KPMG Business Analytics Interview Questions & Answers
Curated interview practice questions with answers. These are practice scenarios, not a verified list of past interview questions.
About KPMG hiring for Business Analytics roles
KPMG is one of the Big Four professional services networks, providing audit, tax and advisory services. KPMG in India and its global delivery centres support clients with technology consulting, data and analytics, risk advisory and cybersecurity. It hires in India for analysts and consultants in data analytics, governance and risk, IT audit, testing and ERP implementation.
For Business Analytics roles, this KPMG practice set focuses most on Product Analytics (6), Business Metrics and KPIs (5), Excel for Business Analysts (5), Process Modelling and BPMN (5). It has 4 easy, 31 medium, 5 hard questions, so you can start easy and work up to the harder ones.
Company type
Consulting
Headquarters
Amstelveen, Netherlands
Main India offices
Gurugram, Bengaluru, Mumbai, Hyderabad, Pune
KPMG Business Analytics interview process
Online assessment or screening test
Technical interview
Managerial or case round
Partner or HR round
Typical pattern based on public candidate experiences. Rounds vary by role, team and year.
Solved for KPMGStart practicing
3 QUESTIONS FREESign up to unlock the full question bank and track your progress.Unlock every question and answer, then continue practising company-wise and package-wise.
✓ 3 questions free • Unlock the rest in seconds
01
A hospital chain's average patient waiting time has risen from 30 to 50 minutes. How would you approach the case?
Clarify where the wait is measured: registration, consultation, diagnostics or billing. Split the problem into demand and capacity. On demand, check patient volume, peak-hour mix, walk-ins versus appointments and new departments. On capacity, check doctor availability, no-shows of staff, consultation time per patient and process bottlenecks such as manual billing. Identify the biggest stage and suggest fixes like slot-based appointments, triage and digital billing.
02
What is the difference between gross margin and net margin?
Gross margin is revenue minus cost of goods sold, divided by revenue, and shows how profitable the product itself is. Net margin is net profit after all expenses, including salaries, rent, marketing, interest and tax, divided by revenue. A company can have a healthy 60 percent gross margin but a negative net margin if it overspends on marketing or overheads, which tells you where to look for fixes.
03
Developers and testers interpret the same user story differently. How do you fix this?
The story is ambiguous, so bring the three amigos together: BA or product owner, a developer and a tester. Walk through concrete examples, including edge cases and error cases, and write them as acceptance criteria, ideally in Given-When-Then format. Add business rules, sample data and mock-ups where needed. Update the story and use the same practice in refinement for future stories, so gaps are found before coding starts.
04
Write DAX measures for Total Sales, Sales Last Year and Year-over-Year growth percent.
Situation: we tested a simplified pricing page expecting higher paid conversion. Task: I owned the experiment design and analysis. Action: after two weeks the variant showed no significant change, and segment checks revealed that heavy users missed the plan comparison table. I documented the result and shared it openly rather than searching for a winning segment. Result: the next version kept the simple layout with an optional comparison, and we learned to interview users before testing.
06
You receive two customer lists from sales and billing and must reconcile them. How do you do it?
Standardise both first: trim spaces, fix case, convert IDs to the same type and remove duplicates. Use a unique key like customer ID, or a combined key of phone and email if IDs differ. Then use XLOOKUP or COUNTIFS to mark each record as matched, only in sales or only in billing, or a Power Query merge with anti joins. Summarise mismatches by type and share the exception list with both owners.
Activation is the point at which a new user first experiences the product's core value, such as completing a first lesson, sending a first payment or creating a first project. The aha moment is the specific early action that strongly predicts long-term retention, found by comparing retained and churned users. Teams then design onboarding to get more new users to that action quickly and track the activation rate.
09
What is a SIPOC diagram and when do you use it before detailed process mapping?
SIPOC stands for Suppliers, Inputs, Process, Outputs and Customers. It is a one-page, high-level table that lists 4 to 7 main process steps along with who supplies what and who receives the outputs. A BA uses it at the start of a process improvement project to agree on scope and boundaries with stakeholders before spending time on a detailed BPMN map. It prevents arguments later about where the process starts and ends.
10
What is Customer Acquisition Cost (CAC) and what is CAC payback period?
CAC is total sales and marketing spend in a period divided by new customers acquired in that period. If you spend 10 lakh and gain 2,000 customers, CAC is 500 rupees. CAC payback period is how many months of gross profit from a customer it takes to recover that CAC. A shorter payback means the business can reinvest faster; many subscription businesses aim for under 12 months.
A data object (page shape with a folded corner) represents information that an activity needs or produces during one process instance, such as an invoice or an application form. It is linked with dotted associations. A data store (cylinder) represents information that persists beyond the process instance, such as the customer database or ERP. Showing them helps developers see which systems are read or updated at each step and helps spot missing integrations.
A self join joins a table to itself using two aliases, so you can compare rows within the same table. A classic business example is an employees table with a manager_id column: joining employees e to employees m on e.manager_id = m.employee_id lists each employee with their manager's name. Another example is finding customers who placed two orders on the same day by joining orders to orders on customer_id and date with different order IDs.
13
Revenue grew 20 percent this quarter but the number of active customers stayed flat. How do you interpret it?
Growth came from more value per customer, not more customers. Break revenue into customers times orders per customer times average order value to see which moved. It could be price increases, larger baskets, a shift to premium products or a few large accounts. Then check risks: is new customer acquisition stalling, is churn being masked, and is growth concentrated in a small segment that could leave?
14
Write Excel formulas to find the second-highest sales value and the name of the rep who achieved it.
// second highest value
=LARGE(C2:C100, 2)
// rep name for that value
=INDEX(B2:B100, MATCH(LARGE(C2:C100, 2), C2:C100, 0))
// with ties, SORTBY gives a clean list:
=TAKE(SORTBY(B2:C100, C2:C100, -1), 2)
15
An A/B test shows a 5 percent higher conversion rate, but revenue per user is lower. Should you ship it?
Not automatically. Find out why: the variant may push cheaper products, bigger discounts or smaller orders, so more people buy but each spends less. Check average order value, margin and long-term metrics like repeat purchase. Decide based on the metric tied to the business goal, usually revenue or contribution per user, which should have been the primary metric or a guardrail. Consider iterating to keep the conversion gain without lowering order value.
16
Tell me about a time a SQL query you wrote gave wrong results. How did you catch it?
Situation: my weekly sales report showed a 12 percent jump that no one could explain. Task: verify the number before it reached leadership. Action: I compared row counts before and after each join and found a new promotions table had multiple rows per order, duplicating revenue. I aggregated promotions per order first and added a test comparing totals with the source. Result: the corrected report went out the same day, and I now check grain on every new join.
It breaks a top metric into drivers: Revenue = Users × Conversion × AOV; Users = New + Returning, and so on. It shows which lever to pull and where a drop came from.
18
Write an Excel formula for a two-way lookup that finds sales for a given product and month.
A sequence flow is a solid arrow that shows the order of activities inside one pool, so it represents work controlled by one participant. A message flow is a dashed arrow with an open circle at its start and shows communication between two different pools, such as an order email from a customer to a seller. A common modelling mistake is drawing sequence flows across pools, which wrongly implies one party controls the other's work.
20
Write non-functional requirements for an online payment page.
NFR-01 Performance: the payment page loads within 2 seconds for 95 percent of requests at 500 concurrent users. NFR-02 Availability: 99.9 percent monthly uptime outside planned maintenance. NFR-03 Security: card data is never stored by the application, and all traffic uses TLS 1.2 or higher. NFR-04 Compliance: the flow meets PCI DSS and RBI payment guidelines. NFR-05 Usability: works on mobile screens from 360 pixels wide. NFR-06 Audit: every transaction attempt is logged with a timestamp.
21
How would you size the market for IT courses in Delhi NCR?
Top-down: NCR population aged 18–30 × share of graduates × share wanting paid upskilling × average fee. Cross-check bottom-up with institutes × batches × students × fees. State your assumptions.
22
Your Power BI report takes 30 seconds to load. How do you speed it up?
Start with Performance Analyzer to see which visuals and DAX queries are slow. Then fix the model: remove unused columns, reduce high-cardinality text, use a star schema and correct data types. Rewrite heavy measures, avoiding FILTER over whole tables and nested iterators. Cut the number of visuals per page, use aggregations or Import instead of DirectQuery where possible, and push transformations upstream into SQL or Power Query.
23
Describe how you would model an employee leave approval process in BPMN.
Use one pool for the company with lanes for Employee, Manager and HR system. Start event: leave request submitted. Task: system checks leave balance. Exclusive gateway: balance sufficient? If no, end event that notifies the employee. If yes, user task: manager reviews. Add an interrupting timer boundary event of 3 days that escalates to the skip-level manager. Gateway: approved or rejected. Approved path updates the leave balance and calendar, then both paths notify the employee and end.
A RAID log tracks Risks, Assumptions, Issues and Dependencies for a project. Each entry has an ID, description, owner, impact, status, mitigation or action, and due date. A risk is something that may happen, an issue is something that has already happened, an assumption is believed true but unverified, and a dependency is something the project needs from others. A BA uses it to record requirement assumptions and dependencies, and reviews it weekly so nothing is silently forgotten.
25
A ride-hailing app's ride cancellation rate rose from 8 to 14 percent. How would you structure the problem?
First split by who cancels: riders or drivers. For rider cancellations, check wait times, pickup accuracy, price surges and changes in ETA. For driver cancellations, check trip distance, payment mode, low fares, destination visibility and incentive changes. Then segment by city, time of day and area. Link to recent changes like a new pricing rule. Recommend fixes for the largest driver, such as better matching or adjusted incentives.
26
Daily active users dropped 10 percent week over week. How do you investigate?
Rule out data issues first: tracking changes, app release bugs, pipeline delays or bot filters. Check whether the drop is seasonal by comparing with the same week last year or a festival calendar. Then segment by platform, app version, country or city, acquisition channel and new versus returning users to locate it. Look at the funnel for the core action. Connect findings to releases, outages or marketing changes, then recommend fixes.
27
Describe a time your analysis found a cost-saving opportunity.
Situation: monthly cloud and software subscription costs had grown 35 percent in a year without anyone tracking why. Task: my manager asked me to review them. Action: I combined invoices with user login data in Excel and found 140 licences unused for over 90 days and two tools doing the same job. Result: we cancelled unused licences and consolidated tools, saving about 18 lakh rupees a year, and I set up a quarterly licence review report.
28
What is the difference between a vanity metric and an actionable metric?
A vanity metric looks impressive but does not guide decisions, such as total app downloads, page views or social followers that only ever go up. An actionable metric links clearly to a decision and behaviour, such as Day-7 retention, conversion rate by channel or revenue per active user. A good test: if the number changes, do you know what to do differently? If not, it is probably vanity.
29
What is the difference between correlation and causation in product analytics?
Correlation means two metrics move together; causation means one actually changes the other. Users who use the wishlist feature may retain better, but they may simply be more engaged users already. Acting on correlation alone can waste effort, such as forcing everyone into the wishlist. To establish causation, run an A/B test, or use quasi-experimental methods such as difference-in-differences when a test is not possible, and always check for confounders like user tenure or acquisition channel.
30
Inventory days rose from 45 to 70 over two quarters. What could be the causes and what would you recommend?
Possible causes include falling sales against unchanged purchasing, poor demand forecasts, bulk buying for discounts, slow-moving or obsolete SKUs, or deliberate stock-building ahead of a festive season or expected price rise. Analyse by SKU, warehouse and age bucket to find where stock is stuck. Recommend clearing aged items, adjusting reorder points using recent demand, and reviewing supplier minimum order quantities. Also quantify the cash locked up and the carrying cost to show urgency.
A UAT test plan defines how acceptance testing will be run. It covers scope and out-of-scope items, business processes to be tested, entry and exit criteria, roles such as testers, UAT lead and sign-off authority, test environment and test data, schedule and cycles, defect management process with severity definitions, communication and reporting, and risks. The BA usually drafts it with business owners so testers know exactly what they are accepting and when.
32
Write a SQL query to find requirements that have no linked UAT test case.
SELECT r.req_id, r.title, r.priority
FROM requirements r
LEFT JOIN test_case_links l ON l.req_id = r.req_id
WHERE l.req_id IS NULL
AND r.status = 'APPROVED'
ORDER BY r.priority;
-- Run this before UAT starts: every approved requirement
-- should trace to at least one test case.
33
A stakeholder's only requirement is 'make the app faster'. How do you turn this into clear requirements?
Ask which screens or actions feel slow, for which users, on which devices and networks, and how it affects the business, such as checkout drop-off. Gather current timings from monitoring data. Then write measurable non-functional requirements, for example 'the search results page loads within 2 seconds for 95 percent of requests on a 4G connection'. Confirm targets with the stakeholder and agree how they will be tested.
34
Guesstimate: how many pizzas are sold in Mumbai on a weekend?
About 20 million people, roughly 10% order pizza on a weekend, about 1 pizza per 2 people in an order. This gives roughly 1 million pizzas, adjusted for outlets and delivery capacity.
A/B testing is a controlled experiment where users are randomly split into a control group that sees the current version and a treatment group that sees a change, such as a new checkout button. You compare a pre-decided primary metric, like conversion rate, between groups. Because assignment is random, a significant difference can be attributed to the change rather than to seasonality, marketing or user mix.
36
What is row-level security in Power BI and how do you set it up?
Row-level security restricts which rows a user can see in the same report. In Desktop you create roles under Manage Roles with a DAX filter, such as 'Region'[Manager Email] = USERPRINCIPALNAME() for dynamic RLS. You test with View as Role, publish, and then map users or groups to roles in the Service. Dynamic RLS with a mapping table scales better than creating one static role per region.
37
Write a DAX measure for the product return rate that avoids divide-by-zero errors.
Return Rate % =
VAR ReturnedQty = SUM ( Returns[Quantity] )
VAR SoldQty = SUM ( Sales[Quantity] )
RETURN
DIVIDE ( ReturnedQty, SoldQty, 0 )
// DIVIDE returns the third argument (0) when SoldQty is zero or blank.
// Format the measure as a percentage in the model.
Situation: a regional manager wanted a custom report changed daily, which would have taken half my week. Task: support him without blocking other priorities. Action: I asked which decisions the report fed and found he needed only four numbers. I said no to the custom version, explained the impact on other work, and added those four metrics with a filter to the existing dashboard. Result: he got his numbers the same day and the team kept its capacity.
40
Your TO-BE process cuts the account-opening cycle time by half, but the compliance team rejects it. How do you handle it?
Ask compliance which specific controls they believe are weakened, and map each regulatory requirement to steps in the old and new flows. Often the control can be kept in a different form, such as automated checks, risk-based sampling or a post-approval audit instead of a manual pre-check. Bring compliance into the redesign rather than presenting a finished model. Document the agreed controls in the TO-BE diagram so audit can trace them. Speed never overrides a mandatory control.
Tips to crack the KPMG Business Analytics interview
Learn basics of IT audit, internal controls and risk frameworks, since advisory roles often combine data with governance work.
Practise structured answers for guesstimates and business problems, explaining assumptions clearly step by step.
Prepare examples of working with stakeholders and meeting deadlines, which are common themes in Big Four interviews.
Keep a one-minute story ready for every project on your CV: problem, your part, tools, result.
Practise the hard questions on this page aloud; KPMG interviewers usually ask follow-ups on your first answer.
Frequently asked questions
How many rounds are there in the KPMG Business Analytics interview?
Candidates usually go through 4 stages: Online assessment or screening test; Technical interview; Managerial or case round; Partner or HR round. The exact rounds change with the role, team, location and your experience, so confirm the process with the recruiter.
Which topics should I prepare for the KPMG Business Analytics interview?
This practice set leans on Product Analytics (6), Business Metrics and KPIs (5), Excel for Business Analysts (5), Process Modelling and BPMN (5). Revise those first, then practise explaining one or two of your own projects end to end, because most KPMG interviewers ask you to walk through real work.
Are these the exact questions asked at KPMG?
No. These are curated practice questions that match the skills and difficulty usually tested in consulting interviews like KPMG's. They are not a leaked or verified list of past questions.
Where does KPMG hire Business Analytics professionals in India?
KPMG has large teams in Gurugram, Bengaluru, Mumbai, Hyderabad, Pune. Openings and locations change often, so check the KPMG careers page and filter by city and experience.
Business Analytics questions for similar companies