Why isn't my product quiz converting?
Usually the quiz is fine. What happens after it is the problem.
Quick answer
Your product quiz is converting a normal share of quiz takers into buyers. The rest are not lost, nobody followed up with them properly. Most stores send quiz non-buyers into the same generic welcome flow as everyone else, which ignores the answers they just gave. Follow-up built from those answers recovers a meaningful share of them.
What conversion rate should a product quiz get?
There is no single number, and chasing one is a distraction. Interact analysed over 80 million quiz submissions and found around 65% of people who start a quiz finish it, and around 37.6% of those hand over an email address. Both are strong figures. Both still leave most quiz takers walking away without buying.
The more useful question is not what percentage buys. It is what happens to everyone who doesn't.
Open your quiz tool and your Shopify orders side by side. Count the quiz completions last month. Count how many of those people ordered. The gap is the number that matters, and for most stores it is a lot larger than they expect. Those people are not cold traffic. They opted in, answered five questions about their problem, and saw a recommendation. They are the warmest non-buyers you have.
Why do quiz takers finish and then not buy?
Because the quiz established that they have a problem, and the product page asked them to trust your solution to it, and nothing in between closed that gap.
The hesitation is specific to what they told you. Someone whose dog has been anxious for three years does not have the same doubt as someone whose dog started last month. The three-year owner has tried things that failed and expects yours to fail too. The one-month owner is not sure it is serious enough to spend on yet.
They both get the same automated email. It answers neither worry, because it was written before either person took the quiz.
We ran this on SettledTail, a dog separation anxiety brand, against a randomised holdout. Personalised follow-up built from each person's quiz answers recovered 18% of non-buyers who would otherwise have gone quiet.
Why doesn't my Klaviyo quiz flow already fix this?
Because a flow branches, and branching is not the same as responding.
A well-built Klaviyo flow can split quiz takers into segments and send each segment a different email. That is real improvement over one generic message. It is also the ceiling. You are still choosing between a handful of pre-written emails, written before you knew who would receive them.
Two limits show up fast. First, a segment cannot reference the specific combination of answers one person gave, only the bucket they fall into. Second, and larger, a flow cannot reply. When someone writes back to your quiz follow-up asking whether it works for their situation, that is the most qualified message in your inbox that day, and most stores have it going to a no-reply address.
How do I improve my product quiz conversion rate?
Start at the end of the funnel, not the start of it.
Most quiz optimisation advice points at the quiz itself. Cut a question, reorder the answers, rewrite the results page. Those changes move the completion rate a few points at best, and completion was never your problem.
The larger gain sits in the part nobody built. Every quiz taker who did not buy is a person who told you about their problem and then heard nothing relevant back. Fixing that costs no ad spend and no traffic. It works with the quiz you already have.
If you do want to change the quiz, change one thing only. Ask at least one question about how bad the problem is and how long it has been going on. That single answer is what makes a useful follow-up possible later.
Doesn't personalised follow-up stop working at scale?
Only if a person is writing every email by hand.
This is the objection that stops most stores from building it. Three hundred quiz takers a month is three hundred emails, and nobody has time for that. So the generic flow stays, because the generic flow is the only thing that scales.
The answer is that the writing gets automated and the judgement does not. Each email is generated from that person's own quiz answers, then a human approves it before it sends. Once the voice is right, approval takes seconds per email rather than the twenty minutes writing one would take. Replies work the same way. The draft is written for you, you decide whether it goes.
That is the part most stores never get to, because building it is a project and running the store is a full time job. It is also why the recoverable revenue is still sitting there.
How much revenue is actually sitting in your quiz non-buyers?
Work it out on your own numbers instead of a benchmark.
Take your monthly quiz completions, subtract the ones who ordered, and multiply the remainder by your average order value. That is the pool. You are not recovering all of it, and any provider promising you that is guessing. But recovering a modest share of a pool that size usually clears the cost of building the system several times over.
The reason this is worth attention over other conversion work is that the traffic is already paid for. You bought the click, you built the quiz, they completed it. The follow-up is the cheapest step in the entire funnel and the one most stores never build.
See it running on your brand
Drop your email. We'll build a demo on your quiz, with your offer and your customer, and send it over.
NO PITCH DECK. JUST THE DEMO.
Hi Sarah,
You mentioned the hot flushes are waking you at night. Losing the sleep makes everything else louder.
How long has it been like this?
Frequently asked questions
Does the quiz itself need changing?
Usually not. If it captures an email and asks diagnostic questions about the person's problem rather than just filtering products, it already collects what a follow-up needs.
How is quiz follow-up different from an abandoned cart email?
Abandoned cart catches people who reached checkout. Quiz non-buyers never got that far. They are earlier in the decision, far more numerous, and they have told you considerably more about themselves.
How soon after the quiz should the first email go out?
Within the hour, while the person still remembers taking it. The only constraint is time of day. Anything landing in the middle of the night sits under forty other emails by morning, so hold it until the recipient's next waking hours.
Will personalised follow-up cause spam complaints?
Plain-text email from a named person that references what someone told you is among the least spam-like messages in most inboxes. Complaint rates on these are typically lower than on bulk campaigns.
How many quiz completions do you need for this to be worth it?
Below roughly a hundred completions a month the numbers get thin. Not because the approach stops working, but because it takes too long to find out whether it did.
