Wabiroidentity and relationship tests built for search and sharing

Context

Personality and relationship tests are a huge, mediocre search category: endless quizzes, forced sign-ups, results with no method. Wabiro takes the subject seriously on the mechanics — short questions, documented scoring, no data collected by default — and on the tone.

Problem

A test only has value if its result circulates. The product therefore has to solve two things at once: a result precise enough to be recognised (“that is exactly you”) and a format simple enough to be shared and compared. And it has to do so without an account, without an email, without storing the answers — because these are personal questions.

What is difficult

Writing short tests that give results people recognise. Fourteen statements for seven flirting styles, sixteen for eight archetypes: precision comes from the wording of the questions more than from their number.

Keeping the privacy promise while allowing sharing and payment. The line is sharp — score and profile on the server, answers never — and it has to stay true with every feature added.

The product as a system

1. Search→
2. Test→
3. Result→
4. Share→
5. Referral

Important product decisions

• Free, no sign-up, complete result. Nothing is asked before the result — no email, no card. An optional magic-link account, with no password, exists only to keep access to a purchased report from another device.
• Deterministic, public scoring, with method pages. Percentiles only appear once two hundred real completions of a test version exist — before that, there is no “you are in the top 10%”.
• Mobile first, built for screenshots: the result card is the unit of sharing, the challenge link the unit of acquisition.
• No diagnosis. The product is a self-reflection and entertainment tool built on documented rules; it never labels anyone toxic or narcissistic, and says so.

Screens

From input to fulfillment — every stage, its layer, what happens there.

Product architecture

The loop: a search intent becomes a two-to-five-minute interactive test; the result becomes a personalised card and a challenge link; the friend who opens it takes the same test and gets a head-to-head with a verdict; they become the second user. The premium report is a one-off, optional purchase.

Tests are content units with a fixed number of questions — twelve to fifty — and an output model: a score out of a hundred, a level, one of eight archetypes, or a profile with strengths and patterns to watch. Two-person tests (compatibility) have each person answer separately and show only where styles converge, never the other’s answers.

Engineering decisions and privacy

Answers stay in the browser’s local storage and the score is computed on the client. If the user creates a share link or buys a report, only the score, the level and the profile name are sent to the server — never the individual answers, never in analytics.

That constraint shapes everything: the scoring engine is a pure, versioned module that runs in the browser, and the same module produces the method pages. A test version is immutable once published, which makes percentiles comparable.

Next.js, TypeScript, Postgres for shared scores and reports, Stripe for one-off purchases, Vercel.

Domain model

Test

Immutable version, questions, output model, public method page generated from the same module.

Result

Score, level, archetype — computed on the client; shared only if the user decides to.

Challenge

Link to the same test, head-to-head with a verdict.

Report

One-off purchase, attached to a passwordless magic-link account.

Current status

In production, payments live. Free tests with no sign-up, premium report at €7.90 and a three-report bundle at €14.90, in English and French; new tests every few weeks.

Business and operating model

Free, no sign-up, complete result; the premium report is a one-off purchase — no subscription, no fake urgency. Growth comes from search and sharing, not advertising; privacy by design — answers never sent to the server — is the constraint that shapes the product.

What shipped

• The loop: search intent → interactive test → personalised result → shareable card and challenge link → second user → premium report
• Deterministic, versioned scoring with public method pages; percentiles only shown once 200 real completions exist for a version
• Mobile-first UX built for screenshots; results and head-to-heads work without an account
• SEO-driven acquisition across test, category and guide pages; one-off premium reports paid through Stripe

Acquisition architecture

Two channels, two mechanics. Search: every test, category and guide is a page that answers a query and leads straight to the test. Sharing: every result carries a challenge link; the friend who opens it takes the test without friction and becomes a user.

The product is bilingual English / French from the start.