Financial analyst interview questions, with worked answers

How to use these questions

Practise every short answer aloud until it sounds like you, not like a textbook. Then add one model or analysis you really built: the question it answered, the key assumption, and what the manager decided. Analyst interviews often follow with a modelling test, so the logic must be yours.

What a financial analyst is expected to know

• The statements: the income statement, the balance sheet and the cash flow statement, how they link, and how the accounting rulesAsked in 18% of financial analyst ads (as of 5 Nov 2026).Read the three financial statements and know how IFRS or US GAAP change the figures. change them.
• Modelling: three-statement models, driver-based forecasts, scenarios and sensitivities in ExcelAsked in 26% of financial analyst ads (as of 5 Nov 2026).Build models others can audit: three statements, scenarios, lookups, pivot tables and checks that flag errors..
• Valuation: discounted cash flow (DCF), the cost of capital (WACC) and multiples such as EV/EBITDA.
• Planning: budgets, forecasts and variance analysis, the work FP&A shares with controllingAsked in 37% of financial analyst ads (as of 5 Nov 2026).Budgets, cost centres and variance analysis, the planning work FP&A shares with controlling..
• Data: SQLAsked in 6% of financial analyst ads (as of 5 Nov 2026).Pull and join data from the ERP or the data warehouse yourself: sales by product, costs by month., Power BIAsked in 6% of financial analyst ads (as of 5 Nov 2026).Build the dashboards of the forecast and the KPIs, from one data model the whole team uses., basic statisticsTrends, seasonality and simple regressions, so a forecast rests on more than last year's figure. for trends and seasonality, and in some teams PythonAsked in 2% of financial analyst ads (as of 5 Nov 2026).Named in some ads, for automating data work and larger analyses; Excel and SQL come first..
• The last mile: presentingTurn a model into three slides and a recommendation a manager can decide on in minutes. a recommendation and writingAsked in 19% of financial analyst ads (as of 5 Nov 2026).Write the commentary behind the numbers, and say what you do not know. the commentary.

The statements, models and valuation

Walk me through the three financial statements and how they link.

What they check: the basic map every model is built on.

Short answer: "The income statement shows sales, costs and profit for a period. The balance sheet shows what the company owns and owes at a date. The cash flow statement starts from net profit, adds back non-cash costs such as depreciation, adjusts for working capital, then adds investing and financing flows. Its ending cash is the cash on the balance sheet, and net profit flows into equity."

The full answer, in five links:

1. Profit to equity. Net profit, less dividends, adds to retained earnings in equity.
2. Profit to cash. The cash flow statement starts from net profit and reverses what is not cash, such as depreciation.
3. Working capital. A rise in receivables or stock uses cash; a rise in payables frees it. Each change on the balance sheet shows in operating cash flow.
4. Investment. Capital spending (capex) adds to fixed assets on the balance sheet and is an investing outflow; depreciation reduces fixed assets and is a cost in the income statement.
5. Financing. New debt or equity brings cash in; repayments and dividends take it out. The ending cash must equal the balance sheet cash, which is the check that the model balances.

A weak answer: listing the three statements without the links. The links are the question.

Follow-up: "Depreciation rises by ten. Walk it through." Profit before tax falls by ten, tax falls by the tax rate times ten, net profit falls by the rest; cash rises by the tax saved, because depreciation is added back; fixed assets fall by ten and equity by the fall in profit, so the balance sheet still balances.

How would you build a three-statement model for a company's five-year plan?

What they check: structure, drivers and checks, so the model can be trusted and changed.

Short answer: "I keep inputs, calculations and outputs apart. I start from the historical statements, then forecast the drivers: volumes and prices for revenue, margins and cost per unit, working capital in days, capex and depreciation, debt and interest. The statements are built from those, cash closes the loop, and a check row shows that the balance sheet balances in every year."

The full answer: One sheet of assumptions, clearly marked, that a manager can change; calculation sheets that only use formulas; outputs and charts. Revenue comes from drivers, not a growth rate typed in. Working capital follows days: receivables as days of sales, stock as days of cost, payables as days of purchases. A debt schedule handles borrowing and interest. Interest on average cash can create a circular reference; switch it off with a toggle or use opening balances. Every year needs the balance check and a cash check.

A weak answer: "I grow everything by five percent." That is a spreadsheet, not a model. Follow-up: "The balance sheet does not balance in year three." Check that every balance sheet change has a matching cash flow line, starting with the one added last.

Walk me through a DCF valuation.

What they check: that you know each step and where the judgement sits.

Short answer: "I forecast the company's free cash flow for five to ten years, meaning the cash its operations make after tax and investment, before paying lenders and owners. I discount each year back to today at the WACC, the weighted cost of its capital, add a terminal value for the years beyond, and get the enterprise value. Less net debt, that gives the equity value."

The full answer, in five steps:

1. Free cash flow. Operating profit (EBIT) after tax, plus depreciation, minus capex, minus the increase in working capital.
2. Forecast period. Long enough for the business to settle into a steady state, often five to ten years.
3. Discount rate. The WACC mixes the cost of equity and the after-tax cost of debt by their weights.
4. Terminal value. Either the last year's cash flow growing at a modest rate forever, or a multiple of the last year's EBITDA; it is usually a large part of the total value.
5. From enterprise to equity. The discounted flows plus the discounted terminal value give the enterprise value; subtract net debt and similar claims for the equity value, and divide by the shares for a value per share.

A weak answer: reciting the formula without saying which assumptions matter most. Follow-up: "Which input would you test first?" The WACC and the long-term growth rate, because the terminal value moves most with them.

How do you estimate the WACC, and what moves it?

What they check: that you understand the cost of capital, not just its formula.

Short answer: "The WACC is the average return lenders and owners expect, weighted by how much each provides at market values. The cost of debt is the interest rate after the tax saving. The cost of equity is usually estimated with CAPM: a risk-free rate plus beta, the share's sensitivity to the market, times the market's extra return over risk-free assets."

The full answer: Higher interest rates raise the risk-free rate and so the WACC. A riskier business has a higher beta. More debt lowers the WACC at first, because debt is cheaper and its interest saves tax, but past a point the risk raises both costs. For a company with no listed shares, take the beta of comparable listed companies and adjust it for the debt level. Use one currency throughout: cash flows in euros need a euro rate.

A weak answer: "We use ten percent." Follow-up: "Rates rise by one point. What happens to the valuation?" The WACC rises, every discounted flow falls, and the terminal value falls most.

What is terminal value, and why does it matter so much?

What they check: awareness that most of a DCF's value sits in one assumption.

Short answer: "Terminal value is the value of all the cash flows after the forecast period, put into one number. It often makes up most of the enterprise value, so a small change in the long-term growth rate or the exit multiple moves the result a lot. I always show it as a share of the total and test it."

The full answer: With the growth method, the last year's cash flow, grown once, is divided by the WACC minus the growth rate; the growth must stay below the economy's long-term growth, or the company would one day be bigger than the economy. With the exit multiple, the last year's EBITDA is multiplied by what similar companies trade at. Do both and compare: if the implied multiple of the growth method is far from the market, one of the assumptions is wrong.

A weak answer: "It is just the last step." Follow-up: "The terminal value is almost all of the value." Lengthen the forecast until the business is steady, and check the growth and margin assumed for the final year.

How do you build scenarios and sensitivities into a model?

What they check: that you show managers a range and the inputs that matter, not one false number.

Short answer: "I pick the few drivers that move the result most, such as volume, price and a main cost. A scenario switch changes them together into a base, a downside and an upside case with a story behind each. A sensitivity table shows the result for a grid of two inputs, such as WACC and growth. Then I say which risk deserves attention."

The full answer: Scenarios are consistent stories: in a downside, volumes fall and discounts rise together. Sensitivities move one or two inputs to show how fragile the result is. A tornado chart ranks the inputs by their effect. Keep the switch on the assumptions sheet, so every output follows it. Point the reader to the break-even, the value of an input at which the decision changes.

A weak answer: "I add ten percent and take ten percent off everything." Follow-up: "The downside breaks the bank's loan terms." Show the cash and the covenant test by quarter, and what actions would avoid the breach.

Forecasts, variances and KPIs

How do you build a revenue forecast?

What they check: that you forecast from drivers the business understands.

Short answer: "I split revenue into its drivers, such as customers times orders times price, or units by product times price, and forecast each with the people who know it: sales for volumes, pricing for prices. I check the result against trends and seasonality in the history, and track how accurate each forecast was."

The full answer: Start from the order book or the pipeline for the near months, and from drivers further out. StatisticsTrends, seasonality and simple regressions, so a forecast rests on more than last year's figure. help with seasonality and trend: compare the same month across years, and use a simple regression where one figure drives another. Combine the two views, the team's and the data's, and write down where they differ. Measure forecast error every month; a forecast that is always too high needs a correction, not a new spreadsheet.

A weak answer: "Last year plus growth." Follow-up: "Sales says the next quarter will be strong; your data says not." Ask what they know that the data does not, such as a signed contract, and show both views to the decision maker.

Revenue came in below forecast. How do you explain it?

What they check: a clear bridge from forecast to actual, and the cause behind each step.

Short answer: "I build a bridge from the forecast to the actual: the effect of volume, price, mix and currency, each as a bar, so the reader sees what caused the gap. Then I find the reason behind the biggest bars with the business, and say whether it reverses next month or changes the full-year forecast."

The full answer: A variance bridge, often drawn as a waterfall chart, starts at the forecast and adds or subtracts each effect until it reaches the actual figure. Currency matters in international groups: a stronger euro lowers the euro value of dollar sales without any change in the business. Separate timing, such as an order that slipped a week, from lasting effects, such as a lost customer. The forecast for the rest of the year must reflect the lasting ones.

A weak answer: "The market was difficult." Follow-up: "Which part do you trust least?" Say it, and how you would confirm it.

What is EBITDA, and what are its limits?

What they check: that you use the measure and know what it hides.

Short answer: "EBITDA is earnings before interest, tax, depreciation and amortisation: operating profit with the non-cash charges for assets added back. It is easy to compare across companies with different debt and tax. But it ignores capex, working capital and tax, so it is not cash, and a company with high EBITDA can still run out of money."

The full answer: Many lenders and buyers use EBITDA multiples, such as EV/EBITDA, and loan terms often use debt over EBITDA. Accounting rules change it: under IFRS 16, lease payments become depreciation and interest, so EBITDA rises; under US GAAP most operating leases still show as one operating cost; under the German HGB most leases stay off the balance sheet. Compare companies on the same basis. Adjusted EBITDA, which removes items a company calls one-off, needs a close look at what was removed.

A weak answer: "EBITDA is the same as cash flow." Follow-up: "Two companies have the same EBITDA. Which is worth more?" The one that needs less capex and working capital to grow, because it turns more of that EBITDA into cash.

How do you choose the KPIs for a business you do not know yet?

What they check: that you find the few figures that drive this business, not a standard list.

Short answer: "I learn how the business makes money: what it sells, to whom, at what cost, and what limits growth. The KPIs follow from that: for a subscription business, new customers, churn and revenue per customer; for a plant, output, scrap and utilisation; and for any company, margin, cash and working capital. I agree each definition with the owner."

The full answer: A KPI tree helps: start from profit or cash at the top and break it into the drivers below, down to figures a team can change. Choose a mix of leading indicators, which warn early, and lagging ones, which confirm. Fewer is better; a dashboard in Power BIAsked in 6% of financial analyst ads (as of 5 Nov 2026).Build the dashboards of the forecast and the KPIs, from one data model the whole team uses. with eight figures people watch beats one with forty nobody reads.

A weak answer: "Revenue, profit and costs." Those are results, not drivers. Follow-up: "Sales and finance define churn differently." Agree one definition, write it down, and report only that one.

How do you analyse working capital and cash flow?

What they check: that you look past profit to the cash a business needs.

Short answer: "I look at the cash conversion cycle: days sales outstanding, how long customers take to pay; days inventory, how long stock sits; and days payables, how long we take to pay. I compare them over time and with peers, find which customers, products or suppliers drive them, and model the cash a change would free."

The full answer: Working capital is receivables plus stock minus payables. In a forecast it follows from days applied to sales and costs, so growth needs cash before it brings any. A thirteen-week cash forecast, built from expected receipts and payments, shows short-term needs; the three-statement model shows the longer view. Watch for one-off effects at period end, such as suppliers paid late to make the balance look better.

A weak answer: "Cash is treasury's job." Follow-up: "Sales want longer payment terms to win a big customer." Model the cash cost of the extra days against the margin of the deal, and show both.

Excel, data and presenting

How do you keep a large Excel model free of errors?

What they check: discipline, because a wrong model is worse than none.

Short answer: "Clear structure, one formula per row across all periods, no numbers typed inside formulas, inputs in one colour and in one place, and check cells that flag any error: the balance sheet balances, totals agree with the source. Before a model is used, someone else reviews it, and I keep a log of versions."

The full answer: Most model errors are simple: a range one row short, a sign the wrong way, a hard-coded number left from a test. Consistent formulas make them easy to spot. Keep calculations simple enough to follow; a long nested formula split into steps is easier to check. Use named inputs or a clear assumption sheet, and lock the cells that should not change. A short summary of the logic at the front helps the next reader.

A weak answer: "I check it carefully at the end." Follow-up: "Your model and the accounts disagree by a small amount." Find it before anyone uses the model; small differences often hide a large error in the opposite direction.

When do you use Power BI or SQL instead of Excel?

What they check: that you pick the tool by the job, and can work with data beyond a spreadsheet.

Short answer: "Excel is for models and one-off analysis. When the same report is rebuilt every month from large data, I move it to Power BIAsked in 6% of financial analyst ads (as of 5 Nov 2026).Build the dashboards of the forecast and the KPIs, from one data model the whole team uses. on a fixed data source, so it refreshes on its own. When I need data from an ERP such as SAPAsked in 22% of financial analyst ads (as of 5 Nov 2026).The ERP many companies run; most analysts pull their actual figures from it. or the data warehouse, I write SQLAsked in 6% of financial analyst ads (as of 5 Nov 2026).Pull and join data from the ERP or the data warehouse yourself: sales by product, costs by month. to pull and join it myself instead of asking for exports."

The full answer: Excel struggles with millions of rows and with many people using one file. SQL handles the joining and filtering at the source; Power BI holds a data model with agreed definitions, so every dashboard shows the same revenue. TableauAnother common dashboard tool, found in some companies instead of Power BI. does the same job in some companies. PythonAsked in 2% of financial analyst ads (as of 5 Nov 2026).Named in some ads, for automating data work and larger analyses; Excel and SQL come first. helps with large or messy data, and with analysis that would be slow by hand. The model logic still often lives in Excel.

A weak answer: "Excel can do everything." Follow-up: "Write in words the query that gives sales by product and month." Take the sales table, join the product table on the product key, group by product and month, and sum the amount.

How do you present a recommendation to senior managers?

What they check: that you lead with the answer and keep the detail for questions.

Short answer: "I start with the recommendation and the one or two figures behind it, then the main risk and what would change my mind. The model stays in the appendix. One clear chart beats a table of numbers. I prepare for the three questions they are most likely to ask."

The full answer: Senior managers have little time and many decisions. A structure that works: the question, the answer, why, the risks, and what is needed from them. Round the numbers and use the same ones throughout. Say what you do not know; it builds trust. If the room disagrees with an assumption, show what the result would be with theirs.

A weak answer: walking through the model sheet by sheet. Follow-up: "The CFO asks for the answer in one sentence." Have that sentence ready before the meeting.

FP&A manager, head of FP&A and finance director

These questions come up for senior financial analyst, FP&A manager, head of FP&A and finance director roles.

As an FP&A manager, how do you run the planning cycle and the team?

What they check: that you can deliver a company's plan on time through other analysts and budget owners.

Short answer: "I set the calendar, the template and the shared assumptions, give each analyst the units they partner, and review their work at fixed points. I keep one model or planning tool as the single source, track open items each week, and make sure leadership gets a clear view of the gaps to target in time to decide."

The full answer: The plan fails in two places: late inputs and too much detail. Agree deadlines with the budget owners' managers, plan on drivers, and keep the number of rounds low. As a manager, review the logic, not every cell, and coach analysts on the story behind the figures. Between planning rounds, the team runs the forecasts and the monthly reporting, and improves one process each quarter.

A weak answer: "I do the hard parts myself." Then the team never learns them. Follow-up: "One unit is always late." Find out why, help once, then escalate with the calendar agreed by leadership.

As head of FP&A or finance director, how do you link the plan to the strategy and the board?

What they check: at the top of the ladder, turning strategy into numbers and numbers into decisions.

Short answer: "I turn the strategy into a few financial targets and the investments behind them, build a three- to five-year plan that leadership agrees, and break the first year into the budget. Each quarter I show the board where we stand against the plan, what changed, and the choices ahead, with scenarios."

The full answer: A strategy says where to play; the long-term plan says what it costs and returns. Link each strategic initiative to its costs, its expected benefits and an owner, and track them like any other figure. The board wants few numbers, honest risks and options, not detail. A head of FP&A also builds the team, the tools and the forecasting habit that make the figures trusted.

A weak answer: "Strategy is the CEO's job; I do the numbers." Follow-up: "The board wants more growth than the plan shows." Show what it would take: the investment, the risk and the cash, and let them choose.

Your CV for these questions

An analyst interview tests the claims on your CV, often with a model of their own, so the lines below prepare the answers above. Place one of each under the job where you did it:

A score against a real financial analyst job then shows which of these lines that employer wants, and which your CV still has to prove.