Finance interview questions — and how to actually answer them

I've been on both sides of the finance interview table — as a candidate early in my career, and more recently as someone who has hired for finance roles and built finance functions from scratch. The gap between a good answer and a great one almost always comes down to the same thing: structure. Lead with the framework, anchor it with a number, and close with a real example from your own work. Interviewers aren't testing whether you've memorised a textbook. They're testing whether you think like a finance professional.

This is a guide to the questions I've actually asked and been asked — across accounting, FP&A, treasury, systems, and behavioural — with model answers that reflect how someone who has done this work actually speaks about it.

A finance interview doesn't test what you've memorised. It tests how you think about money, risk, and the information that connects the two.

Accounting fundamentals

"Walk me through the three financial statements and how they connect."

Start with the P&L: revenue less all costs gives you net income. That net income flows in two directions. First, it lands on the balance sheet as an increase to retained earnings under equity — so any profit you make is permanently recorded in the net worth of the business. Second, it becomes the starting line of the cash flow statement.

From that starting point, the cash flow statement makes three adjustments. You add back non-cash charges like depreciation and amortisation — these reduced net income but never left the bank. You then account for working capital movements: if debtors grow, cash is absorbed; if creditors grow, cash is released. Finally, you subtract capex — the actual cash paid for fixed assets, which never touched the P&L directly. The result is operating cash flow.

The ending cash balance from the cash flow statement then reconciles to the cash line on the balance sheet. So the three statements form a closed loop: P&L feeds retained earnings on the balance sheet and operating cash flow on the cash flow statement; capex flows to fixed assets on the balance sheet; debt drawdowns and repayments flow through both financing cash flows and the liabilities side of the balance sheet.

"What's the difference between profit and cash flow?"

Profit is an accounting concept — it's accrual-based, meaning revenue is recognised when earned and expenses when incurred, regardless of when cash moves. Cash flow is the actual movement of money in and out of the business.

The gap between the two is driven mainly by working capital. A manufacturing business that sells on 90-day credit terms could show strong profit on the P&L while its bank account is empty — because the cash from those sales hasn't arrived yet. Add in the fact that it's paying suppliers in 30 days, and you have a profitable business that's technically running out of cash. This is how solvent businesses go into liquidity crisis.

The other driver is capex: buying a machine for ₹2 Cr hits the cash flow statement immediately but only hits the P&L as depreciation spread over, say, ten years. So in year one, cash is ₹2 Cr worse than profit suggests. Understanding this gap — and managing the working capital cycle actively — is one of the most practically valuable skills in finance.

"Deferred tax — what is it, in one minute?"

Deferred tax arises because the income you report in your financial statements and the income the tax authority taxes you on are calculated differently, and those differences unwind over time. The classic example is depreciation: companies often depreciate assets faster for tax purposes (using WDV at higher rates) than for accounting purposes (SLM at lower rates). In early years, taxable income is lower than accounting income, so you pay less tax now than your P&L expense suggests — that difference is a deferred tax liability, because you'll owe it eventually. The reverse — where you've paid more tax now than your P&L reflects — creates a deferred tax asset.

The key distinction: it's not real cash today. It's the tax consequence of timing differences between accounting and tax treatment, recognised now so the P&L isn't distorted.

FP&A and analysis

"How would you build a budget vs actual variance process?"

Start with the drivers, not the line items. A variance report that just shows "revenue was ₹12 L lower than budget" is useless. You want to decompose it: was it volume, price, or mix? For manufacturing, was it the yield rate, the utilisation, or the raw material cost? Building the budget at the driver level first means your variance analysis is pre-structured — the drivers are the same, you're just filling in actuals.

On process: I run a T+5 close cadence — books closed by the fifth working day, variance pack distributed to cost-centre owners by T+5, written explanations due from owners by T+7. The pack uses a materiality threshold — typically ±5% or ±₹10 L, whichever is lower — so owners are only explaining variances that actually matter. Below threshold, flag it but don't escalate.

The discipline of making cost-centre heads own their variances in writing every month is underrated. It forces them to understand their numbers in a way that a shared Finance dashboard never does. Within three months of running this at a new entity, the quality of the next budget cycle improves significantly — because people have been forced to explain what actually drove their costs every month.

"CAGR vs IRR vs XIRR — when do you use each?"

CAGR (Compound Annual Growth Rate) is the simplest: it measures the annualised growth rate between two endpoints, assuming smooth compounding. Use it for benchmarking business growth over time — revenue, AUM, headcount. The formula is (End/Start)^(1/n) − 1. It ignores intermediate values and says nothing about the path — two businesses can have the same CAGR but very different volatility in between.

IRR is the discount rate that makes an investment's NPV equal to zero — it's a project-level return metric. Use it when evaluating capex projects, acquisitions, or any investment with an upfront outflow and future inflows. The limitation: it assumes all interim cash flows are reinvested at the IRR itself, which is often unrealistic for high-IRR projects. It also assumes regular, annual periods.

XIRR solves the period problem. It calculates the IRR for cash flows that occur on irregular dates — which is almost always the reality in actual investments. Use XIRR for portfolio returns, loan assessments, or any scenario where cash flows don't land neatly on anniversary dates. In Excel, you just pass the cash flow column and the date column. I always default to XIRR over IRR for real-world analysis.

Treasury & banking

"What is DSCR and what's a healthy level?"

DSCR — Debt Service Coverage Ratio — measures whether a business generates enough operating cash flow to cover its debt obligations. The formula is Net Operating Income ÷ Total Debt Service, where debt service includes both principal repayments and interest due in the period.

A DSCR of 1.0x means the business earns exactly enough to cover its debt — no margin. Banks typically want to see 1.25x to 1.5x as a minimum covenant for term loans, and higher for project finance or infrastructure lending where cash flows are more variable. Anything below 1.0x means the business is covering debt from reserves or new borrowing, which is unsustainable.

In practice, I pay attention to DSCR both at sanction — it's part of the credit appraisal — and on a rolling quarterly basis, because a business that passes DSCR at sanction can drift below covenant if EBITDA compresses or if a large bullet repayment is due. Banks monitor it; so should the CFO.

"How does buyer's credit work and why use it?"

Buyer's credit is a short-term foreign currency loan that an Indian importer raises through a domestic bank to pay an overseas supplier at sight. Instead of paying from your own rupee working capital, the bank draws down foreign currency funds from an overseas correspondent — typically at SOFR plus a spread — and remits them directly to the supplier. You then repay the bank in rupees at the due date, absorbing the exchange rate.

The reason to use it: the all-in cost is usually lower than domestic rupee borrowing. If SOFR is 5% and you add a 60–80 bps bank spread, you're borrowing at roughly 5.6–5.8% in dollar terms. On a rupee equivalent basis — accounting for hedging cost or expected depreciation — that's often still cheaper than a CC limit at 8–9%. The other benefit is that it frees up your domestic working capital limit for other uses. The risk, of course, is currency: if the rupee depreciates sharply before repayment, your effective cost rises. That's why buyer's credit ideally pairs with a forex hedge.

ERP & systems

"You're implementing an ERP — what do you do first?"

Write the SOPs before you touch the configuration. This is the single most important sequencing decision in any ERP implementation, and it's the one most commonly skipped.

The temptation is to open the ERP, start creating item masters and posting groups, and figure out the process as you go. The result is an ERP that encodes your current confusion rather than the process you actually want. Every rework in an ERP — changing a dimension structure, reversing a posting group, rebuilding approval workflows — takes ten times as long after transactions have been posted as it would have taken before go-live.

The right sequence: map the business process end-to-end first (purchase requisition → PO → GRN → quality release → vendor invoice → payment approval → bank payment). Write a one-page SOP for each step. Assign a process owner. Then, and only then, map each SOP step to an ERP transaction. When I implemented Microsoft Dynamics 365 Business Central for a newly acquired entity, we spent the first three weeks in a conference room writing SOPs with the warehouse, QC, and accounts teams before a single item master was created. The go-live was cleaner than any implementation I've seen that went straight to configuration.

Behavioural

"Tell me about a time you built something from scratch."

Situation: When a large Pharma API company executed a full asset buyout, I inherited an entity with no SOPs, no ERP, no banking relationships, and no audit trail. The finance function was genuinely zero — there was no chart of accounts, no close process, no MIS. The workforce was in place, but the financial infrastructure wasn't.

Task: My mandate was to get the entity to full, audit-ready financial operations as quickly as possible — including a live ERP, functional banking, a structured close process, and a statutory audit completed.

Action: I sequenced the build deliberately. Week one was entirely cash and banking — current account, AR migration, daily cash position on Excel. Weeks two through six were SOPs and ERP: I wrote the process documentation first, then mapped it to Business Central, then configured. Months three and four were close cadence — T+5 target from month two, achieved by month four. Banking facilities — working capital limit, ECLGS 5.0, buyer's credit — were built in parallel with the group's relationship bankers. IFC framework and internal audit cycle were set up by month six.

Result: Full statutory audit completed and signed off in seven months from the buyout date. Banking limits reached approximately ₹125 Cr. The entity went from zero financial infrastructure to a live Power BI MIS consumed by promoters and management in real time — all within the financial year of acquisition.

"Describe a process you automated."

The most impactful automation I've run was rebuilding the RBI compliance reporting process for BRC (Bank Realisation Certificates) and EDPMS (Export Data Processing and Monitoring System) reconciliation. Previously, this was a manual, spreadsheet-driven process: someone would download export data from the ERP, download bank entries, manually match them, and then build a reconciliation file for submission. It took roughly two days per cycle and was error-prone — mismatches were caught late, and late EDPMS reporting attracts RBI scrutiny.

I rebuilt it in Python: the script ingests the ERP export ledger, the bank's SWIFT confirmation file, and the EDPMS portal data; matches on invoice number, date range, and amount within a tolerance; flags unmatched items with a root-cause category (timing difference, short payment, forex gain/loss); and outputs a clean Excel with the reconciliation summary and a pre-formatted submission file. The two-day cycle became a two-hour one. More importantly, the error rate dropped to near zero because the matching logic is deterministic and consistent — no human transcription errors.

Closing advice for candidates

The questions above cover most of what gets asked in finance interviews at the manager and senior manager level. But the meta-skill is more important than any individual answer: think out loud, in structure. Interviewers forgive wrong numbers far more readily than they forgive incoherent thinking. If you don't know the exact DSCR threshold a specific bank uses, say "banks typically require 1.25x to 1.5x as a minimum covenant" — that's a real, defensible answer that shows you understand the concept.

Prepare two or three real examples — things you've actually built, fixed, or automated — and practice telling each one in the STAR format at roughly 90 seconds. Most behavioural questions are answered adequately in two minutes and well in three. Longer than that and you've lost the room.

Finally, ask good questions at the end. The quality of your questions signals the quality of your thinking more than almost anything in the interview. "How does the finance function here interact with the business unit heads?" or "What does the close cycle look like today?" tells the interviewer you've already started thinking about the job — not just the interview.