Walmart Data Analyst Interview Questions
Data analysts at a very large grocery and general retailer work with sales from a huge number of stores, prices that must stay low, and fresh food that loses value every day it sits. Prepare to describe queries in plain words, to design fair store tests, to compare prices across pack sizes, and to explain findings to people who run stores. What to show here: small savings found in data and repeated across many stores. Each question below has what the interviewer is checking, an answer frame and a sample spoken answer to adapt.
How Walmart hires, in its own words, and its questions for every job →
1. Tell me about a time your analysis found a saving that looked tiny for one item or one store but mattered once it was repeated many times.
- Describe the data and what you were looking for.
- Explain the small effect and how you checked it was real.
- Show how you sized it across the whole business.
- Share what was done and the result.
“At my last job I looked at how often stores reordered paper bags. Each store ordered a little more than it used, and the extra was thrown away when designs changed. For one store the waste was almost nothing. I checked several months to be sure it was a pattern, then multiplied it across all stores and showed the total to the supply lead. We changed the reorder rule to follow real use. The result was your result. Since then I always ask what a small number becomes when every store does the same thing.”
Weak answers only chase large, obvious numbers or scale up a small effect without checking it holds in other stores.
2. Describe a time you were asked for an answer from a very large data set faster than you could do the full analysis. How did you choose between a rough answer now and an exact one later?
- Describe the question and the deadline.
- Explain what a quick answer could and could not tell them.
- Show how you labelled the answer and its limits.
- Say what you did afterwards to confirm it.
“A buying team asked me on a Friday whether a weekend promotion on cooking oil had lifted sales the previous time it ran, and they needed it before the day ended to place an order. Running the full comparison across every store would take until Monday. So I took a sample of stores across regions and sizes, compared the promotion week with the weeks around it, and gave them a range with a clear note that it was a sample. They placed the order. On Monday I ran the full analysis, which came out your result, and I sent it to them with the difference explained.”
Weak answers give a rough number as if it were exact, or refuse to answer until the perfect analysis is done.
3. You have daily sales by store and by item. Describe in words the query that finds stores where fresh produce sales fell this week compared with last week while the rest of the store held steady.
- Join sales to the item table to label produce and the rest.
- Sum sales per store, per group and per week.
- Put this week beside last week for each store.
- Keep stores where produce fell and the rest did not.
“First I would join sales to the item table so each row is marked as produce or other. Then I would GROUP BY store, that label and the week, and SUM the sales. To compare weeks, I would place this week and last week side by side, using a self join on store and label or a window function such as LAG. Then I work out the change for each group. The final filter keeps stores where produce dropped past a chosen limit and other sales stayed within a small band. I would also remove stores that were closed for some days.”
Weak answers compare total store sales only, or forget stores with missing days, which look like a fall.
4. Fresh food that does not sell in time is marked down or thrown away. Which numbers would you track to show whether a store is ordering the right amount of fresh food?
- State the goal: enough fresh food with little waste.
- Pick a waste measure and an empty-shelf measure.
- Decide how to compare stores of different sizes fairly.
- Add a check on shopper experience.
“I would track two things together. One is waste: the value of fresh food thrown away or sold at a marked-down price, as a share of fresh food received. The other is lost sales: how often an item had no stock during hours when it normally sells. A store can look great on waste by ordering too little, so neither number is read alone. I would compare each store with similar stores, by size and area. As a check, I would watch whether shoppers who buy fresh food keep coming back each week.”
Weak answers measure only waste, which rewards empty shelves, or compare a small store with a very large one directly.
5. The price of a basic grocery item is lowered in some stores. How would you tell whether it brought in more shoppers, or whether the same shoppers simply bought that item instead of a similar one?
- Choose test stores and similar stores that keep the old price.
- Compare before and after in both groups.
- Look at the whole category and the basket, not one item.
- Check how long the effect lasts.
“I would pick test stores and match each with a similar store that keeps the old price, so weather and season affect both. Then I compare the change in each group before and after. Sales of the item will rise, which proves little. The real questions are whether the whole category grew or similar items fell by the same amount, and whether the number of baskets containing any item from the category went up. I would also look at total visits to those stores. Last, I would keep watching, because shoppers may only have stocked up early.”
Weak answers celebrate higher sales of the one item and never check the similar items beside it on the shelf.
6. The same washing powder is sold in a small box, a large box and a pack of several, each with its own product code. How do you compare its price fairly across stores and over time?
- Link the codes that belong to one product family.
- Turn every size into one common unit.
- Work out price per unit and check odd values.
- Record the rules so others get the same answer.
“I would first build a mapping from each product code to a product family, using brand, name and size fields, and review unclear matches by hand. Next I would convert every size to one unit, such as weight in grams or number of washes, because size is often typed as free text. Then price per unit is simply price divided by that amount. I would look at the highest and lowest values, since they usually show a wrong unit, like a pack counted as one box. I would save the mapping as a shared table, so every report uses it.”
Weak answers compare shelf prices directly or trust the size field without checking how it was typed in.
7. What interests you about analysing grocery and household sales for a retailer built on low prices, and why Walmart?
- Say what you enjoy in retail data.
- Tie it to low prices and the number of stores.
- Link it to your past analysis work.
- Say what you hope to learn.
“I enjoy data that describes ordinary life, and few things are more ordinary than the weekly food shop. At Walmart the same question can be asked across a huge number of stores, so a finding can be tested properly and then used widely. I also like that the goal is clear: keep prices low, which means finding waste and removing it. In my past work I studied your project, and the part I liked most was seeing a store change what it did. I want to learn how fresh food ordering is planned.”
Weak answers speak about big data in general and never mention groceries, prices, stores or shoppers.
8. A regional manager wants to remove a slow-selling grocery item from every store. Your data shows the few shoppers who buy it have some of the largest baskets. What do you do?
- Check the finding is real and not caused by a few shoppers.
- Estimate what is at risk if those shoppers go elsewhere.
- Present both sides with options.
- Suggest a small test before a full decision.
“First I would check my own finding. I would see how many shoppers it covers and whether they buy the item every trip or only sometimes. If it holds, I would go to the manager with both sides: the shelf space the item uses, and the value of the baskets that contain it. I would not claim those shoppers will leave, because I cannot know that. Instead I would suggest removing the item in a small group of stores and watching whether those shoppers keep visiting. Then the decision rests on what happened.”
Weak answers stay silent because the manager is senior, or state as fact that shoppers will leave.
No figures on this page: pay depends on the level, the city and the offer in front of you. When the offer comes, the free salary negotiation script helps with that part.
The questions above are the prep. On the real call, ClapAssist helps with the ones you did not practice.
ClapAssist is an AI interview assistant for Mac and Windows. It listens to the interview on your computer and shows you what to say, in short lines you can read while you talk. Your live interview audio and screen are never stored. Your resume and notes are saved to your account so the app fills them in on any computer. It is hidden from screen share on every plan, including Free; only you can see it.