"How did you hear about us?" — the one survey question that pays for itself

If you could ask your customers exactly one question, it should be this: "How did you hear about us?" It's the single most valuable thing a Shopify store can learn — the one answer with direct budget consequences — and almost nobody asks it well. This guide covers why it matters, where to place it, how to write the options, and how to turn the answers into smarter spending instead of a spreadsheet nobody opens.

Why this is the highest-ROI question you can ask

Every other survey question tells you something nice to know. Attribution tells you where to put your next dollar. When a buyer says "a friend told me" or "I heard you on a podcast," you've just learned which channels are quietly driving revenue — the ones no dashboard is crediting. Multiply that across a few hundred orders and you have a real, customer-sourced read on what's working, updated continuously and for free.

The attribution gap analytics can't close

Your analytics stack is good at what it can see: clicks it can tag, sessions it can stitch together, last touches within a tracking window. But a growing share of how people discover brands happens where pixels can't follow — a friend's recommendation, a creator's video watched in-app, a podcast ad heard in the car, an insert in someone else's package. After the iOS privacy changes, even the trackable slice leaks. The result is a systematic under-counting of exactly the channels that build brands.

The point isn't to replace your analytics — it's to fill the gap. Ask the customer, and you recover the discovery channels UTMs and last-click models can't. One brand running post-purchase surveys found that a striking share of its buyers first discovered it through paid social that its own reporting had badly under-credited — the kind of finding that changes a media plan.

Where to ask: thank-you page vs email

Placement is the difference between rich data and crickets. The thank-you (order-confirmation) page is the highest-attention moment you will ever get from a customer — they just paid you, and they're waiting for confirmation. Ask there and response rates are dramatically higher than a follow-up email that competes with a full inbox.

Email surveys still have a place for questions that need distance (like "how's it working out a month in?"), but for attribution you want the answer while the discovery is fresh — the second the order is placed.

How to write the answer options

The options are the survey. Get them right and analysis is trivial; get them wrong and you drown in "Other." A few rules that consistently work:

• List your real channels, not generic ones. "TikTok," "a podcast," "a friend or family member," "an influencer I follow," "saw it in another order's packaging" — specific beats vague.
• Always include "Other" with a text box. It catches the channels you didn't predict, and those write-ins are often where the next big idea hides.
• Randomize the order. Position bias is real — the first option wins votes just for being first. Shuffle so your data reflects reality.
• Keep it to one screen. Five to eight options people can scan. If you need more nuance, ask a quick follow-up instead of a 15-option wall.

Anatomy of a survey that converts

Here's what a high-performing attribution survey looks like on the thank-you page — one clean question, scannable options, an "Other" escape hatch, and a gift at the end that turns the survey itself into a second order.

Turn the answer into action

Data you can't act on is just trivia. The whole point of asking is to change what you do next, so send every answer into the tools where decisions actually happen:

• Klaviyo — each answer becomes an event and a typed profile property, so "heard about us on TikTok" can build a lookalike-seed segment or trigger a tailored welcome flow the same day. (See the Klaviyo setup →)
• GA4 — pipe stated attribution alongside click data to sanity-check where your click models over- or under-credit.
• Shopify tags & Flow — write the answer to the customer record so every downstream app can use it, and trigger automations off it.
• Your media plan — the simplest and most valuable: shift budget toward the channels customers actually name, and stop over-crediting the last click.

Five mistakes that tank your data

1. Asking too late. A survey emailed three days later measures memory, not discovery. Ask at the moment of purchase.
2. Too many questions. Completion drops with every extra tap. Lead with attribution; make everything else optional.
3. Fixed option order. Without randomization, you're measuring position bias as much as truth.
4. No "Other" field. You'll never discover the channel you didn't list — and that's often the interesting one.
5. Letting answers die in a dashboard. If the data doesn't reach Klaviyo, GA4, or your ad decisions, you asked for nothing.

Do the opposite of each and you'll have, within a week, a customer-sourced attribution feed that makes every ad dollar a little smarter — for the price of one well-placed question.