> For the complete documentation index, see [llms.txt](https://docs.zigpoll.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.zigpoll.com/guides/how-to-build-a-post-purchase-survey.md).

# How to Build a Post-Purchase Survey Customers Actually Answer

Design a post-purchase survey that gets answered: which moment to ask in, which questions earn their place, and how to keep the response rate high

The thank-you page is the best survey placement in ecommerce. The customer just gave you their money, their attention is still on you, and every answer attaches to a real order with a real value. Post-purchase surveys on the thank-you page typically see response rates in the tens of percent, against low single digits for an on-site pop-up (see [Response Rate Benchmarks](/response-rate-benchmarks.md)).

That makes it tempting to ask everything at once. Don't. The surveys that work treat the thank-you page as a scarce resource and spend it on the two or three questions that will change a decision this quarter.

{% hint style="info" %}
This guide is the strategy. For the mechanical setup, see the [Post Purchase Survey](/tutorials/post-purchase-survey.md) tutorial.
{% endhint %}

***

## There are two moments, not one

A post-purchase survey is really two surveys that happen at different times, and most mistakes come from asking the wrong question in the wrong moment.

| Moment             | When                                             | What the customer can answer honestly                                                                     | Delivery                                           |
| ------------------ | ------------------------------------------------ | --------------------------------------------------------------------------------------------------------- | -------------------------------------------------- |
| **At purchase**    | Thank You page, seconds after checkout           | How they found you, why they bought, what almost stopped them, who it's for, which brands they considered | **Thank You & Order Status**                       |
| **After delivery** | Email or SMS, a few days after the order arrives | Whether the product met expectations, how delivery went, fit, quality, whether they'd recommend you       | **Order Delivered** or **Order Fulfillment** email |

The customer on the thank-you page has not seen the product. Ask "How satisfied are you with your purchase?" there and you measure the checkout, not the product. Ask "How did you hear about us?" in an email a week later and recall has already decayed.

Run both. Keep them separate.

***

## Pick your questions by the decision they feed

Every question on the thank-you page should trace back to a decision someone will make with the answer. The questions below are the ones that reliably do:

| Question                                 | Decision it feeds                                      | Guide                                                                               |
| ---------------------------------------- | ------------------------------------------------------ | ----------------------------------------------------------------------------------- |
| **How did you hear about us?**           | Where to spend the next marketing dollar               | [Marketing Attribution](/guides/how-to-measure-marketing-attribution.md)            |
| **What almost stopped you from buying?** | What to fix on the site                                | [Why Visitors Aren't Buying](/guides/how-to-find-out-why-visitors-arent-buying.md)  |
| **What convinced you to buy today?**     | What to lead with in copy and ads                      | [Why Customers Buy](/guides/how-to-find-out-why-customers-buy.md)                   |
| **Who are you buying for?**              | How to segment email, when to start seasonal campaigns | [Gift Buying](/guides/how-to-measure-gift-buying.md)                                |
| **Which other brands did you consider?** | How to position against competitors                    | [Competitors](/guides/how-to-find-out-which-competitors-you-are-winning-against.md) |

Pick the one or two whose decisions are live right now. Attribution is the usual starting point because it is the one that moves budget; the rest can rotate in as questions come up.

A reasonable first survey is three questions: attribution, one more from the table, and a **Long Answer** catch-all such as "Anything else you'd like us to know?" The catch-all is where bugs, broken promo codes and product requests show up unprompted.

***

## Order the questions by importance, not by logic

Completion falls with every question, and it falls fastest on the first two or three. Two things follow:

* **Put the question you most need answered first.** If attribution is why the survey exists, it goes first, even when a warm-up question would feel more natural. People who drop off after the first question still gave you the thing you needed.
* **Open with a click, not a paragraph.** The first question should be a **Single Choice** or a **Binary choice**. A text box as question one is the single most common reason a post-purchase survey underperforms its benchmark.

Longer surveys are not automatically worse. Zigpoll counts one response per person however many questions they answer, so a six-question survey where most people stop at three still gives you three questions' worth of data on every respondent. What matters is that the questions people reach before they quit are the ones you care about.

***

## Write answer options that produce a decision

Most post-purchase surveys fail at the answer options, not the questions.

* **Make options specific.** "Social media" is not a channel you can fund. Instagram, TikTok and YouTube are.
* **Make options distinct.** Don't list "Instagram post," "Instagram story" and "an influencer on Instagram" as separate choices. Overlapping options split one signal three ways.
* **Keep each list to one kind of thing.** A list that mixes channels ("Google") with motivations ("the discount") can't be charted.
* **Always include "Other."** Enable the write-in. Zigpoll reads the write-ins on attribution questions automatically and files them under a channel, so "insta" and "chat gpt" land where they belong. Elsewhere, Z-GPT Insights clusters them into themes.
* **Include an honest escape.** "I don't remember" on an attribution question, "None of these" on a motivation question. Without one, people guess, and a guess is noise you can't separate from signal.
* **Prune.** After a few hundred responses, retire any option under about 3–5% and promote the write-ins that keep appearing. Review the "Other" pile monthly.

***

## Show different customers different surveys

A first-time buyer and a fifth-time buyer should not see the same survey. The first-timer can tell you how they found you and what nearly stopped them. The repeat buyer already told you that; ask them why they came back instead.

Create one survey per segment and use **Shopify Targeting Rules** (**Settings → Targeting**) with the **New Customers** and **Returning Customers** presets. Zigpoll serves the first live survey whose rules match, top-down in your Surveys list, so put the specific surveys above the catch-all. The [Layered Post Purchase Surveys](/tutorials/layered-post-purchase-surveys.md) tutorial walks through it, and [First-Time Buyers into Repeat Customers](/guides/how-to-turn-first-time-buyers-into-repeat-customers.md) covers what to ask each group.

The same mechanism handles subscription orders (**Contains Subscription**), high-value orders (**Customer Total Spent**), specific products, and shipping country.

***

## Keep the response rate up

* **Add the Zigpoll block high on the Thank You page.** Buyers who have to scroll past the order summary mostly don't.
* **Decide whether you want the Order Status page.** The block can also sit on the Order Status page, which customers revisit from the confirmation email for weeks. That collects more responses, but some arrive long after the purchase. For a pure moment-of-purchase read, keep the block on Thank You only.
* **Offer something at the end.** A [Promo Code question](/tutorials/adding-discount-codes-to-your-survey.md) at the end of the survey lifts completion and sets up the second order. Put it last so it rewards finishing.
* **Send one reminder.** [Nudge Non-Responders](/polls/nudge-non-responders.md) emails people who saw the survey but didn't answer, once, and skips anyone who did.
* **Check the Typical line.** Every survey dashboard shows a benchmark range for its delivery type. Below the range usually means a text question is first, or the survey is too long.

***

## Reading the results

Every post-purchase response carries its order. Zigpoll shows **total order value** and **average order value** for each answer option ([Post Purchase Order Stats](/tutorials/order-stats.md)), so each question can be read two ways: by how many people chose an answer, and by how much revenue those people represent. The two rankings often disagree, and revenue is the one to act on.

Beyond that:

* Segment by **new vs. returning**, by **product**, and by **discount used** before you draw a conclusion. The blended number is usually an average of groups that want opposite things.
* Read the write-ins. Z-GPT Insights will summarize them into themes once you have volume.
* Ask Z-GPT Chat questions in plain language: "What products are most purchased by customers who found us through TikTok?"

***

## Push answers where they'll get used

A survey answer that stays in a dashboard changes nothing. Use [Question Logic](/questions/question-logic.md) actions to send each answer where the decision gets made:

* **Shopify** — **Add Customer Tag** or **Add Order Metafield**, so the answer appears next to the order where your merchandising team already works, and so [Shopify Flow](/shopify-app/shopify-flow.md) can act on it.
* [**Klaviyo**](/integrations/klaviyo.md) and other email platforms — answers become profile properties, which become segments, which become different flows for gift buyers, subscribers, and people who said they wanted a product you don't make yet.
* [**Slack**](/integrations/slack.md) — route "what almost stopped you" answers to the channel where the people who can fix the site will see them.
* [**Google Sheets**](/tutorials/sync-survey-responses-to-google-sheets.md) — for the month-over-month view.

***

## Related

{% content-ref url="/pages/kf9a6Ys5Y7CYvVupiXEi" %}
[What Brands Have Learned with Zigpoll](/guides/what-brands-have-learned.md)
{% endcontent-ref %}

{% content-ref url="/pages/WK89O0wfWTf4yzln5xzV" %}
[Post Purchase Survey](/tutorials/post-purchase-survey.md)
{% endcontent-ref %}

{% content-ref url="/pages/Iwn5ywxmSvccwtT1njra" %}
[How to Measure Marketing Attribution](/guides/how-to-measure-marketing-attribution.md)
{% endcontent-ref %}

{% content-ref url="/pages/6679EuMiJiYVakSdeJgM" %}
[How to Measure the Delivery and Product Experience](/guides/how-to-measure-the-delivery-and-product-experience.md)
{% endcontent-ref %}

{% content-ref url="/pages/bNh5dnZIwu9WipcijsKg" %}
[Layered Post Purchase Surveys](/tutorials/layered-post-purchase-surveys.md)
{% endcontent-ref %}


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.zigpoll.com/guides/how-to-build-a-post-purchase-survey.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.