The "How Did You Hear About Us?" Survey: A Complete Playbook

A "How did you hear about us?" survey asks customers to describe how they discovered your business. On Shopify, you can ask after checkout on the Thank you page and when customers revisit their Order status page. A short question can reveal remembered recommendations, creator mentions or offline discovery that a referral report does not explain.

The useful result is customer context. Survey responses depend on recall, wording and who chooses to respond, so use them alongside analytics and campaign records when deciding what to investigate next.

Install OrderSurvey or book a 20-minute setup call to get help with the question and both Shopify placements.

Product details reviewed October 25, 2026.

Decide whether you are measuring discovery or the purchase trigger

These questions can produce different, equally useful answers:

What you want to learn	Question
Where a new customer first encountered the brand	How did you first hear about us?
What prompted today's visit or order	What brought you to our store today?
Why a previous customer returned	What brought you back today?

A buyer may first discover you through a friend, follow you on Instagram, and return after an email. A first-discovery survey and an analytics report that credits the final visit can therefore disagree without either being a complete account of the purchase.

Use one clear question for the decision you want to make. Avoid changing its wording each week if you want to compare the answers over time.

Start with a short choice list and room for unexpected answers

For a first-discovery question, an example list is Google search, Instagram, TikTok, a friend or family member, a creator or podcast, ChatGPT or another AI assistant, Other, and I don't remember.

Adapt that list to your store. Include relevant sources you have not paid for, and review Other write-ins for gaps. Keep uncertainty separate from a missing option. Our 15 answer options and copy-ready example explains how to make the choices easier to distinguish.

A single-select question asks the customer to choose one primary source. If your question allows several sources, make that clear and remember that the resulting percentages may add to more than 100%. Neither format removes recall or selection bias.

Use branching for a useful second question

Conditional follow-ups help identify the source behind a broad category:

• A creator or podcast: Which creator or show?
• Other: Where did you hear about us?
• An event or store: Which event or location?

Create the main question and the follow-up in OrderSurvey, then connect the relevant answer in the visual branching builder. Preview both a route that shows the follow-up and one that skips it. Keep the follow-up optional unless the detail is necessary to interpret the response.

Asking for a creator's name can tell you what respondents remember. It cannot establish the creator's incremental sales impact on its own.

Launch it on the Thank you and Order status pages

1. Create the survey and choose its questions, answers and branching rules.
2. Select First orders only if you want new-customer discovery feedback. All orders and Repeat orders only serve different audiences.
3. Select the desired pages, check any schedule or targeting conditions, enable the survey and save.
4. In Shopify's checkout and accounts editor, add and save OrderSurvey's block on Thank you.
5. Switch to Order status, add the block there if selected, and save that placement separately.
6. Use a matching approved test order to verify display, branching and successful submission. Confirm the response appears in OrderSurvey.

Installing the app or enabling the survey does not place the blocks automatically. Shopify describes this in its checkout app instructions. Thank you and Order status UI extension blocks are supported on Basic and higher plans.

First-order targeting requires Shopify order context. A survey limited to first orders stays hidden if the context cannot determine whether the order qualifies. Use a qualifying order for that test; an editor preview alone does not prove the complete flow works.

The full setup walkthrough includes placement checks. Book setup help if you would like us to walk through them with you.

Read the answers as evidence, not a channel-ROI calculation

For each reporting period, record the question, its answer options, the audience, and the number of completed responses. Compare periods with the same definitions, and label any changes.

An illustrative example: 8 of 40 respondents choose Podcast. Report 8 responses, or 20% of respondents. Do not present this as 20% of all buyers or proof that podcasts caused those purchases. Nonrespondents may have a different mix of discovery sources.

Useful next steps could include checking which shows people named, reviewing campaign dates, and repeating the question over a longer period. A discrepancy with an ad dashboard is a reason to investigate different attribution windows and touchpoints. It does not prove that the dashboard is over-crediting a channel or that a campaign should be cut.

If you compare answers with order revenue, define the matching orders and revenue measure explicitly, and count each order once. Revenue associated with a reported discovery source is still an association. Measuring incremental return needs additional evidence about costs and what would have happened without the campaign.

Use Unlimited when your reporting workflow needs it

OrderSurvey's Free plan includes up to 100 responses per rolling 30 days, conditional branching and audience targeting. It is a practical way to launch and check the question.

Unlimited is $24 USD/month and includes unlimited responses, CSV export, webhook alerts and priority support. A team that needs regular spreadsheet reporting may benefit from export even before reaching the Free response limit.

The response export contains question-and-answer rows and order identifiers. Filter to the main discovery question before tallying channel choices; follow-up rows are not additional buyers. Order revenue is not included in that response CSV, so any revenue comparison needs a separate, correctly matched order dataset.

Compare the plans, install OrderSurvey, or book a 20-minute setup call. We can help you choose one useful question, check both blocks and verify the first submitted response.