datasets because the database can stop searching once a matching row is found. NOT EXISTS also avoids some of the NULL-related behavior that can occur with NOT IN.
4. How do you find the second-highest salary without using LIMIT or TOP?
Answer:
One approach is:
SELECT MAX(Salary)
FROM Employee
WHERE Salary < (SELECT MAX(Salary) FROM Employee);Another approach uses DENSE_RANK():
WITH RankedSalaries AS (
SELECT Salary,
DENSE_RANK() OVER (ORDER BY Salary DESC) AS rank_val
FROM Employee
)
SELECT Salary
FROM RankedSalaries
WHERE rank_val = 2;DENSE_RANK() is useful when multiple employees can have the same highest salary because it does not skip the next rank.
5. What is the use of the LAG() function in SQL?
Answer:
LAG() allows you to access a value from a preceding row without using a self-join. It is particularly useful for comparing current-period values with previous-period values.
For example, it can be used to calculate month-over-month sales growth.
6. How would you calculate month-over-month growth in SQL?
Answer:
First aggregate the metric by month, then use LAG() to obtain the previous month's value.
The basic calculation is:
((Current Month - Previous Month) / Previous Month) × 100
A robust query should also handle NULL or zero previous values to avoid incorrect calculations.
7. How do you identify duplicate records in SQL?
Answer:
A common approach is to use ROW_NUMBER() with PARTITION BY.
ROW_NUMBER() OVER (
PARTITION BY EmpName
ORDER BY JoinDate DESC
)Records with a row number greater than 1 can then be identified as duplicates according to the selected business rule.
8. How would you delete duplicate records while retaining the most recent record?
Answer:
Use a CTE with ROW_NUMBER(), partition the data according to the duplicate definition, and order records by the timestamp in descending order.
The newest record receives row_num = 1, while older duplicates receive values greater than 1 and can be targeted for deletion.
Crack Data Analyst Interviews with Real Company Questions
Prepare for your next Data Analyst interview with 750+ curated interview questions covering SQL, Python, Excel, Power BI, Statistics, Machine Learning, and HR interviews—all organized in one structured guide.
Power BI and DAX Interview Questions
The research highlights Power BI modeling, query folding, DAX contexts, time intelligence, and report navigation as important areas for Data Analyst candidates.
9. What is a Star Schema?
Answer:
A Star Schema has a central fact table containing metrics and foreign keys, surrounded by dimension tables containing descriptive information such as customers, products, and dates.
It creates a simple structure where dimensions connect directly to the fact table.
10. What is the difference between a Star Schema and a Snowflake Schema?
Answer:
A Star Schema keeps dimension tables relatively denormalized and directly connected to the fact table.
A Snowflake Schema further normalizes dimensions into multiple related tables. The research emphasizes using Star Schema structures for performant Power BI models.
11. What is Query Folding in Power BI?
Answer:
Query Folding is the process through which Power Query translates transformation steps into operations that can be executed by the underlying data source.
This allows processing to happen closer to the source and can reduce the amount of data transferred into the Power BI environment.
12. Why is Query Folding important for large datasets?
Answer:
It can reduce data movement and allow the source database to perform supported transformations rather than bringing unnecessary data into the Power BI model.
For large enterprise datasets, this can help improve refresh and processing efficiency.
13. What is Row Context in DAX?
Answer:
Row Context means DAX is evaluating an expression for the current row. It naturally occurs in calculated columns and iterator functions such as SUMX().
Row Context identifies the current row but does not itself act as a filter on the entire model.
14. What is Filter Context in DAX?
Answer:
Filter Context is the subset of data used by a DAX calculation. It can come from slicers, filters, cross-filtering, and the fields used in report visuals.
15. What is Context Transition in DAX?
Answer:
Context Transition converts an existing Row Context into a Filter Context.
The research identifies CALCULATE() and CALCULATETABLE() as the functions that trigger this behavior.
16. What are Time Intelligence functions in Power BI?
Answer:
Time Intelligence functions allow analysts to perform calculations across time periods, such as year-to-date, quarter-to-date, and previous-period comparisons.
Examples include:
- TOTALYTD()
- SAMEPERIODLASTYEAR()
- DATEADD()
- DATESQTD()
A proper continuous Date/Calendar table is important for accurate time-intelligence calculations.
17. What is the difference between a calculated column and a measure?
Answer:
| Calculated Column | Measure |
|---|---|
| Evaluated row by row during data refresh | Calculated dynamically when a visual queries it |
| Works naturally with Row Context | Works with dynamic Filter Context |
| Increases model storage | Stores calculation logic rather than materializing every result |
| Useful for categories and relationship-related attributes | Useful for aggregations, ratios, percentages, and KPIs |
The research recommends measures for dynamic analytical calculations such as percentages and KPI calculations.
18. What is the difference between Drill-Down and Drill-Through in Power BI?
Answer:
Drill-Down moves through levels of a hierarchy within a visual, such as Year → Quarter → Month.
Drill-Through takes the user to a separate report page containing details filtered for the selected item.
19. What are Tooltips in Power BI?
Answer:
Tooltips provide additional information when a user hovers over a visual. They can display supporting metrics or even additional visual content without overcrowding the main report page.
Python and Pandas Interview Questions
Python, particularly Pandas, is highlighted for data manipulation, exploratory analysis, cleansing, and automation.
20. What is the difference between a Series and a DataFrame in Pandas?
Answer:
A Series is a one-dimensional labeled data structure, similar to a single column.
A DataFrame is a two-dimensional tabular structure containing rows and columns. A DataFrame can contain multiple Series sharing the same index.
21. How do you handle missing values in Python?
Answer:
The appropriate approach depends on the amount and nature of the missing data.
Common methods include:
- dropna() to remove missing observations
- fillna() for statistical imputation
- Algorithmic methods such as K-Nearest Neighbors imputation
The choice should consider the impact on bias, variance, and the validity of the analysis.
22. When would you use median instead of mean for missing-value imputation?
Answer:
Median can be preferable when a numerical feature contains significant outliers because it is less affected by extreme values than the mean.
23. What are the different approaches to handling missing data?
Answer:
The research identifies three broad approaches:
- Listwise deletion — remove records containing missing values.
- Statistical imputation — replace missing values using mean, median, or mode.
- Algorithmic imputation — predict missing values using methods such as K-Nearest Neighbors.
Each approach involves different statistical and computational trade-offs.
Master the Data Analyst Interview Process
Stop guessing what they will ask. Get instant access to 750+ curated interview questions covering SQL, Python, Excel, Power BI, Statistics, Machine Learning, and HR interviews.
Statistics and Machine Learning Interview Questions
The research also highlights fundamental statistical modeling and machine learning concepts that candidates may need to explain clearly.
24. What are the main assumptions of Linear Regression?
Answer:
The key assumptions identified in the research are:
- Linearity
- Independence of errors
- Homoscedasticity
- Normality of residuals
- No problematic multicollinearity
Violations can affect the validity and interpretation of regression results.
25. What is the Bias-Variance Tradeoff?
Answer:
Bias represents error caused by an overly simplified model and can result in underfitting.
Variance represents excessive sensitivity to the training data and can result in overfitting.
The objective is to find an appropriate level of model complexity that generalizes well to unseen data.
26. What is the difference between normalization and standardization?
Answer:
Normalization, commonly Min-Max scaling, transforms values into a bounded range such as 0 to 1.
Standardization transforms values using their mean and standard deviation, producing a mean of 0 and standard deviation of 1.
The appropriate method depends on the algorithm and characteristics of the data.
Excel Interview Questions
Excel remains relevant because analysts may work with spreadsheet-based client data and use it for rapid analysis and reporting.
27. Why is INDEX-MATCH useful compared with VLOOKUP?
Answer:
INDEX-MATCH provides greater flexibility because the lookup column does not have to be the leftmost column. It can also avoid some structural limitations associated with hardcoded column positions in VLOOKUP.
It is therefore useful when working with changing or larger datasets.
28. How would you create a dynamic dashboard in Excel?
Answer:
A practical structure is:
- Keep the raw data in organized tables.
- Use Pivot Tables for calculations and summaries.
- Create Pivot Charts for visualization.
- Connect Slicers and Timeline controls to the relevant Pivot Tables.
- Use the final layer as the presentation dashboard.
This allows users to interactively filter multiple dashboard elements.
Behavioral and Stakeholder Interview Questions
Technical ability is only one part of the evaluation. The research emphasizes communication, ambiguity management, stakeholder conflict, analytical curiosity, and business impact.
29. How do you handle vague requirements or missing information?
Answer:
Start by clarifying the underlying business objective rather than immediately focusing on the technical request.
Then:
- Identify missing requirements.
- Document assumptions.
- Create a quick prototype or wireframe where useful.
- Share it with stakeholders.
- Establish a feedback loop before investing heavily in implementation.
This helps reduce misunderstandings and ensures the analytical solution addresses the actual business need.
30. Why do you want to join Accenture?
Answer:
Avoid focusing only on Accenture's size or reputation. A stronger answer connects your career goals with the intersection of business strategy, analytics, technology, and consulting.
You can discuss your interest in working across different projects, industries, and enterprise technology environments while explaining how the role fits your professional goals.
How to Prepare for an Accenture Data Analyst Interview
Based on the supplied research, preparation should not focus exclusively on memorizing SQL syntax.
Prioritize these areas:
Practice joins, window functions, ranking, duplicate handling, time-based calculations, and query-performance concepts.
Understand Star Schema modeling, Query Folding, DAX contexts, measures, calculated columns, and report navigation.
Focus on data manipulation, missing-value handling, and exploratory analysis.
Be comfortable explaining regression assumptions, bias-variance tradeoffs, and feature scaling.
Know lookup techniques and how to structure interactive dashboards.
Accenture's research emphasizes the ability to handle ambiguity, communicate analytical findings, and work with client stakeholders.
During technical interviews, do not simply provide the final query or answer. Explain your assumptions and reasoning as you work through the problem.
Frequently Asked Questions About Accenture Data Analyst Interviews
Accenture Data Analyst interviews can cover SQL, Power BI, DAX, Python, Pandas, statistics, machine learning fundamentals, Excel, and behavioral or stakeholder-management scenarios. The research emphasizes practical problem-solving and the ability to explain assumptions, not just theoretical knowledge.
Yes. The supplied research identifies SQL as a foundational skill for the Data Analyst role. Candidates may be tested on advanced querying, window functions, duplicate handling, joins, time-based calculations, and query performance.
Focus on window functions such as LAG(), ROW_NUMBER(), and DENSE_RANK(), subqueries, EXISTS versus IN, duplicate detection, ranking problems, and analytical calculations such as month-over-month growth.
The supplied research identifies Power BI as an important interview domain, particularly around semantic modeling, DAX, Query Folding, and report design.
Prepare Star Schema versus Snowflake Schema, Query Folding, Row Context, Filter Context, Context Transition, Time Intelligence, calculated columns versus measures, Tooltips, Drill-Down, and Drill-Through.
The research indicates that Python, particularly Pandas, can be tested for data manipulation, exploratory data analysis, data cleansing, and automation. The exact emphasis can vary by role.
Candidates may be asked about Pandas data structures such as Series and DataFrames, as well as approaches for handling missing data and their statistical trade-offs.
The supplied research includes statistics and machine learning fundamentals among the areas candidates may encounter. Topics include linear regression assumptions, bias-variance tradeoffs, and normalization versus standardization.
Excel remains relevant in the research because analysts may work with spreadsheet-based client data and use Excel for rapid analysis and reporting. Candidates should understand lookup techniques and dashboard construction.
Yes. The research places significant emphasis on stakeholder communication, ambiguity, conflict resolution, project ownership, and business impact. Behavioral and managerial discussions can therefore be an important part of the evaluation.
Use the STAR framework—Situation, Task, Action, and Result—when it fits the question. For scenario-based questions, emphasize what you personally did, how you communicated with stakeholders, and what measurable outcome resulted.
Start by identifying the underlying business objective. Then explain how you would identify missing information, document assumptions, create a quick prototype or wireframe when appropriate, and establish a feedback loop with stakeholders before proceeding with major implementation work.
The research describes Accenture's evaluation as combining technical problem-solving with consulting and stakeholder-management skills. Candidates are expected to translate ambiguous business requirements into analytical solutions and clearly communicate their reasoning.
The supplied research states that the overall lifecycle typically spans three to eight weeks, although the duration can depend on factors such as client project urgency and internal administrative approvals.
Prioritize SQL first, followed by Power BI/DAX, Python/Pandas, statistics and ML fundamentals, Excel, and stakeholder-management scenarios. Most importantly, practice explaining your assumptions and connecting analytical solutions to business requirements rather than only memorizing technical syntax.
Final Takeaway
Preparing for an Accenture Data Analyst interview requires a combination of technical depth and consulting-oriented communication.
SQL remains a major foundation, while Power BI, DAX, Python, statistics, Excel, and stakeholder management broaden the evaluation. The research also emphasizes that candidates should be able to connect analytical work with business requirements rather than treating data analysis as purely technical work.
Use these 30 questions as a focused preparation checklist, but practice explaining your answers in your own words. For scenario-based questions, structure your responses around the situation, task, action, and result where appropriate.