Post-Purchase Surveys
What surveys catch, where they lie, how to weight them.
We added a "how did you hear about us?" box to the thank-you page mostly as a joke. Within a month, 14% of answers said "podcast". We did not advertise on any podcast. A founder we had sent free product to had mentioned us on one, twice, and Google brand search had been quietly taking the credit for all of it.
The one-sentence version
A post-purchase survey asks the customer what tracking cannot see: the touch that never involved a click. It is the only leg of attribution that catches word of mouth, podcasts, and the reel someone watched twice and never tapped.
The setup
Same orders, two stories
One hundred orders. Tracking assigns each to its last click. Then each customer answers one question. The two pictures of the same week barely resemble each other.
Try it: the same 100 orders
What tracking saw vs what customers said
Meta
22
41
18
Direct / none
19
Google looks huge because brand search is the last click. Meta looks small because the reel was never clicked.
Tracking is precise and blind. Surveys are fuzzy and honest. Neither is the answer alone.
Why it matters
Surveys catch the demand creators
The channels that create demand (reels, TikTok, podcasts, creators) are the ones tracking under-credits, because they rarely get the last click. The channels that harvest demand (brand search, retargeting, email) are the ones tracking over-credits. Surveys pull the credit back toward the beginning of the journey.
Ask one question
Multiple choice, randomised order, an 'other' box. On the thank-you page, not in an email three days later.
Keep the answer on the order
An answer is only useful if it stays attached to that order's tracked journey, so the two can be compared.
Never budget on raw answers
'Instagram' at 47% does not mean Instagram deserves 47% of spend. It means tracking's 22% was too low. The model decides where between.
The podcast founder got a bigger box of free product. Brand search got a budget cap. Both were the right call, and neither was visible in Ads Manager.
The TrueROAS version
One of three legs
TrueROAS ships the survey, stores the answer on the Shopify order next to its server-side journey, and lets the per-store model weigh them against each other. Where tracking and survey agree, credit is confident. Where they disagree, that disagreement is the most useful signal in your account.
Honest footnote
When the survey is enough on its own
Testing a brand-new channel with a few hundred dollars: the survey alone will tell you if anyone noticed. Set up proper tracking before you scale, not after.
Questions people actually ask
Straight answers
What response rate should I expect?
20 to 40% with a single question on the thank-you page, lower by email. Enough for channel-level signal within a few weeks at normal order volume.
Can the survey tell me which campaign?
No. Customers say 'Instagram', not 'Advantage+ Shopping campaign 3, creative B'. The survey sets the channel weight; tracking supplies the campaign and creative.
Do people lie?
Rarely on purpose. They misremember and they simplify. 'Google' often means 'I Googled you after seeing your ad'. That is exactly why survey answers should be cross-checked, not used raw.
Which tools do this?
Standalone survey apps (Fairing, KnoCommerce, Zigpoll) collect answers well but stop there. TrueROAS includes the survey and uses each answer inside the attribution of that specific order.
See your own numbers, verified against your orders.
Free under $10k/month in revenue. 14-day trial above that. Nothing changes in your ad accounts. You will know within a week which platform has been telling stories.
"I built this because I was making $50k-a-month decisions on numbers I knew were wrong. If TrueROAS does not show you something Meta or Google got wrong in the first two weeks, tell me and I will help you cancel."
Rasmus Arvidsson
Founder, TrueROAS. Answers support himself.
Keep going