60 Supply Chain Analyst interview questions to ask job applicants

Ask a supply chain analyst candidate to walk you through one number they changed: which metric, what it read before, what it read after, and what they did to move it. That single question separates people who have run a forecast from people who have read about one. The 60 questions below are organised the way an interview actually runs, and each block tells you what a strong answer contains, because a question you cannot score is just conversation.

This page is written for the person doing the interviewing. If you are hiring your first or second analyst and there is no recruiter sitting between you and the shortlist, the hard part is not finding questions. It is knowing which answer is good.

TL;DR

• Pick 10 to 12 questions, not 60. Ask every candidate the same ones in the same order, and score each answer on the same scale, which is what makes a comparison mean anything.
• Weight the technical block heaviest. Excel, SQL and SAP are all flagged In-Demand by O*NET for this occupation, and they are the skills most easily faked on a resume.
• A strong answer carries a number, a baseline and a consequence. A weak answer carries a process description with no result attached.
• Pay the market seriously: the median wage for logisticians was 82,320 dollars in May 2025, and employment is projected to grow 18 percent from 2025 to 2035, so good analysts have options.
• Score before you discuss. Reviewers who compare notes before writing scores converge on whoever spoke most confidently.

What does a supply chain analyst do?

A supply chain analyst measures how goods move and money follows, then changes the plan when the numbers say the plan is wrong. In practice that means demand forecasts, inventory targets, supplier performance, freight cost, and the reporting that tells everyone else which of those is on fire this week.

The U.S. Bureau of Labor Statistics tracks the role under logisticians, who held about 255,100 jobs in 2025 at a median annual wage of 82,320 dollars, with about 26,600 openings projected each year over the decade. Most jobs expect a bachelor's degree, though the BLS notes less education can be enough for some of them. Worth knowing before you write the job ad: an 18 percent projected growth rate means your offer is competing, not being received gratefully.

The task list matters more than the title. O*NET's occupational profile for the same SOC code lists 22 core tasks, and two of them are the ones hiring managers forget to interview for: "Redesign the movement of goods to maximize value and minimize costs" and "Explain proposed solutions to customers, management, or other interested parties through written proposals". The job is analysis plus persuasion. A candidate who can do the first and not the second will produce correct work nobody acts on.

On tools, O*NET flags Microsoft Excel, SAP software, Microsoft Office and Microsoft Outlook as In-Demand, and lists SQL, Microsoft Access, SQL Server, Visio and Primavera among its Hot Technologies for the occupation. So when a resume says "proficient in Excel", that is the claim most worth testing and least worth believing.

One scoping note before the questions. A supply chain analyst is not a supply chain manager. The analyst owns the numbers and the recommendation; the manager owns the people and the call. If you are really hiring the latter, the questions for a supply chain manager are a different set, weighted toward leadership rather than modelling.

60 supply chain analyst interview questions

Sixty questions is a bank to draw from, not a script to read. The table shows how the blocks divide and roughly how much weight each one deserves for a mid-level hire. Shift the weighting if the role leans one way: a planning-heavy job pushes technical higher, a job sitting between procurement and sales pushes communication higher.

Experience and method questions

These open the interview. The aim is to find the edge of what the candidate has actually done, quickly and without making them defensive.

1. Walk us through one metric you owned end to end. What was it, what did it read when you took it on, and what did it read a year later?
2. Which part of the supply chain have you spent the most hours inside: demand planning, inventory, procurement, or transport?
3. What does your week look like when nothing is broken?
4. Which report that you built got used the longest, and who used it?
5. Describe a forecast you got badly wrong. What did you miss?
6. How do you find out whether the data you were handed is trustworthy before you build on it?
7. What is the largest data set you have worked with, and in what tool?
8. Which supply chain KPIs do you consider overrated?
9. How did your last employer measure your performance?
10. What did you inherit that was broken, and did you fix it or work around it?
11. Which decision in your last role did you influence that you are least sure was right?
12. What would you want to see in the first two weeks here before you recommend anything?

Behavioral questions worth asking

Behavioral questions earn their place when they force a specific past event rather than a policy statement. If an answer starts with "generally what we would do is", the candidate has dodged it. Ask again for the one time.

1. Tell us about a time you found an error in someone else's analysis. How did you raise it?
2. Describe a cost reduction you can quantify. What was the baseline, and how was the saving verified?
3. Tell us about a supplier that kept missing commitments. What changed as a result of your work?
4. Describe a time you pushed back on a target you thought was unachievable.
5. Tell us about an inventory decision that tied up more cash than it should have.
6. Describe the last time you had to explain a number to someone who did not want to hear it.
7. Tell us about a project you could not finish. What stopped it?
8. Describe a time you automated something manual. How many hours a week did it return?
9. Tell us about a disruption that hit while you were on the job. What was your first move in the first hour?
10. Describe a disagreement with a planner or buyer about what the data meant.
11. Tell us about a time you changed your own recommendation after new data arrived.
12. Describe the hardest handover you have done, to a person or to a system.

Situational questions for incomplete data

Analysts are paid for the days when the data is late, partial, or contradicted by someone on the phone. These questions put the candidate on one of those days.

1. A supplier confirms a shipment will be three weeks late. Finance wants a cash impact by the end of the day. Where do you start?
2. Two systems disagree on on-hand stock for a top-selling item. How do you resolve it before anyone reorders?
3. Demand for one product jumps 40 percent in a fortnight. How do you decide whether it is a trend or noise?
4. Your forecast says cut the order, the sales director says hold it. What do you do?
5. A carrier raises freight rates 12 percent mid-contract. What options would you put in front of the business?
6. You discover a report everyone trusts has been double counting returns for months. What is your sequence?
7. The warehouse says a stock count is accurate. Your numbers say it is not. How do you settle it?
8. You have one week to cut 5 percent of landed cost on a category. What do you look at first?
9. A new product launch has no history. How do you build the first forecast and say how wrong it might be?
10. Half the data you need sits in a system you do not have access to. What now?

Working style and collaboration questions

Weight this block lowest, but do not cut it. An analyst whose work nobody reads is an expensive report generator.

1. Who outside your team read your work most often, and what did they do with it?
2. How do you present a recommendation someone is going to dislike?
3. What do you do when two stakeholders ask for opposite things in the same week?
4. How do you keep a weekly report useful after the first month, when people stop opening it?
5. How do you prefer to receive feedback on your analysis?
6. Describe how you work with procurement, planning and finance differently.
7. When have you had to say "the data cannot answer that"?
8. What do you do when you are asked for a number you know is going to be misused?
9. How do you keep track of the recommendations you have made and whether they landed?
10. What do you want from a manager in your first six months?

Supply chain analytics interview questions to ask

This is the block most interviews get wrong. Teams ask "how strong is your Excel?" and accept "very" as an answer. Ask instead for the mechanics, because the mechanics cannot be bluffed by someone who has only watched a colleague do it.

Sixteen questions, grouped by what they actually probe. Give this block about a third of your total weight for a mid-level analyst hire, and expect to lose some candidates here who interviewed beautifully in the behavioral block.

1. Explain the difference between a safety stock and a reorder point to someone in finance.
2. How do you calculate forecast accuracy, and which measure do you prefer? Why that one?
3. What does a MAPE of 30 percent tell you, and what would you do about it?
4. When would you choose a moving average over exponential smoothing?
5. Walk us through how you would size safety stock for an item with lumpy demand and a 10-week lead time.
6. What is the practical difference between service level and fill rate?
7. How do you work out total landed cost, and which component do people most often leave out?
8. What does an ABC analysis change about how you manage a category?
9. In Excel, how would you join two tables with no common clean key?
10. Which Excel functions do you reach for when a data set is too big for VLOOKUP to stay fast?
11. Write the logic, in words or SQL, for finding every SKU whose on-hand stock exceeds 90 days of cover.
12. In SQL, what is the difference between a left join and an inner join, and when has that distinction bitten you?
13. What ERP have you pulled data from, and what was unreliable about the way it stored dates or units?
14. How do you handle a data set where 5 percent of rows have missing costs?
15. You have 30 seconds and one chart to show inventory is too high. What chart do you build?
16. How would you check whether a supplier's on-time delivery number has actually improved or just been measured differently?

Two of these are deliberately mean. Question 11 asks for logic rather than syntax, so a candidate who thinks in sets can pass it on paper without a keyboard. Question 16 is the one that finds analysts who have been burned before, since a reporting definition change is the most common fake improvement in supplier scorecards.

Pro Tip: Run the technical block as a short working session rather than a verbal quiz. Hand over a messy 200-row extract with duplicate SKUs, blank costs and two date formats, and ask the candidate to tell you the top five overstocked items in 20 minutes, talking as they go. You learn more from watching someone clean data than from any answer about cleaning data. If a live session is not practical, a scored SQL assessment gets you the same signal before the first call.

Supply chain analyst interview questions and answers

An interviewer does not need a model answer to recite. They need to know what a good one contains, so they can recognise it when it arrives in a shape they did not expect. The pattern to listen for is simple: a number, where the number came from, what was done, and what happened next. Answers missing the last part are the most common near-miss.

What to listen for in experience answers

Scope is what you are testing here, and candidates inflate it without meaning to. Someone who "managed inventory across EMEA" may have maintained the report that someone else acted on. The question that settles it is who had to agree before anything changed. An analyst who names the buyer, the planner and the finance partner they had to convince was in the room. An analyst who says "we decided" and cannot say who else was in the we usually was not.

Listen for the shape of the number too. "We cut inventory by 20 percent" invites one follow-up: twenty percent of what, measured when, and against what target. Strong candidates answer that instantly because they lived with the figure for months. Weaker ones reach for a range, then revise it. That is not dishonesty, it is distance from the work, and distance is the thing you are trying to measure.

What to listen for in behavioral answers

You want one event, in order, with the candidate's own hands in it. The common failure is the policy answer: asked about a supplier missing commitments, the candidate explains how supplier management should work. Ask again for the one supplier, and the one week. Candidates who have done the work drop into detail quickly, including detail you did not ask for, like the name of the part or the day the shipment was due.

Mark up the answers that include a cost. Behavioral questions in supply chain are really questions about tradeoffs, and a candidate who says "we expedited by air, which cost about 14,000 dollars and saved a launch date" is showing you they understood what the decision was worth. One that says "we expedited the order" has described a task.

What to listen for in situational answers

The first move is the signal. Strong candidates bound the problem before they solve it: which orders are affected, which customers, how much cash, by when. Weak candidates start by telling you who they would escalate to, which is a reasonable instinct and tells you nothing about analysis. Ask what they would have in front of them before they picked up the phone.

Then listen for whether they say what they do not know. Situational questions have deliberately incomplete information, and the best answers name the missing piece and the assumption they would make in its place, with a plan to check it. A candidate who answers a question with no data gaps acknowledged has either misread the question or does not notice gaps, and the second is a genuine risk in a forecasting role.

What to listen for in technical answers

Correctness matters less than you expect, and reasoning matters more. Plenty of good analysts will fumble the formal definition of a measure they use weekly, then explain exactly when it misleads them. That is the better answer. What should worry you is fluent terminology with no operational consequence attached: a candidate who defines MAPE crisply but cannot say what they would change at 30 percent has learned the vocabulary from a course rather than from a planning cycle.

For the tool questions, listen for friction. Real Excel and SQL work is full of it: inconsistent units, dates stored as text, a join that silently duplicates rows. A candidate who mentions the annoyance before you ask has done the work. One who describes everything as straightforward has not done it at this scale, which is fine for a junior hire and a problem for a senior one.

A worked example of scoring one answer

Take question 1, the metric you owned. Here are three illustrative answers to the same question, scored on the scale further down this page. They are composites written to show the scoring, not real candidates.

Scores 2. "I owned our inventory reporting. Every Monday I pulled the stock position from SAP, built the cover report by category, and circulated it to planning and the warehouse team. I kept it accurate and on time for about two years." This is a clear description of an activity with no result and no decision attached. Accurate reporting is genuinely valuable, but nothing here shows the candidate changed anything.

Scores 4. "I owned days of cover for the spares category. It was running at 96 days when I picked it up, against a target of 60. I rebuilt the reorder points using actual lead-time variability instead of the standard 6-week assumption, which was wrong for two of the three suppliers, and we got to 71 days over about eight months without the stockout rate moving." Baseline, action, result, and a number with a source. The analysis is specific enough that you could check it.

Scores 5. The same answer, plus: "What I would do differently is sequence it the other way. I changed the reorder points first and only cleaned the lead-time data afterwards, so for the first two months the new points were built on the same bad assumption and planning lost some confidence in the numbers. Now I would fix the input data before touching any parameter." The candidate has told you something that does not flatter them, which is the hardest evidence to fake and the best predictor you will get in a 45-minute conversation.

Notice what separates the 4 from the 5. It is not more detail about the achievement. It is one piece of self-criticism with a decision attached, and interviewers routinely score it lower than the polished version because it sounds less confident. That is the bias the written anchors exist to catch.

Three things are worth marking down hard, whatever else the answer contains. A candidate who cannot state the baseline for any improvement they claim has probably not owned it. A candidate who has never been wrong in a way they can describe has either not been trusted with a decision or is managing you. And a candidate who talks only in tools, never in money, will struggle the first time a CFO asks what the recommendation is worth.

One honest caveat on the whole exercise. Interview answers measure how well somebody describes work, which is correlated with doing the work but is not the same thing, and the gap is widest for candidates who are nervous, interviewing in a second language, or coming from a smaller team where they did everything and named none of it. That is the argument for pairing the interview with a work sample rather than scoring harder.

How should you structure the interview?

Ask every candidate the same questions in the same order, and score every answer against the same scale. That is the whole method. The U.S. Office of Personnel Management describes a structured interview as one where all candidates are asked the same predetermined questions in the same order and all responses are evaluated using the same rating scale and standards for acceptable answers.

It sounds bureaucratic. It is the difference between a hiring decision you can defend and a preference you can only describe. Three rounds is usually enough for this role:

1. Screen, 25 minutes. Four questions from the experience block plus two from working style. You are checking scope and communication, not depth.
2. Technical, 60 minutes. Six to eight questions from the analytics block, ideally with the messy data exercise. This round should be run by whoever will review the candidate's work.
3. Team and judgment, 45 minutes. Three situational questions and two behavioral ones, with the stakeholders the analyst will have to persuade in the room.

Keep it to 10 or 12 scored questions in total across the rounds. More than that and interviewers stop writing scores and start forming impressions, which is the thing the structure exists to prevent. If your process runs longer than three rounds for an analyst hire, the extra rounds are usually buying reassurance rather than information, and good candidates drop out while you are buying it.

One more rule that costs nothing: have each interviewer write their scores before the debrief, not during it. Reviewers who talk first converge on whoever was most articulate, and the quietest competent candidate in your pipeline loses every time.

How do you score answers without guessing?

Use a five-point scale with written anchors, and define the anchors before you meet anyone. The scale below is a starting point you can lift as is. What matters is that every interviewer reads the same words for what a 3 means.

Two constraints are worth knowing before you build the scorecard, because they apply to interviews and not only to tests. Interview questions are a selection procedure. The U.S. Equal Employment Opportunity Commission's guidance on employment tests and selection procedures states that the test or selection procedure must be job-related and its results appropriate for the employer's purpose, and that where a procedure screens out a protected group, the employer should determine whether there is an equally effective alternative with less adverse impact. That is a direct argument for cutting the charming-but-irrelevant question from your list.

The second is the four-fifths rule, set out in 29 CFR 1607.4, which treats a selection rate for any race, sex or ethnic group below four-fifths (or 80 percent) of the rate for the highest group as evidence of adverse impact. Read it for what it is: a practical enforcement rule of thumb the agencies use, not a definition of what is lawful, and not a measure of whether your questions predict performance. If your analyst pipeline is small, the ratio will swing on a single decision, which is a reason to watch the pattern over several hires rather than to calculate it once and relax.

This is the part Testlify is built around: assessment scores, reviewer ratings, interview notes and AI summaries land in one structured view, and the decision stays with the hiring team. Testlify's own product copy states it plainly: AI scores and insights are for guidance only, and human judgment is for final decisions. Score aggregation makes disagreement visible, which is the point. It does not resolve it for you, and a platform that claimed otherwise would be selling you a worse process.

Which questions fit adjacent supply chain roles?

The same searches that bring people here often turn out to be for a neighbouring role. The questions transfer, but the weighting does not. Here is what changes.

Logistics analyst interview questions

Weight freight and carrier data over demand planning. Useful additions: how do you build a landed cost model across three modes; what does on-time in-full actually measure at your last employer, and who defined it; how would you decide between air freight and a stockout; which carrier scorecard metric is most often gamed; and how do you model the cost of a routing change that saves two days.

Inventory analyst interview questions

Push hard on inventory mathematics and cycle counting. Add: how do you set cycle count frequency by ABC class; what is your approach to ageing and obsolescence provisions; how do you tell slow-moving from dead stock; what causes most of your inventory record inaccuracy; and how would you reduce stock by 15 percent without raising stockouts.

Supply technician interview questions

This role is about accuracy and process discipline rather than modelling, so drop the SQL questions entirely. Ask: walk us through your receipt and put-away process; what do you do when a delivery does not match the paperwork; how do you keep a stores area countable; how do you handle an urgent issue request that breaks procedure; and what is the last process error you found and reported.

If the role you are filling is really a coordinator, the supply chain coordinator question set is weighted for scheduling and supplier communication instead.

When should you use a skills assessment?

Before the first interview, not after. The reason is arithmetic rather than preference: the technical block carries the most weight and takes the most interviewer time, so running it on candidates who cannot clean a spreadsheet is the single most expensive habit in an analyst hiring process. Score the skill first, then spend the hour on judgment and fit.

For this role, three assessments do most of the work. A supply chain management assessment covers the domain knowledge: inventory logic, planning concepts, procurement basics. A spreadsheet or SQL test covers the tooling O*NET flags as In-Demand for the occupation. And a numerical reasoning or cognitive test tells you whether someone will cope when the problem is new, which matters more for a first analyst hire than for a fifth. Testlify's test library carries role-based and custom assessments you can combine into one invite, with weighted scoring so the technical section can count for more than the rest.

Where it genuinely does not help: a senior hire coming in to build a planning function, where the thing you are buying is judgment and stakeholder credibility that a 30-minute test will not see. Test them anyway if you like, but weight the interview and the references heavily, and do not let a mid-range test score talk you out of a strong operator.

If the job ad still needs writing, the supply chain analyst job description template and the wider logistics and supply chain hiring guide cover the requirements and the market context that shape which of these questions matter most.

Hire supply chain analysts on evidence

Pick 10 questions from the blocks above, write your five anchors before you meet anyone, and send a scored assessment ahead of the first call so the interview is spent on judgment instead of verification. If you want to see how the scoring and the reviewer workflow fit together on real candidates, book a 30-minute demo and bring the role you are hiring for.

Frequently asked questions (FAQs)