TCS Data Analyst Interview Questions (2026)

1. 1TCS NQT60 minMedium

   Online qualifier on TCS iON covering numerical/verbal/reasoning ability plus a coding section; the cutoff and score band route candidates to Ninja, Digital, or Prime interviews.

2. 2Technical Round (TR)40 minMedium

   Panel probes CS fundamentals (OOP, DBMS, OS basics), the candidate's final-year or work projects, and one or two simple programs written live; laterals get grilled on their claimed stack.

3. 3Digital/Prime Coding Interview45 minHard

   For Digital and Prime track candidates, a harder round with DSA problems, code optimization, and questions on cloud or newer technologies.

4. 4Managerial Round (MR)30 minMedium

   A senior manager tests composure with scenario and mild stress questions — missed deadlines, disagreement with a lead, willingness to work in any technology or location.

5. 5HR Round25 minEasy

   Discussion of the service agreement, relocation and shift flexibility, salary structure, and background verification documents.

1. Q1

   Write a SQL query to find duplicate customer records in TCS's banking client's loan-portfolio reporting. How would you decide which record survives?

   MediumSQL roundduplicate detectionTCS-specific

   Context: TCS banking client's loan-portfolio reporting

   How to answer: Use GROUP BY/HAVING or ROW_NUMBER over a business key, define survivorship rules, and push the fix upstream. State the business question, the data you would pull, and the checks that make the number trustworthy. Walk through the query or steps concretely, naming tables, joins, filters, and edge cases. Close with how you would validate the result and explain it to the stakeholder.

2. Q2

   Using window functions, how would you add a 7-day rolling total and a rank within category to TCS's retail client's daily sales dashboard?

   MediumSQL roundwindow functionsTCS-specific

   Context: TCS retail client's daily sales dashboard

   How to answer: Contrast aggregate vs window behavior, frame clauses, and partition choices. State the business question, the data you would pull, and the checks that make the number trustworthy. Walk through the query or steps concretely, naming tables, joins, filters, and edge cases. Close with how you would validate the result and explain it to the stakeholder.

3. Q3

   For TCS's telecom client's churn reporting, when would an INNER JOIN silently drop rows a LEFT JOIN would keep? Show the row counts you would check.

   MediumSQL roundjoinsTCS-specific

   Context: TCS telecom client's churn reporting

   How to answer: Explain join semantics with unmatched keys, and the count-before/count-after habit. State the business question, the data you would pull, and the checks that make the number trustworthy. Walk through the query or steps concretely, naming tables, joins, filters, and edge cases. Close with how you would validate the result and explain it to the stakeholder.

4. Q4

   Your monthly GROUP BY summary for TCS's insurance client's claims data mart does not match the source system's total. How do you reconcile them?

   HardSQL roundaggregation and reconciliationTCS-specific

   Context: TCS insurance client's claims data mart

   How to answer: Compare grain, filters, timezone/date boundaries, late-arriving rows, and duplicates. State the business question, the data you would pull, and the checks that make the number trustworthy. Walk through the query or steps concretely, naming tables, joins, filters, and edge cases. Close with how you would validate the result and explain it to the stakeholder.

5. Q5

   Which results change when key columns in TCS's e-commerce client's returns analysis contain NULLs — filters, joins, COUNT vs COUNT(col), averages?

   MediumSQL roundNULL handlingTCS-specific

   Context: TCS e-commerce client's returns analysis

   How to answer: Walk NULL semantics through WHERE, JOIN keys, aggregates, and COALESCE defaults. State the business question, the data you would pull, and the checks that make the number trustworthy. Walk through the query or steps concretely, naming tables, joins, filters, and edge cases. Close with how you would validate the result and explain it to the stakeholder.

6. Q6

   After adding a join, revenue in TCS's healthcare client's appointment no-show reporting doubled. Debug the query and prove the fix.

   HardSQL roundjoin fan-out debuggingTCS-specific

   Context: TCS healthcare client's appointment no-show reporting

   How to answer: Detect one-to-many fan-out, pre-aggregate or dedupe the many side, verify with row counts. State the business question, the data you would pull, and the checks that make the number trustworthy. Walk through the query or steps concretely, naming tables, joins, filters, and edge cases. Close with how you would validate the result and explain it to the stakeholder.

7. Q7

   When would you restructure a nested-subquery report for TCS's logistics client's delivery SLA tracking into CTEs, and what does it cost?

   MediumSQL roundCTEs and subqueriesTCS-specific

   Context: TCS logistics client's delivery SLA tracking

   How to answer: Readability, reuse, debugging layer by layer; note optimizer behavior differences. State the business question, the data you would pull, and the checks that make the number trustworthy. Walk through the query or steps concretely, naming tables, joins, filters, and edge cases. Close with how you would validate the result and explain it to the stakeholder.

8. Q8

   Find the top 3 products per region by revenue in TCS's credit-card client's spend-category dashboard, handling ties explicitly.

   MediumSQL roundranking problemsTCS-specific

   Context: TCS credit-card client's spend-category dashboard

   How to answer: ROW_NUMBER vs RANK vs DENSE_RANK, and which tie behavior the business actually wants. State the business question, the data you would pull, and the checks that make the number trustworthy. Walk through the query or steps concretely, naming tables, joins, filters, and edge cases. Close with how you would validate the result and explain it to the stakeholder.

9. Q9

   Your XLOOKUP against the master sheet in TCS's banking client's loan-portfolio reporting returns #N/A for valid IDs. What do you check?

   MediumExcel and BI roundlookupsTCS-specific

   Context: TCS banking client's loan-portfolio reporting

   How to answer: Trailing spaces, number-vs-text types, exact match mode, and TRIM/VALUE cleanup. State the business question, the data you would pull, and the checks that make the number trustworthy. Walk through the query or steps concretely, naming tables, joins, filters, and edge cases. Close with how you would validate the result and explain it to the stakeholder.

10. Q10

    Build a pivot summary for TCS's retail client's daily sales dashboard: what goes in rows, values, and filters, and when does a pivot beat formulas?

    MediumExcel and BI roundpivot tablesTCS-specific

    Context: TCS retail client's daily sales dashboard

    How to answer: Structure the source as a flat table, choose aggregations, and know pivot refresh limits. State the business question, the data you would pull, and the checks that make the number trustworthy. Walk through the query or steps concretely, naming tables, joins, filters, and edge cases. Close with how you would validate the result and explain it to the stakeholder.

11. Q11

    List the cleaning steps you would run on a raw extract for TCS's telecom client's churn reporting before any analysis.

    MediumExcel and BI rounddata cleaning in ExcelTCS-specific

    Context: TCS telecom client's churn reporting

    How to answer: Duplicates, blanks, types, date formats, merged cells, and a repeatable order of operations. State the business question, the data you would pull, and the checks that make the number trustworthy. Walk through the query or steps concretely, naming tables, joins, filters, and edge cases. Close with how you would validate the result and explain it to the stakeholder.

12. Q12

    In Power BI, when does TCS's insurance client's claims data mart need a DAX measure instead of a calculated column?

    HardExcel and BI roundDAX measuresTCS-specific

    Context: TCS insurance client's claims data mart

    How to answer: Row context vs filter context, storage cost, and totals that only measures get right. State the business question, the data you would pull, and the checks that make the number trustworthy. Walk through the query or steps concretely, naming tables, joins, filters, and edge cases. Close with how you would validate the result and explain it to the stakeholder.

13. Q13

    Design the Power BI model for TCS's e-commerce client's returns analysis: which tables are facts, which are dimensions, and why not one flat table?

    MediumExcel and BI rounddata modelingTCS-specific

    Context: TCS e-commerce client's returns analysis

    How to answer: Star schema, relationship directions, and what breaks when everything is one wide table. State the business question, the data you would pull, and the checks that make the number trustworthy. Walk through the query or steps concretely, naming tables, joins, filters, and edge cases. Close with how you would validate the result and explain it to the stakeholder.

14. Q14

    The client says the TCS dashboard for healthcare client's appointment no-show reporting takes a minute to load. How do you make it fast?

    MediumExcel and BI rounddashboard performanceTCS-specific

    Context: TCS healthcare client's appointment no-show reporting

    How to answer: Reduce cardinality, trim visuals, pre-aggregate, and measure before optimizing. State the business question, the data you would pull, and the checks that make the number trustworthy. Walk through the query or steps concretely, naming tables, joins, filters, and edge cases. Close with how you would validate the result and explain it to the stakeholder.

15. Q15

    Which KPIs go on the first screen of TCS's logistics client's delivery SLA tracking, and what do you deliberately leave out?

    MediumExcel and BI rounddashboard designTCS-specific

    Context: TCS logistics client's delivery SLA tracking

    How to answer: Lead with decisions the client makes weekly; park detail in drill-throughs. State the business question, the data you would pull, and the checks that make the number trustworthy. Walk through the query or steps concretely, naming tables, joins, filters, and edge cases. Close with how you would validate the result and explain it to the stakeholder.

Practice these with instant AI feedback in a live mock interview → Start a TCS Data Analyst mock

Clear the NQT; revise programming fundamentals and your projects; prepare HR questions

Indicative Data Analyst pay in India: ~₹6–22 LPA (role-level range, not a TCS-specific figure).

Other roles at TCS

Data Analyst interviews at other companies