Post purchase survey questions get answered when there are few of them, they arrive at the right moment for what they are asking, and the customer can tell the answer will actually go somewhere. A five-question survey sent the day after checkout, before the product has even arrived, gets ignored or answered carelessly no matter how well the questions are worded.
What makes a question actually get answered
The biggest lever is not question wording, it is count. Response rates drop sharply once a survey goes past three questions, and the drop is steepest between question one and question two if the first question does not feel worth the time. Most useful post-purchase surveys should ask one focused question, occasionally two, and rarely three.
The second lever is timing matched to the question's subject. A question about delivery experience only makes sense once the package has arrived. A question about whether the product met expectations only makes sense after the customer has had a real chance to use it, which varies by category: a phone case can be judged within days, while a mattress or a supplement needs weeks before an honest answer is even possible.
The third lever is specificity. "How was your experience?" is easy to skip because it is vague and the customer cannot tell what changes based on their answer. "Did the size run as expected?" or "Was setup clear from the included instructions?" gives the customer a concrete thing to react to and gives the brand an answer it can actually act on.
Which post purchase survey questions to ask, and when
| Trigger point | Example question | What it's for |
|---|---|---|
| Delivery | "Did your order arrive as expected?" | Catches shipping damage, delays, or wrong-item issues early enough to fix. |
| Unboxing/setup | "Was everything you needed included and easy to set up?" | Surfaces packaging or instruction gaps before they turn into support tickets or returns. |
| Fit or sizing (apparel, footwear) | "Did this fit true to size?" | Feeds sizing guidance back into the product page to reduce future returns. |
| Product satisfaction (post-use window) | "Has the product done what you expected so far?" | A genuine read on product performance, timed after real use rather than right at delivery. |
| Repeat intent | "Would you buy this again?" | A simple forward-looking signal that correlates with actual reorder behavior more than a satisfaction score alone. |
Step-by-step way to run a post-purchase survey program
Pick one trigger to start, not five. Delivery-arrival surveys are usually the easiest first step since timing is unambiguous (the tracking system tells you when it landed) and the question is low-effort to answer.
Write the single question in plain, specific language, and test it on a coworker who has never seen the product before sending it to customers. If they cannot tell exactly what is being asked, the question needs to be narrower.
Decide the response mechanism before writing the send: a one-tap scale, a yes/no with an optional comment field, or a short open text box. Tap-based responses get far higher completion than open text, so save open text for a secondary optional field, not the primary ask.
Route the answers somewhere they change something. A survey whose answers land in a spreadsheet nobody reviews is worse than not asking, since it trains customers that responding does not matter. Set a weekly or biweekly review cadence with whoever owns the product, fulfillment, or content changes those answers might inform.
Space out triggers across the post-purchase relationship rather than stacking them. A customer who gets a delivery survey, a setup survey, and a satisfaction survey within the same ten days will start ignoring all three. Sequence them naturally: delivery near arrival, satisfaction weeks later once real use has happened, repeat intent near a natural reorder window if the category has one.
Keep any review request separate from private feedback surveys. The FTC's endorsement guides address expectations around soliciting and disclosing reviews, and several marketplaces have their own stricter rules about connecting incentives to review content. A private satisfaction survey is not the same instrument as a public review request, and treating a survey answer as a review without separate, explicit consent creates both a trust issue and, in some marketplace contexts, a policy issue.
Common mistakes
The most common mistake is sending one long survey trying to cover delivery, setup, satisfaction, and repeat intent all at once. Each of those belongs at a different point in time, and cramming them together produces a survey nobody finishes.
The second is sending the satisfaction question too early, often right at delivery, before the customer has actually used the product enough to have an honest opinion. An early satisfaction question either gets skipped or gets a shallow, non-predictive answer.
The third is survey fatigue from over-triggering. Sending a survey after every single touchpoint (delivery, support ticket, and every order) trains customers to ignore brand messages generally, which hurts response rates on the surveys that actually matter.
The fourth is collecting answers with no visible follow-through. If a sizing survey never changes the product page's fit guidance, and a setup survey never improves the included instructions, customers eventually notice the pattern and stop bothering to answer. Treat the answers as an input to real product and support decisions, not only a metric to report internally.