Post-Purchase Survey Template (Checkout, Delivery, Two Weeks)
A post-purchase survey template split into three short sends: four questions on the confirmation screen, four when the parcel lands, six two weeks after that. Sellers also call it a post-sales questionnaire, an after-sales survey, a post-sale survey or a post-transaction survey. Read the post purchase survey questions below, answer them here to see how each send runs, copy any block, or open all fourteen in the editor.
- 14questions
- 3short sends
- About 4 minfor all three
- One orderstart to finish
Three sends: what the checkout, the parcel and the product can each answer
One customer, one order, three measurements. The reason to split them is not tidiness: a question about the product cannot be answered by somebody who is still looking at an order confirmation.
The split is not our idea. The national index the United States uses to track customer satisfaction writes the stage onto its own questions, labelling one set pre-purchase and another post-purchase, because what somebody expected and what they got are two measurements taken at two different times.[1] The same report keeps product quality and service quality apart, repeating its questions once for each rather than averaging them into one number.[1] These three sends are that separation, moved onto a calendar.
Send 2 is fired by the parcel, not by a date. The largest seller-rating programme works the same way. Google Customer Reviews emails the buyer a few days after the estimated delivery date, the merchant has to pass that date in the opt-in module for the survey to be scheduled at all, and the survey is one question long.[2] A trigger you can point at beats a delay you guessed.
Two weeks is when the decision is live. A shopper in the United Kingdom or the European Union has fourteen days to withdraw from a distance contract, counted from the day the goods reach their hands rather than the day the money moved.[4] That fortnight is exactly when somebody is deciding whether to keep the thing. A United States buyer has no such clock, so there the timing is a design choice, and a fortnight sits well inside the three-month window the national index uses for an online order.[1]
Waiting costs something too. Professional survey design treats the gap between an event and the question about it as a known source of error, in detail lost and in events being placed in the wrong week.[5] That is the reason each send fires on its own event instead of being batched into one email later, and the reason the third send asks about a decision rather than about a moment.
| The send | What fires it | What it can answer | What it cannot |
|---|---|---|---|
| 1. The confirmation screen | The order is placed. | How the buying went: how hard it was, where they came from, whether they are new. | Anything about the parcel or the product. Neither exists yet. |
| 2. On delivery | The carrier marks the order delivered. | Whether it arrived when promised, complete, and in one piece. | Whether the thing is any good. Most boxes are not open yet. |
| 3. Two weeks after delivery | Fourteen days after that delivery event. | The product against what they expected, and whether they are keeping it. | How the checkout felt. That memory is a fortnight old. |
The one mistake this template exists to stop. Asking about the product before it has arrived. A question list with no send attached to any item invites exactly that: copy it straight into an email and the product question lands on the confirmation screen, where the only honest answer a buyer can give is a shrug.
None of the three runs past six questions. In a web experiment, a longer stated questionnaire got fewer starts and fewer finishes, and its later answers came back shorter and more alike.[6] Three short surveys are what stops the fourteenth question being the worst-answered one.
Two neighbouring triggers are not this survey. When what fired the survey is a help request rather than an order, the post-service survey asks about the fix and the effort it took. When you want the whole relationship rather than one transaction, the customer experience survey questions run across the stages a customer passed through instead of the three moments of one order.
All fourteen questions are written and every scale is set, so you can make an online survey out of any one send, change the wording to match how you talk to buyers, and have it live before the next order goes out.
The 14 post purchase survey questions, one block per send
Fourteen questions in three blocks. Each item says what it is for and how to read a bad answer to it. Copy one send, or copy all fourteen.
Send 1: on the confirmation screen (4 questions, about a minute)
Fired by the order. Everything here is answerable with the product still in a warehouse, which is the test every item in this block has to pass.
- How easy was it to place this order?Effort at the checkout, not effort in getting help. Read it as the share who chose Very easy; a drop belongs with the payment screen, not the catalogue.
- Is this your first order with us?One tap that makes the other thirteen readable by segment. A first order and a fifth are not the same experience, and nothing else here separates them.
- How did you first hear about us?Asked at the order because that is the last moment the buyer is still thinking about how they got here. Treat it as a rough read rather than a channel report: it is one self-reported answer.
- Anything about ordering we should fix?Last and optional, so skipping it costs nothing. Read these before the effort score: one named field explains more than an average does.
Send 2: when the parcel lands (4 questions, under a minute)
Fired by the carrier delivery event rather than by a date. Everything here is about the parcel, and nothing here is about whether the thing inside it is any good.
- Did your order arrive when you expected it?The promise, checked. A United States seller who cannot ship inside the window it advertised owes the buyer the choice to cancel and be refunded,[3] so a late count is a number somebody is already meant to hold.
- Was your order complete?Kept apart from condition on purpose. Missing and damaged are two failures with two different owners, and one item asking both produces an answer you cannot act on.
- What condition did it arrive in?Packaging, handling and the last mile in one item. Anything below Good belongs next to the open answer under it.
- Anything about the delivery we should fix?Where the specific complaint lives, while the box is still on the floor.
Send 3: two weeks after delivery (6 questions, about two minutes)
Fired fourteen days after the delivery event. The first item goes first so you can see who is answering from use and who is not before you read anything else.
- Have you used it yet?First, because it changes what the next four answers mean. A Not yet is still worth having: it is a product still in its box a fortnight after it arrived.
- How satisfied are you with the product itself?The product, not the shop and not the parcel. Split it on the used-it-yet answer above before you average anything.
- How well does it match what you expected before you bought it?The gap between the listing and the object, which is the one thing that cannot be asked before delivery. Three named options rather than a scale, so it cannot be read as a satisfaction score.
- Are you keeping it?A real outcome rather than a downbeat extra: retailers surveyed by the National Retail Federation and Happy Returns estimated that 16.9 per cent of their 2024 sales would be returned.[7]
- Would you buy this product again?Product-level, and deliberately not a scale. Whether they would use the shop again, and whether they would recommend you, are two other questions that live on two other pages.
- What would have made this better?Better than asking for any comments, because it asks for one change. Tag each answer to the item it is about, then count the tags.
Nothing here asks for a name or an email. You already know which order each answer belongs to, so the survey does not need to ask, and the person who set it up is the person who reads what comes back. Tell buyers that in the invitation and you have told them everything they need.
Three jobs this survey hands over. When the parcel is the subject in its own right and somebody will act by carrier or by route, the delivery survey questions go well past four items. When you want one number that stays comparable across a checkout, a ticket and a delivery, the CSAT survey reuses one scale at each. The recommend question, with its arithmetic and its banding, belongs to the net promoter score survey.
Three sends do not add up to one number: what to do with what comes back
These fourteen questions produce no single score, and they are not meant to. That is the trade: three answers instead of one number, each one about something the other two cannot reach.
| What you have | How to read it | What it is not |
|---|---|---|
| Send 1 on its own | The share who chose Very easy, against your own previous month. One question, one line. | Not a satisfaction score. Nobody has been asked how they feel about anything yet. |
| Send 2 on its own | Three counts: late, incomplete, damaged. Report them as counts until you have a few hundred orders behind them. | Not an operations scorecard. There is no carrier, route or driver in these four questions. |
| Send 3 on its own | Split every answer on the used-it-yet item first. People who have used the thing and people who have not are two different groups. | Not comparable with send 1. Different questions, different options, a fortnight apart. |
| One order across all three | The pattern is the finding. An easy checkout, a late parcel and a kept product is a different story from all three going well. | Not an index. Averaging a five-option effort item with a three-option return question gives a number that means nothing. |
| The open answers | Read them before any chart, and keep the send they came from attached to each one. | Not a vote. Six people naming the same field is one fix, not the biggest one. |
If what you actually want is one comparable number, this is the wrong template, and the block above says where to find the right one. One satisfaction scale reused at the checkout, the ticket and the delivery keeps a score comparable across all three; three different surveys deliberately give that up in exchange for questions each moment can answer.
On response rates, we will not quote you one. There is no measurement of our own to give, and the figures on offer around this search come without a source behind them. What has been measured is that a longer stated questionnaire gets fewer starts and fewer finishes,[6] and what is obvious is that somebody who does not have the thing yet has nothing to say about it. Both of those are inside your control. A benchmark is not.
When what you want is the reasons in a buyer own words rather than any of these counts, the customer feedback form is open boxes and nothing else, and it is not tied to one order.
Post-purchase survey FAQ
When is the best time to send a post-purchase survey?
Three times, because there are three different things to measure. Ask about the buying on the confirmation screen while the order is in front of the buyer; ask about the parcel when the carrier marks it delivered; ask about the product two weeks after that. There is no single correct delay: the window is set by what was bought. The national satisfaction index screens buyers of an online order on the last three months and buyers of a new washing machine on the last three years.
How many questions should a post-purchase survey have?
Four, four and six here, and no send runs past six. Length is one of the few things about a survey that has been measured: in a web experiment, a longer stated questionnaire got fewer starts and fewer finishes, and its later answers came back shorter and more alike. If you only have room for one send, make it the second: whether the parcel turned up as promised, complete and undamaged is the question a buyer can always answer.
Is a post-sales questionnaire the same thing?
Yes. Post-sales questionnaire, after-sales survey, post-sale survey and post-transaction survey all name one thing: a survey fired by somebody buying from you. This page says post-purchase survey throughout because it is the clearest of the set, and because it names the trigger rather than the department. The questions are the same whichever word you use for them.
What is post-purchase evaluation?
It is the stage of buying where the customer compares what turned up with what they expected before they paid. That comparison is the entire reason the third send exists, and it is the match-what-you-expected item in the third send, in plain English. A post-purchase evaluation survey and a post-purchase survey are the same instrument under two names, so use the fourteen questions above for either.
How is a post-purchase survey different from a satisfaction score or a recommend score?
By what it produces. A CSAT survey is one satisfaction scale reused at several moments so the number stays comparable. An NPS survey is one recommend question with its own arithmetic and banding. This is three surveys fired by three events, and it deliberately produces no single score. Use it when you want three answers; use one of the others when somebody has asked you for a number.
What is the difference between a post-purchase survey and a product review?
Who it is written for. A review is public and aimed at the next shopper; this is private and aimed at you, so it can ask about a missing item or a late parcel without that sitting on the product page. Both wait for delivery, for the same reason: neither can be answered by somebody who does not have the thing yet.
These questions follow an order that went through. The mirror image, an account that closed rather than opened, belongs in the lost customer survey, which asks what the alternative was and whether anything would have changed the outcome.
An order arrives and the transaction closes. A construction project has a defined end with paperwork attached, and the construction customer survey asks at that end.
Methods and sources
What this template is based on
Five sources carry the timing argument: the national satisfaction index methodology, which labels its own items pre-purchase and post-purchase, measures product quality and service quality separately, and sets recall windows by category;[1] Google Customer Reviews, whose seller survey is scheduled off the estimated delivery date;[2] the United States shipping rule that turns a missed delivery date into a cancellable order;[3] the fourteen-day withdrawal period that runs from physical possession of the goods in the United Kingdom and the European Union;[4] and survey methodology on what elapsed time does to recall.[5] The question count follows a web experiment on questionnaire length.[6]
Question licensing
There is no named or licensed post-purchase instrument, so all fourteen questions here are ours to copy, edit and translate freely. Nothing reproduces the national index item wording: that model is proprietary and its score is a weighted index built from a national sample, so this page describes what it separates rather than reprinting its questions, and no number this template produces is comparable with a published index score.
Limits and disclosure
None of these sources was measured on this template. The index methodology is a 2008 federal filing describing a national measurement programme, not a rule for one shop.[1] The fourteen-day withdrawal window is United Kingdom and European Union law with no United States equivalent, so for a US seller the two-week send is a design choice rather than a legal one.[4] The recall evidence comes from crime victimisation interviewing: the mechanism transfers and the magnitudes do not.[5] The questionnaire-length finding is one web experiment and only its abstract is public.[6] The returns figure is a trade body reporting retailers own estimates, not an official statistic.[7]
SuperSurvey makes the editor this template opens in. Everything on this page is free to use with it or without it.
References
- American Customer Satisfaction Index. ACSI Methodology Report, June 2008. Filed as a federal information-collection document. reginfo.gov
- Google. "Google Customer Reviews FAQ" and "Customer reviews opt-in and survey", Merchant Center Help. support.google.com
- Federal Trade Commission. Business Guide to the FTC's Mail, Internet, or Telephone Order Merchandise Rule. ftc.gov
- The Consumer Contracts (Information, Cancellation and Additional Charges) Regulations 2013, regulation 30; and Directive 2011/83/EU, Article 9. legislation.gov.uk
- Bureau of Justice Statistics. Survey Methodology for Criminal Victimization in the United States. bjs.ojp.gov
- Galesic, M. and Bosnjak, M. "Effects of Questionnaire Length on Participation and Indicators of Response Quality in a Web Survey." Public Opinion Quarterly 73, no. 2 (2009): 349-360. academic.oup.com
- National Retail Federation and Happy Returns. "2024 Consumer Returns in the Retail Industry", press release, December 2024. nrf.com
Put the first send live before the next order
Fourteen questions in three blocks, about four minutes across all of them. Open them in the editor, cut what does not apply to what you sell, and start with the send that fires on delivery.
Use this template