Relatio — offer page redesign
A test task for Relatio, a relationship coaching app: review the offer page users see after the quiz and redesign it to convert better.

Overview
Relatio sells a personal relationship plan. Users take a quiz in a web funnel and then land on the offer page, where they pick a plan and pay.
Challenge: the brief asked for a short competitor review, a backlog of five hypotheses, the three strongest built into the layout, new Before/After visuals and a redesigned plan chart, all explained in a Loom.
My role: I did the whole task on my own: competitor review, hypotheses, the redesign, AI-generated Before/After visuals and a Loom walkthrough of every decision.
Result: a redesigned offer page with all five hypotheses applied (the brief asked for three), a plan roadmap instead of the abstract chart and three pricing variants ready for an A/B test.
What was the problem?
The original page had the right blocks, but users had to do too much work before they could trust it:



What did I find out?
I reviewed offer pages from four web2app products, MyAddMile, MusesAcademy, Lovon and MuscleMonster, and noted the blocks that push users toward buying.
Key findings

What did I expect to change?
Why this solution?
Keeping it honest
How did I solve it?
Personal Before/After and a plan roadmap
- Before/After pairs a visual with status bars from the quiz; the shift from red to green shows the gap between now and the goal at a glance
- The roadmap replaces the abstract chart with a concrete path: Today, Week 1 communication, Weeks 2-3 trust, Week 4+ reconnecting
- Names from the quiz, Anna and Mike, keep the message consistent from the ad to the quiz to the paywall; I generated the Before/After visuals with AI in the page's style


Offer, pricing and proof
- An applied promo code, Anna_jun26, with 76% off and a timer, so users don't have to look for a discount
- Three pricing variants for an A/B test: price per day with a crossed-out anchor, the full charge with a per-week price and a SAVE badge, and the full price with clear billing
- The rating block shows 4.8 out of 5, 6.4K reviews and 100K+ users instead of an unexplained number




Where it stands
This was a test task, so nothing went live. If the page shipped, I'd A/B test the pricing variants first and watch conversion and refund rates.
Key takeaways
- Quiz data is the cheapest personalization. The user has already told you their name, problem and goal, so the paywall should use them.
- Honest framing matters on a paywall. Goal-style After bars and real review counts build trust without setting up refunds.
- Next. Test the three pricing variants against each other before changing anything else.


