Your product is live and the traffic is there, but the numbers do not follow. I locate the drop-offs with data and user research, rebuild the steps that lose people, and test the change before it goes to everyone.
“Why aren’t they converting? Where do we lose them? What do we fix first?”
Guesswork is expensive. Analytics show where the funnel leaks, research shows why it leaks there, and a tested change shows whether the fix actually works — before it ships to your whole audience.
Three phases, each ending with something the next one needs: evidence, then a design, then proof.
Measure the funnel and then diagnose it. Analytics show which step loses people; session recordings and interviews show why it happens there and not somewhere else.
Turn the findings into ranked, testable hypotheses and rebuild the weak steps around them — clickable, and detailed enough to put in front of people.
Run the new version against the old one, keep what wins, and feed what you learned into the next round instead of a report nobody opens.
The first stage is finding where the money leaks. Funnel analytics, session data and user interviews together show which steps lose people and why — so we work on the step that actually costs you, not the one that is easiest to change.
Then we rebuild the weak steps. Every screen is tied to one conversion hypothesis, drafted quickly and checked against the evidence from the first stage. You get a clickable design you can put in front of people, not a deck.
A change only becomes a result once it is measured. We run the new version against the old one, watch the metric that matters and keep what wins. What we learn feeds the next round instead of disappearing into a report.
Evidence first, then design, then proof. You end up with a product that converts better and a team that can say why.
Sprint or ongoing retainer
Conversion work pays off when a product is already live and people are already arriving. Three situations bring most clients here.
Ads and content bring people in, but sign-ups or orders stay flat. Every additional euro spent on acquisition makes the leak more expensive.
You get a ranked list of where and why people drop out.
Each step belongs to someone else, the analytics go unread, and the roadmap is decided by whoever argues best in the room.
You get evidence that settles the argument.
The campaign works, the landing page does not. Optimising spend hides the problem for one more quarter instead of fixing it.
You get a tested change against the metric you report.
Almost never for the reason the team assumes first. Conversion is lost at a specific step, and analytics can show you which one within days: where people arrive, where they stop, and what they did just before. The reason behind it needs a second method — session recordings and a handful of user interviews. Without that pairing, you are left with guesswork in a confident voice.
The audit stage is typically two to three weeks and ends with a ranked list of problems. Redesigning the weak steps takes another three to five weeks, depending on how many there are. A/B testing then needs enough traffic to reach a reliable result, which depends on your volume rather than my calendar.
For a clean statistical test, yes — roughly a few thousand visitors per variant through the step being tested. Below that we still change what the evidence points at, but we validate it with moderated usability sessions instead of a split test, and watch the metric before and after.
Conversion optimisation asks why people stop. UX design decides what they should meet instead. Kept apart, the first produces button-colour experiments and the second produces redesigns nobody measures. Here they are one programme: the evidence sets the agenda, the design answers it, and the test decides.
Yes, and that is the usual case. I work with your analytics people, your developers and your product owner, and hand over the hypotheses, the designs and the test setup so the loop keeps running after I leave. A team that can do this without me is the point, not a side effect.
A free 30-minute call. No pitch, no pressure — just an honest look at your product, your users and whether I can help.