Case study · Onedio
A new test and quiz app for Onedio, built from scratch in Flutter for iOS and Android, with personalized feeds and results made to be shared.
Neon Apps designed and built Mygen for Onedio from scratch, an iOS and Android test and quiz app on a single Flutter codebase. Onedio’s own app had already turned tests into a habit, but that experience was the same for everyone. Mygen is a separate product. It starts from what the person tells the app about themselves and ends in a result made for them. We ran the work end to end and delivered it in a four-month project.
Mygen is the test and quiz app Neon Apps built for Onedio in a four-month project in 2025. Onedio’s own app had already made tests a habit, but Mygen is a separate product, built from scratch for iOS and Android. It asks for a name, a photo and a few interests before the first question, so the result at the end is built around the person who took the test, and it arrives with a share sheet and a version already sized for stories. Everything runs on one Flutter codebase, and the tests themselves are data, so Onedio’s editors can publish a new test without waiting for an app update.
Onedio is one of Turkey’s best-known digital content platforms, and describes itself as the country’s largest. Its app has been in the store since 2012, carrying news, video and the tests its readers keep coming back for. That habit is what Mygen was built on. Neon Apps ran the work end to end, scoping it with Onedio’s editorial team, designing it and building it for iOS and Android.
Onedio’s own app had already made tests a habit.
People knew where to find them, but the experience was the same for everyone. The same feed, the same result at the end, and nothing in it that belonged to the person who took the test. The brief was to build a new app around that habit, not to patch the old experience. A feed that adapts to the person using it, a result built from the answers they gave, and a sharing layer designed for a phone screen. They wanted one partner for all of it, from product scoping and design to the Flutter build and the store work. We started with a scoping session with their editorial team.
Onedio’s app had already built the habit of taking tests. What it did not have was a product built around the person taking them.
Every user saw the same feed. Nothing was collected about what someone liked, so the app could not learn what that person would want to take next.
Every test ended in the same result for everyone. The job was to build a result around the person’s own name, photo and answers.
Screenshots were already the habit. The app had to be faster than taking one, and hand people an image that already looks right in a story.
A user does four things. They find a test, take it, get a result and share it. The habit already covered the first two, so we started from the result card and worked backwards, then built the feed so it adapts to the person using it.
Discovery, question, result, share. Each screen is there because the next one depends on it.
Name, photo and interests come first, in three quick steps. Those are the inputs the result card is built from, which is what makes it specific to that person.
Single choice, photo questions, polls and true or false all run on the same engine. A new test became a content job instead of a design job.
The Neon Apps team of 85 people at the Istanbul office.
We sat down with Onedio’s editorial team, listed the test formats they actually publish, and agreed on what the engine had to cover.
Home, onboarding, the question screens and the result were designed as one flow, in the order a user meets them.
One Flutter codebase for both platforms, with tests and questions coming from data rather than being written into the app.
Handover inside the four months, including the content setup Onedio’s team now runs on its own.
Mygen for Onedio, front to back: one Flutter codebase, the test library, the question engine and the sharing layer, delivered in a four-month project.
Ten screens in one continuous flow. Every screen below is from the shipped design, in the order a user sees them.
The library sits in categories for personality, career, character, horoscope and polls, and the feed reorders itself around the interests picked during onboarding. Editorial picks run across the top, including the most-solved tests of the year.
Name, photo and interests take three quick steps. They are what let the feed start relevant and the result call you by name.
Photo questions, polls and true-or-false all end in the same place. The user gets a result card built around their own answers, with their name and photo on it, a share sheet and a version already cropped for stories.
Mygen was not built to run on paid installs. The product was meant to bring its own users, which puts the weight on four things.
Onboarding is three steps, so even the very first test ends in a personalized result. No two results look quite the same, which is what gets a first session finished.
Sharing is one tap and the story version needs no cropping. Users bring the next users, with no media budget behind them.
A large library, clear categories and picked-for-you rows mean there is always one more test to take.
New tests are content, not releases. Onedio publishes as often as it likes and the app is current the same day.
None of this needs a media budget. The result screen does the work. It is personal enough to post, and every post is where the next user comes in.
Flutter let us build iOS and Android from a single codebase and spend the time on the flow instead of writing it twice. The personalization layer is what the whole build turns on. The engine reads each question type and builds a result around the profile of the person answering.
iOS and Android ship together, and a fix lands in both at once.
The feed and the results are built around user signals. Interests from onboarding and previous results feed the engine, so the app gets more accurate the more it is used.
The share sheet and the story export were part of the first build, not something added later.
Neon felt like our own product team, not an agency. They understood what Mygen needed to be, made the hard calls with us, and shipped it end to end.
Kaan Kayabalı
Co-Founder at Mediazone
Neon Apps is a mobile-first studio with 85 people in house, across Istanbul and New York. Product, design and engineering sit in the same office, so a four-month schedule holds.
Neon Apps is a mobile app development company based in Istanbul and New York, building iOS, Android and cross-platform products for startups and enterprises since 2019. The team is 85 people in house, has completed 600+ projects for 300+ clients, and those products have reached 100 million users with a 98% client satisfaction rate. Neon Apps is also an official RevenueCat partner, which is the subscription infrastructure behind much of that app growth work.
Got a project? We build world-class mobile and web apps for startups and global brands.
