FP&A interview questions from the candidate's chair

Explore the Finance interviews hub and the finance careers guide.

Most FP&A interview questions sound technical and turn out to be about judgment, the same judgment that separates each rung of the FP&A career path. Most people who reach a panel can clear the three statements, budget versus forecast and a variance. The real test comes after the number, when a senior leader wants a different answer or the panel asks who you are. For the full loop, the interviews guide maps every round.

What are they really checking in an FP&A interview?

They're checking whether you can be trusted with a number someone is going to act on.

I've interviewed for manager seats and hired analysts, so I've asked a few of these as well as answered them. For analyst seats, what a financial analyst interview is really scoring is a separate page.

The U.S. Bureau of Labor Statistics describes financial managers as people who spend much of their time analyzing data and advising senior managers. Advising is the word to notice.

There's a line I use when I'm hiring, and it's worth saying from the candidate's chair.

"Walk me through how the three statements connect"

This one is a warm-up, so keep it under a minute.

Net income from the income statement flows into retained earnings on the balance sheet and starts the cash flow statement. Non-cash items like depreciation get added back, changes in working capital adjust cash, capex sits in investing, and debt and equity moves sit in financing. Ending cash lands back on the balance sheet, and you're done.

Then stop talking, which is the hard part.

Finish with one line on where it breaks in real life, like a big accrual reversal that makes profit and cash tell two different stories in the same month. The SEC's guide to financial statements makes the same distinction.

"What's the difference between a budget and a forecast?"

A budget is the plan you commit to, and a forecast is your best current guess.

The budget gets set once a year and people get measured against it, while the forecast moves every time the business does.

Every budget I've ever seen has a plug in it.

Nobody has a plan for the plug.

Say that in the room and the interviewer knows you understand a budget is a negotiated number, not a forecast with a nicer font.

A practical example, because I've been on both ends of that negotiation. For five-plus years I was the one receiving targets from the top, and for about a year and a half I was the one giving them. The old loop went like this: we'd get the target, put in some plugs and say, okay, we'll figure this out next year, and then the first forecast of the year admitted the miss.

In the interview I'd say you can accept a target as an ambition without presenting it as a committed plan. So I showed leadership the bottom-up build, the drivers behind it, and what would need to be true to hit their number: the extra cost, the timing and who owns each piece. It wasn't the answer they wanted, and we still got the target.

Make the stretch explicit up front, before the first forecast admits it for you.

"Tell me about a variance you investigated"

They want to see you go past the math to the reason, and then to what changed because of it.

My best example is a sales versus target report. The ask was simple, pull actuals against target by rep, and it was a 10-minute calculation. What was really being decided on the other side of that spreadsheet was whether someone kept their job.

Then one line in the results looked wrong.

The reps in Canada hit their target every single time, and Canada was our most saturated market.

A saturated market should be the hardest place to hit a number, so I dug in. For a couple of reps, around 60 percent of their sales were purely inbound. Meanwhile some reps who looked weak were selling brand new products with sales cycles longer than a year.

I presented both views. If we just look at sales, this is the answer. If we look at sales efficiency, this is the answer.

A target tells you one thing. It doesn't tell you everything about how someone is performing, and that one red flag flipped the decision the report was built to support.

Give your variance story that shape:

Point 4 is the one most people skip, and it's the only one the interviewer cares about.

"Tell me about a time a senior leader brought you something urgent"

They're checking whether you can say no without saying no.

I separate requests into urgent and important, because every request I've ever got in finance was quote unquote urgent. So before I open a spreadsheet, I ask one question: when are you presenting this?

A senior leader once wanted a revenue number in the forecast that I knew they would miss. I didn't say no. I said, "I'm protecting you."

Then I walked them through what happens when the number doesn't land and they're the one in the room explaining it.

The first move is usually a question. When a leader pushes a number I don't agree with, I ask something fundamental they probably haven't thought about, because I'm coming from the finance seat. Say they tell me we're hiring two new salespeople this quarter, budget or not, and that's final. My first question is, do we have enough pipeline coverage for those salespeople to actually execute on the estimates you're bringing forward? That gets them thinking the way you think.

The other version of this question is the forecast that misses, and the business says it was finance's forecast. I ran into that early on, and after it I never again built a number in isolation. I built it with the business. If a business leader says they're going to hit a revenue number next year, I say, this is what has to be true for you to hit it. Are you still comfortable with this or not? When the actuals come in, those are assumptions we both agreed on.

"Tell me about a process you improved"

This is where "improved reporting" goes to die, so bring counts.

We had 7 teams each calculating the same scorecard, each with its own spreadsheet and its own way of doing it. Call it 30 to 40 spreadsheets, give or take.

Step one was standardizing the inputs so every team was feeding the same thing, and step two was a script to load them. Then the fun part, using AI to model them, and the refresh went from hours to minutes.

Have a lesson ready, because the follow-up is always what you'd do differently. Mine is that the worst thing you can do is automate something and do it efficiently the wrong way. Standardize first, automate second.

The other half came from tools I build for myself. One workflow wouldn't run well, and diagnosing the whole thing in one shot never worked. Breaking it into steps and finding the bottleneck in each one did.

So when a process breaks, name the slow step before you touch anything.

"Tell me about a failure"

Pick a real one, because a fake weakness gets spotted in about 4 seconds.

Mine was a planning process I simplified so well it stopped working.

The old process was heavy, so I cut it down, and I went a little bit overboard. The plan stopped capturing things the business actually needed to see.

The fix was listening more. I went back to which outputs were actually being used, and where, and made the inputs reflect that. The longer version, and the meeting I should have had first, is in my 30-60-90 day plan guide.

The lesson I'd say out loud is that a process isn't finished just because you've designed it. It has to be usable.

"How do you develop analysts?"

For manager roles this one carries weight, because they want to know your team's work will get used.

I had analysts whose numbers were always right and whose reports nobody acted on.

We fixed it with two words: so what?

I told them the what and the why are only half of it. Now I want the so what and the what next.

When a budget was going to miss, I didn't want a walkthrough of what happened. I wanted three options they'd already backed with research and conversations with the business, the one they'd recommend, and why. One analyst I coached that way was promoted to senior analyst about a year and a half later.

Getting to the number is the easy part. Saying what it means, and what to do about it, is the job.

"How do you present numbers to people outside finance?"

They're checking whether anyone outside finance will actually read what you build.

When I joined one team, the monthly management pack was an Excel file rebuilt from scratch every month. Nobody read it. Any question about history got the same answer, which was let me dig out last April's file.

We moved it into one live view in our BI tool. One place, all the history, no new file each month.

The numbers were right and the format was the problem, and showing you can tell those two apart is the whole answer.

Outside finance I'd rather show a variance as percent growth than as a dollar amount. Sometimes the absolute number means nothing if you don't know the baseline.

Sometimes the format problem is one word. On a forecast deck, part of a budget was labelled available. It wasn't. That money "has been released and is no longer in budget," so I had the label changed. A reader who sees available assumes the money is still there.

Show what was true before, too. In FP&A a lot of decisions get made on the fly, and versions go back and forth. Knowing what was true before, and why it changed, is just as important as the final number. You also have to understand the history.

What is the case question testing, and how did I lose one?

A case tests how you'd run the conversation with a leader who doesn't have the answer yet, and I lost a case round by forgetting that.

I went straight into the calculations. I was using my accounting brain. Afterward, the interviewer's feedback was that I needed to get to the other person's issues faster. The number was assumed. What got scored was how I led the conversation.

So slow down and lead it like a working session with a budget holder:

• Say how you'll set it up before you calculate anything.
• Name your assumptions and ask if they're fair.
• Get the number.
• Say what it means for the business, what you'd recommend, and what would change your mind.

How do FP&A manager and director interviews differ?

The questions sound the same at every level, and what changes is the answer they want.

I saw senior director on a posting once and wrote it off as out of my league. The title made the call before I'd read the work. Then the first interview showed me what the work really was, and I wanted it. Don't let the title decide. Read what the seat owns, then decide which of the three answers below it wants from you.

For the director answer, bring a what-if. One quarter, I wanted to know what our forecast would do if we changed the exchange rates, so I asked for a sensitivity, a table where I could change an input and see the result move. It was there to show whether we should update the rates at all. That business was heavily FX related, so you had to know which sensitivity mattered, and picking the right one shows you know the business. Name the decision your what-if was built for.

If you haven't managed people yet, answer the leadership question with a time you led without the title: a close you ran, a new analyst you got up to speed. If you're not sure what the manager seat owns week to week, the FP&A manager job, from inside it walks through it.

What's the hardest question in an FP&A interview?

The hardest questions are the follow-ups, and the personal ones come once the panel is confident you can do the job.

The surface questions are easy if your stories are organized and you know them in and out.

In one in-person round with three people, the controller said she could see I was a builder and a finance person, so who was I? I told her I'm the person who can sit in a technical room, understand the language and what's going on, then turn around and explain it to someone non-technical, and back again.

Then she asked which way I lean. I paused, then said I don't think I'd have to choose, because that is my superpower. If I had to pick one lane, I'd be miserable.

She told me afterward there was no right answer. She was curious, and she saw a bit of herself in me.

That's the fit part, and it works both ways. You've been reading the seat since the posting, and in this round the seat reads you back. Before it, write one true line on who you are at work and the part you'd hate to give up.

More FP&A interview questions, and what they're really asking

The rest of the loop is shorter questions, and each one still has a second question under it.

• "Why FP&A?" They want what you're moving toward, and "I'm tired of close" doesn't count. Mine is that every company has a fundamental finance problem it's still trying to fix. If everything tied and the forecast was perfect, the whole finance department would be run by AI agents. It's messy enough to need your skills and your judgment, and I like solving problems.
• "Walk me through a model you built." They're checking you understand the business behind the formulas. Start with the question the model answered, then the drivers, then what came out.
• "How strong is your Excel?" They want to know you can work without someone checking behind you. Name what you actually use.
• "What BI or planning tools have you used?" They're asking whether they'll have to train you. Name the tool, then what you built in it.
• "What do you do when a forecast turns out wrong?" They're checking that you explain a miss instead of hiding it. Name the driver that moved it and what you changed after.
• The KPIs you'd track for a business like theirs. It checks your prep. Pick 3 and tie each to a decision someone makes.
• "How do you make sure your numbers are right?" They want a checking habit. Describe the tie-out you do before anything leaves your desk.
• "What's the difference between FP&A and accounting?" It tests whether you can explain your own job in plain words. Accounting tells you what happened. FP&A tells you what's likely next and what to do about it.
• "How do you work with the accounting team at close?" They're checking you respect the close. Say what you need from them, when you need it, and what you give back. The controller interview guide covers the other side of that table.
• "What would you do in your first 90 days?" They want to see a plan. Bring a short 30-60-90 day plan: learn the numbers, fix one thing, own one process.
• "What are your salary expectations?" It checks your homework. Say the posted range works and where in it you'd put yourself. Skip the single number. They anchor to it. For Canada, see Job Bank's wage reports. For FP&A ranges by level, see what FP&A pays from analyst to director.

What should you ask at the end of an FP&A interview?

Ask questions that show you're already thinking about the seat.

Two I've actually asked: where do you need this person to make the most difference, and what's the least glamorous part of the job? The second answer tells you what the recurring work looks like.

A few more:

• Who reads the numbers this team produces, and what they do with them.
• When the plan gets set, and how much it moves after that.
• What changed in the last reforecast, and why.
• What the person in this seat will be doing in month 3 that nobody is doing today.

Skip anything you could have read on their careers page, and skip pay until they raise it.

How do you prepare for an FP&A interview in one evening?

Five true stories and the numbers inside them will carry you further than 45 memorized answers.

Write each one in 5 lines with a number in it, then say them out loud until you can do each in 90 seconds. Most FP&A questions are one of these 5 stories wearing a different hat.

Tonight, pick one story and say it out loud.