Data Analyst Interview Questions & Answers for Freshers (2026) – SQL, Power BI, Python & Scenario-Based

Data Analytics is the systematic practice of examining large volumes of raw data to surface patterns, relationships and actionable insights that guide business decision-making. It covers the full pipeline — from collecting and storing data to cleaning, querying, visualising and communicating findings. Unlike simply "looking at data," analytics connects each observation back to a business question, making it inherently goal-oriented.

In interviews, acknowledge both terms and clarify that the role of a Data Analyst spans both.

• Descriptive Analytics — summarises past events. Example: Monthly sales report showing revenue by region.
• Diagnostic Analytics — investigates why something happened. Example: Identifying that Q3 sales dipped because of supply chain delays.
• Predictive Analytics — forecasts likely future outcomes. Example: Predicting customer churn probability using a logistic regression model.
• Prescriptive Analytics — recommends specific actions. Example: Suggesting optimal inventory levels to prevent stockouts based on demand forecasting.

KPIs (Key Performance Indicators) are quantifiable metrics that measure how effectively a business achieves its goals. Selecting the right KPIs requires three steps:

1. Understand the business objective — e.g., "increase customer retention" points to metrics like churn rate and renewal rate.
2. Ensure measurability — the metric must be captured reliably in existing data sources.
3. Confirm actionability — stakeholders must be able to act on the metric, not just observe it.

Common examples include revenue growth rate, customer acquisition cost, Net Promoter Score and order fulfilment time.

Data Cleaning (also called data wrangling or data pre-processing) is the process of detecting and correcting errors, inconsistencies and gaps in a dataset before analysis. Issues addressed include:

• Missing values — rows where fields are empty or NULL
• Duplicate records — the same transaction or customer appearing more than once
• Format inconsistencies — dates stored as text, mixed case names, varying decimal separators
• Outliers — values far outside the expected range that may skew results

Analysts spend an estimated 60–80% of their time cleaning data. Skipping this step produces misleading insights, which can cost businesses significantly.

Practical example:

• Primary Key — a column (or combination of columns) that uniquely identifies every row in a table. It cannot be NULL and cannot contain duplicate values. Example: employee_id in an Employees table.
• Foreign Key — a column in one table that references the Primary Key of another table, establishing a relationship between them. Example: department_id in the Employees table pointing to department_id in a Departments table.

Foreign keys enforce referential integrity — you cannot add an employee with a department that does not exist.

NULL represents the absence of a value — it is not zero or an empty string. Key rules:

• Use IS NULL or IS NOT NULL to filter — never = NULL
• Use COALESCE(column, default_value) to substitute NULLs with a fallback
• Aggregate functions like SUM() and AVG() automatically ignore NULL rows

Window functions perform calculations across a set of rows related to the current row without collapsing them into a single group (unlike GROUP BY). They are heavily tested in senior and mid-level interviews.

Common window functions: ROW_NUMBER(), RANK(), DENSE_RANK(), LAG(), LEAD(), SUM() OVER().

VLOOKUP searches the leftmost column of a range for a match and returns a value from a specified column to the right.

Limitation: VLOOKUP can only look to the right and breaks if columns are inserted or reordered.

INDEX-MATCH is the preferred alternative — it can look in any direction, is faster on large datasets and remains stable even if columns shift.

• Use VLOOKUP for quick, one-off lookups on small, stable tables
• Use INDEX-MATCH for production reports and dynamic dashboards

Power BI is Microsoft's cloud-connected business intelligence platform used to ingest data from multiple sources, model relationships, create DAX-calculated measures and publish interactive dashboards to stakeholders — all without writing application code.

In a typical analyst workflow:

1. Connect Power BI to SQL databases, Excel files or APIs
2. Transform and clean data using Power Query (M language)
3. Build a data model with defined relationships and calculated columns
4. Design report pages with charts, slicers and KPI cards
5. Publish to Power BI Service for stakeholder access

Python extends what Excel and SQL can do, particularly for large datasets and automation:

• Pandas — data manipulation (filtering, grouping, pivoting)
• NumPy — numerical computations
• Matplotlib / Seaborn — creating charts and visualisations
• Scikit-learn — building basic predictive models
• OpenPyXL / xlrd — reading and writing Excel files programmatically

Python is not mandatory for every entry-level role, but knowing it gives freshers a measurable edge in competitive hiring rounds.

I would approach this in four steps:

1. Diagnose the pattern — Are values missing randomly or for a specific region, product or time period?
2. Consult the source — Talk to the data owner or check ingestion logs to determine whether it is a collection error or a genuine zero-revenue situation.
3. Choose a treatment strategy:
   • If missing at random and volume is low (<5%): remove those rows
   • If there is a logical substitute: impute with mean/median of the same category
   • If missing signals a real event (e.g., store closed): flag with a separate Boolean column
4. Document the decision — Record what was done and why, so downstream stakeholders understand the data's limitations.

Performance issues in Power BI typically fall into three categories:

• Data volume: Import only the columns and rows required. Use filters at the Power Query stage, not the visual level.
• DAX inefficiency: Replace calculated columns with measures where possible; avoid iterating functions like SUMX on large tables.
• Visual overload: Remove redundant visuals and cards. Each visual fires a separate query against the data model.
• Relationships: Ensure relationships use integer keys rather than text strings.
• Indexing: For DirectQuery mode, confirm the underlying SQL tables have appropriate indexes.

Use Power BI's built-in Performance Analyser to isolate which visual or query is the bottleneck before making changes.

I use a simple two-axis framework — business impact vs. effort:

1. Identify which report feeds a time-sensitive decision (e.g., a board presentation vs. a routine weekly update)
2. Estimate completion time and flag any dependencies (data still being loaded, access issues)
3. Communicate proactively with each stakeholder — confirm revised timelines rather than silently missing deadlines
4. Deliver the highest-impact report first, then move sequentially

In a real interview, walk through a specific example from your project experience if you have one.

Structure your answer around three pillars:

1. Technical readiness — "I have hands-on experience writing complex SQL queries, building Power BI dashboards and cleaning datasets using Python and Excel."
2. Business thinking — "I focus on what the numbers mean for the business, not just the numbers themselves. In my project, I translated a sales trend into a specific budget recommendation."
3. Growth mindset — "I actively follow developments in the analytics space — tools like Microsoft Fabric and AI-assisted querying — and I am eager to bring new efficiencies to your team."

Keep it concise (90 seconds), specific and evidence-backed.

SQL consistently ranks as the single most tested technical skill across entry-level Data Analyst interviews. Nearly every company with a structured database requires analysts to query data independently. Prioritise SQL above all other tools when preparing for your first role.

Yes. Many freshers secure Data Analyst roles by demonstrating practical project experience in lieu of professional experience. A strong GitHub portfolio with documented SQL scripts and a Power BI or Tableau dashboard project can substitute effectively for work history on your resume.

Not universally. Many entry-level roles in Chennai require only SQL, Excel and one BI tool. However, knowing foundational Python — particularly Pandas for data manipulation — gives you a clear advantage when applying to product companies, startups and IT services firms that handle large or complex datasets.

Freshers typically receive between ₹3 LPA and ₹6 LPA. Candidates with Python skills, a Power BI certification and at least two documented portfolio projects consistently land closer to the upper end. Service-based companies like TCS and Infosys tend to offer structured bands, while product companies may offer higher variable pay.

Both are valued, but Power BI has a larger share of job postings targeting freshers in Chennai, partly because of its integration with Microsoft 365 — a suite already deployed across most large enterprises. Tableau remains the standard in global analytics-heavy organisations. If you can only invest time in one, Power BI offers a faster return on interview readiness.

With consistent daily practice of two to three hours, most dedicated learners reach interview readiness in three to five months. Enrolling in a structured programme — such as a Data Analyst course in Chennai with real-time projects and mock interviews — can compress this timeline significantly by providing guided feedback and industry exposure.