16 Problem Solving Interview Questions and Sample Answers (2026)

Last updated: November 2026

Quick Answer

Problem solving interview questions test whether you can define a messy problem, find its real cause, weigh options, act without perfect information, and check that your fix worked. Interviewers listen for your process more than your answer. Rehearse the 16 questions below out loud in OphyAI Interview Practice, and use Interview Copilot for real-time answer suggestions when the live interview runs on Zoom, Teams or Google Meet.

#	Question	What the interviewer is testing
1	Tell me about a time you solved a difficult problem at work	Your full process, start to finish
2	How do you approach a problem you have never seen before?	A repeatable method
3	Tell me about a time you found the root cause of a recurring problem	Digging past symptoms
4	Describe a time you had to solve a problem with incomplete information	Comfort with ambiguity
5	Tell me about a time your first solution did not work	Ownership and iteration
6	Describe a decision you made based on data	Analytical judgment
7	You have three urgent tasks and time for two. What do you do?	Prioritization under pressure
8	Tell me about a time you spotted a problem before it happened	Initiative
9	How would you estimate how many coffees a cafe sells in a day?	Structured reasoning out loud
10	Tell me about a time you solved a problem with limited resources	Resourcefulness

Why employers ask so many problem solving questions

Problem solving is the attribute US employers most often say they look for. In NACE’s Job Outlook 2025 survey, nearly 90% of employers said they want evidence of problem-solving skills on graduates’ resumes, ahead of teamwork at nearly 80%. O*NET, the US Department of Labor’s occupation database, defines “complex problem solving” as “identifying complex problems and reviewing related information to develop and evaluate options and implement solutions,” and lists it for 910 occupations. That definition is a good checklist: identify, research, generate options, choose, implement. A strong answer shows each step.

Behavioral vs situational vs analytical

You will meet three shapes of problem solving question. Behavioral questions ask what you did (“Tell me about a time…”); answer with one real STAR story (Situation, Task, Action, Result). Situational questions ask what you would do (“What would you do if…”); answer with your steps in order, then a short real example. Analytical questions hand you a small puzzle or a business problem on the spot; think out loud, state assumptions and structure the problem before you calculate. MIT’s career office suggests weighting a STAR answer about 20% Situation, 10% Task, 60% Action and 10% Result. For problem solving, put most of that 60% on how you diagnosed the problem, not only on what you did to fix it.

The Questions, With Sample Answers

Sample answers below are illustrations for a fictional candidate with a few years in operations and analytics roles. Their numbers are examples, not data. Swap in your own true stories.

1. “Tell me about a time you solved a difficult problem at work.”

Why they ask it: It is the broadest version of the question and lets them see your whole process: how you framed the problem, what you checked, what you chose and how you measured the result.

Sample answer: “At my last job, our warehouse missed same-day shipping on about 12% of orders for two months, and customers were complaining. I owned the fulfillment report, so my manager asked me to find out why. I pulled six weeks of late orders and sorted them by time, product and picker. Most of the late orders were placed between 1 and 2 p.m., right when the lunch break left the floor short-staffed. I proposed staggering lunch breaks across two shifts, tested it for one week in one zone, and then rolled it out. Late shipments fell to under 3% within a month, and the staggered schedule is still in place.”

Weak answers: A story with no diagnosis, where you jumped to a fix, or a “problem” so small it shows nothing, such as fixing a printer jam.

Practice this in a voice mock in Interview Practice and see whether the follow-up questions can find the step where you diagnosed the cause.

2. “How do you approach a problem you have never seen before?”

Why they ask it: They want a repeatable method, not luck. This is a process question, so describe your steps, then prove them with a short example.

Sample answer: “I use four steps. First, I define the problem in one sentence and confirm it with whoever raised it, because half the time the real problem is different. Second, I gather facts: data, the people closest to it, and what has already been tried. Third, I list two or three options and compare them on cost, speed and risk. Fourth, I pick one, set a way to measure it, and check back after a set time. When our billing system started rejecting a new client’s invoices, I used exactly that: the ‘rejection’ turned out to be a tax code the system did not recognize. We added the code and the client’s invoices went through the same week.”

Weak answers: “I just figure it out,” or a long list of steps with no example to show you actually use them.

3. “Tell me about a time you found the root cause of a recurring problem.”

Why they ask it: Many people fix symptoms. Interviewers want proof you keep asking “why” until you reach something you can actually change.

Sample answer: “Our customer support team kept getting the same password reset tickets, about 300 a month. The quick fix was a better help article, and we had already tried that twice. I took 50 recent tickets and traced each one back. Asking ‘why’ a few times led to the real cause: the reset email was landing in spam for two large email providers because of a sender setting. I brought the evidence to our IT lead, who changed the setting. Reset tickets dropped by about two thirds the next month. The lesson I took was to read the tickets themselves, not just count them.”

Weak answers: Stopping at the first cause (“users forgot their passwords”), or describing a root cause analysis someone else did.

4. “Describe a time you had to solve a problem with incomplete information.”

Why they ask it: Real problems rarely come with full data. They are testing whether you can act sensibly, state your assumptions and adjust as you learn.

Sample answer: “A supplier told us on a Friday that a key part would be late, but could not say by how much. I had to decide whether to delay a customer order worth about $80,000. I listed what I knew and what I didn’t, then called two other suppliers to see who could supply a partial batch. One could cover 40% of the order by Tuesday. I told the customer the situation honestly, offered a split shipment, and set a Monday check-in with the original supplier. The customer accepted the split, the rest arrived ten days later, and we kept the account. I would rather make a reversible decision early than wait for certainty.”

Weak answers: Waiting for all the facts before doing anything, or guessing without saying you were guessing.

5. “Tell me about a time your first solution did not work.”

Why they ask it: Ownership and iteration. They want to see that you notice failure quickly, own it and try again with what you learned.

Sample answer: “I built a new intake form to reduce errors in our onboarding paperwork. After two weeks, errors had not moved at all, and my manager asked what happened. I sat with three new hires as they filled it out and saw the real issue: they had the form, but not the documents it asked for, so they guessed. My form solved the wrong problem. I added a checklist email that went out three days before the start date, listing exactly what to bring. Errors fell by about half over the next quarter. Now I watch someone use a fix before I call it done.”

Weak answers: Blaming the users or the team, or a “failure” that was secretly a success. Our guide to answering “tell me about a time you failed” covers this in depth.

6. “Describe a decision you made based on data.”

Why they ask it: Analytical judgment. They want to hear which data you used, how you checked it, and whether you were honest about its limits.

Sample answer: “My team wanted to add a second weekend shift at our store because Saturdays felt busy. Before agreeing, I pulled three months of sales and foot traffic by hour. Saturdays were busy, but only from 11 a.m. to 3 p.m.; the evening hours were quieter than weekdays. A full extra shift would have cost about $1,500 a month. Instead, I proposed one extra person for a four-hour midday block. We tried it for six weeks, wait times at checkout dropped, and we spent less than half the cost of a full shift. I also told the team the data could change at the holidays, so we agreed to review it in November.”

Weak answers: Naming a tool (“I used Excel”) instead of a decision, or using data only to support a choice you had already made.

Rehearse one data story in a mock round in Interview Practice, then check the transcript for the numbers you actually said.

7. “You have three urgent tasks due today and time for two. What do you do?”

Why they ask it: This is situational. They are testing how you prioritize, and whether you communicate early instead of quietly missing a deadline.

Sample answer: “First, I would check what each task is for: who is waiting, what happens if it slips, and whether any of them can be done partly. Then I would rank them by impact and by how hard the deadline really is. If one task is a client deliverable and another is an internal report, the client usually comes first. I would tell the owner of the third task right away, explain why, and offer a new time or a partial version today. Last quarter this happened to me with a client proposal, a pricing update and a team slide deck. I finished the first two and sent my manager a draft deck with a note, and she moved the meeting a day. Nobody was surprised.”

Weak answers: “I’d just work late and do all three,” which avoids the question, or silently letting one slip.

8. “Tell me about a time you spotted a problem before it happened.”

Why they ask it: Initiative. The best problem solvers prevent problems, and interviewers want proof that you look ahead.

Sample answer: “While preparing for a product launch, I noticed our order form only accepted five-digit ZIP codes, but the launch included a Canadian retail partner. Nobody had flagged it because testing used only US addresses. I tested a few Canadian postal codes, confirmed they failed, and raised it with the developer and the project lead two weeks before launch, with screenshots. The fix took a day. On launch day, Canadian orders went through without problems, and the project lead added international address tests to the launch checklist. Since then, I read every launch plan asking who it leaves out.”

Weak answers: A problem you spotted but did not raise, or one you only noticed after it had already happened.

9. “How would you estimate how many coffees a cafe sells in a day?”

Why they ask it: It is an analytical question with no single right answer. They are testing whether you break a problem into parts, state assumptions and sanity-check your result out loud.

Sample answer: “I’d split the day into time blocks, because demand isn’t even. Say the cafe is open 7 a.m. to 7 p.m. In the morning rush, from 7 to 10, maybe one customer a minute, so about 180. Midday, from 10 to 2, maybe one every two minutes, about 120. The afternoon, from 2 to 7, maybe one every three minutes, about 100. That’s roughly 400 customers. If the average customer buys about 1.2 drinks, that’s around 480 coffees a day. To sanity-check, 480 drinks over 12 hours is 40 an hour, which a two- or three-person counter can handle. I’d refine it by asking about location, since a train station cafe looks very different from a suburban one.”

Weak answers: Freezing because you don’t know the real number, or giving one number with no visible reasoning.

A text mock in Interview Practice is a good place to try one estimate a day and see whether your assumptions are clear.

10. “Tell me about a time you solved a problem with limited resources.”

Why they ask it: Resourcefulness. Most teams have less budget, time or staff than they want, and interviewers want to see you work with what you have.

Sample answer: “Our nonprofit needed to survey about 400 program participants, and we had no budget for a survey tool or extra staff. I set up a free online form, wrote it at a sixth-grade reading level, and asked our three front-desk volunteers to offer it on a tablet at check-in for two weeks. To raise responses, I worked with a local bakery that donated cookies for anyone who finished it. We collected 260 responses, well above the 100 we had hoped for, and the results shaped our next grant application, which was funded.”

Weak answers: Saying the problem could not be solved without more money, or a story where someone else found the resources.

11. “Tell me about a time you disagreed with others on how to solve a problem.”

Why they ask it: Problem solving at work happens with other people. They want to see that you argue with evidence, listen, and commit once a decision is made.

Sample answer: “Our team was fixing slow month-end close, and my manager wanted to hire a contractor. I thought the delay came from late approvals, not workload. Instead of arguing in the meeting, I asked for two days to check. I mapped the last three closes and showed that invoices sat waiting for sign-off for an average of four days. I proposed an approval deadline and an automatic reminder first, with the contractor as a backup. My manager agreed to try it for one month. Close time dropped from nine days to six, and we didn’t need the contractor.”

Weak answers: Going along silently and complaining later, or “winning” the argument without testing your idea.

12. “A customer reports a problem you cannot reproduce. What do you do?”

Why they ask it: This is situational and common in support, IT, product and operations roles. They are testing patience, structured questioning and follow-through.

Sample answer: “I’d start by thanking the customer and asking for specifics: exactly what they did, when, on which device, and what they saw, ideally with a screenshot. Then I’d try to match their setup as closely as I can. If I still can’t reproduce it, I’d check logs or records for their account at that time and look for other reports of the same thing. I’d give the customer a clear next update time, even if I don’t have an answer yet. In my last support job, one ‘random’ checkout error only appeared for customers using a specific browser extension; we found it because two customers sent screenshots that showed the same toolbar.”

Weak answers: Closing the ticket as “cannot reproduce,” or implying the customer is wrong.

13. “Tell me about a time you had to make a quick decision under pressure.”

Why they ask it: Some problems can’t wait for analysis. They want to see that you stay calm, protect what matters most, and review the decision afterward.

Sample answer: “Thirty minutes before a client webinar with 200 registrants, our presenter’s internet failed. I had to choose between delaying, canceling, or presenting myself from her slides. I knew the content because I’d built the deck, so I called her to confirm the key points, told attendees we’d start five minutes late, and presented while she joined by phone for questions. The webinar ran with one short delay, and the feedback scores were in line with our usual sessions. Afterward I set up a rule that every webinar has a backup presenter who has rehearsed the deck.”

Weak answers: A decision that was quick but reckless, or one with no reflection on what you’d change.

14. “How do you decide whether to solve a problem yourself or escalate it?”

Why they ask it: Judgment. Escalating everything wastes your manager’s time; escalating nothing can hide serious risks.

Sample answer: “I ask three questions: Is it within my authority? Can I fix it in a reasonable time? And what happens if I’m wrong? If it’s within my role and low risk, I fix it and tell people afterward. If it touches money, legal risk, safety or a customer relationship, I raise it early, with the facts and a suggested fix. Last year I noticed a vendor had charged us twice for the same shipment. I could have asked them for a refund myself, but the vendor was in a contract renewal, so I flagged it to my manager with the invoices and a draft email. She sent it as part of the renewal talks and we got the credit.”

Weak answers: “I always try to handle it myself” or “I’d go straight to my manager,” with no criteria for which.

15. “Tell me about a process you improved.”

Why they ask it: Improving a process is problem solving without a crisis. They want to see that you notice waste and fix it in a way that lasts.

Sample answer: “New clients at my agency waited about three weeks for their first report because setup was spread across four teams with no single checklist. I interviewed one person from each team, wrote out every step and found two handoffs where work sat for days. I built a shared checklist with an owner and a due date for each step, and set up a short weekly check-in for open setups. Within two months, average setup time fell from about 21 days to 12. The checklist is now part of our standard onboarding, and new team members are trained on it.”

Weak answers: An improvement nobody else adopted, or one with no before-and-after result.

16. “If you joined and found sales had dropped 15% in one region, how would you figure out why?”

Why they ask it: An analytical, on-the-spot business problem. They want a structured breakdown, not a guess, and they often ask follow-ups to test how you would narrow it down.

Sample answer: “I’d start by confirming the number: over what period, compared to what, and whether reporting changed. Then I’d split the drop into parts. Is it fewer customers or smaller orders? Is it across all products or a few? All stores or one? Next I’d compare with outside factors: a competitor opening, a price change, weather or staffing. I’d talk to the regional manager and two or three store leads, because they often know in minutes what data takes days to show. Once I had a likely cause, I’d test a fix in one area and compare it to the rest before rolling it out. I’d give my manager a short update after the first week, even if it’s only what I’ve ruled out.”

Weak answers: Jumping straight to “run a promotion” before diagnosing anything.

To hear how your structure sounds under follow-up questions, run this one in an Interview Practice case-format mock.

How to Practice These

Build a small story bank before the interview: a hard problem you solved, a root cause you found, a fix that failed, a data-based decision, a prioritization call and a process you improved. MIT and Dartmouth’s career offices both suggest preparing 3-5 versatile stories that can cover several themes, and each story above can answer more than one question. Then practice two analytical questions out loud, because thinking aloud is a separate skill from telling a story.

• OphyAI Interview Practice runs voice or text mock interviews with follow-up questions, a transcript and per-answer feedback in behavioral, technical, system design and case formats; the free plan’s 5-credit grant covers one session capped at 15 minutes.
• OphyAI Interview Copilot gives real-time answer suggestions during live Zoom, Microsoft Teams and Google Meet interviews, with a live transcript and Whisper Mode overlay, in 35 languages.

Start practicing →

Frequently Asked Questions

What are the most common problem solving interview questions?

The most common are “Tell me about a time you solved a difficult problem,” “How do you approach a problem you’ve never seen before?”, “Describe a time your first solution didn’t work,” “Tell me about a decision you made with data,” and “How do you prioritize when everything is urgent?” Most are behavioral, so prepare real STAR stories, plus one or two situational and analytical examples.

How do I answer problem solving questions with the STAR method?

Set the situation and your task in two or three sentences, then spend most of the answer on your actions: how you defined the problem, what facts you gathered, which options you considered and why you chose one. Close with a measurable result and what you learned. MIT’s career office suggests about 60% of the answer should be your actions.

What if I have no work experience to draw on?

Use school projects, part-time jobs, volunteering, sports, clubs or family responsibilities. Interviewers care about how you think, not where the problem happened. Pick a situation with a real obstacle, such as a group project behind schedule or a club event that lost its venue, and show the steps you took and the result you got.

How long should a problem solving answer be?

Aim for about one to two minutes for a behavioral story. Analytical questions, such as an estimate or a business problem, can run longer because you are thinking out loud, but check in with the interviewer every minute or so. If they want more detail on your diagnosis or your options, they will ask a follow-up question.

What do interviewers look for in a problem solving answer?

They look for a clear process: you defined the problem, found the real cause instead of a symptom, considered more than one option, made a choice and checked whether it worked. They also listen for honesty about assumptions and mistakes, and for a specific result. A good answer sounds like a method you could use again.

What are red flags in problem solving answers?

Common red flags are jumping straight to a solution without diagnosing the problem, stories where “we” did everything and your role is unclear, blaming others when a fix failed, no measurable result, and answers to analytical questions with a final number but no visible reasoning. Vague claims such as “I’m a natural problem solver” with no example also score poorly.

How does OphyAI help with problem solving interview questions?

Interview Practice runs mock interviews by voice or text and asks follow-up questions, such as how you knew the cause was real, so you find gaps in your stories before the interview. It gives per-answer feedback and a transcript. During live video interviews on Zoom, Teams or Google Meet, Interview Copilot suggests answers in real time with a live transcript.

Related Guides

Problem solving questions show up in behavioral, case and analytical rounds at these employers:

• Amazon interview guide: Leadership Principle questions such as Dive Deep.
• McKinsey interview guide: case interviews plus the Personal Experience Interview.
• Google interview guide: the general cognitive ability interview.
• Deloitte interview guide: case and behavioral rounds.
• Accenture interview guide: case interviews and scenario questions.
• BCG interview guide: structured case interviews.
• Capital One Power Day case interview: business case rounds.

For the wider question set, see our behavioral interview questions and answers, situational interview questions, teamwork interview questions and business analyst interview questions.

Sources and verification notes

Checked November 2026. Each page was opened in this run. Sample answers are illustrative; their numbers are examples, not data.

• NACE: The key attributes employers are looking for on graduates’ resumes: Job Outlook 2025 survey; nearly 90% of employers seek evidence of problem-solving skills, nearly 80% seek strong teamwork skills.
• O*NET OnLine: Complex Problem Solving: US Department of Labor occupation database; definition quoted above; 910 occupations listed for the skill.
• MIT CAPD: The STAR method for behavioral interviews: suggested split of 20% Situation, 10% Task, 60% Action, 10% Result; prepare about 3-5 flexible stories.
• Dartmouth Guarini Graduate Career: Using the STAR method: past experiences used to predict future job performance; identify 3-5 examples that address multiple common interview themes.