Recommendations people believe
Five different signals have to combine into one ranked list that feels right to the member, not like a random shuffle.
What goes wrong: matches that ignore stated preferences, and members stop opening the app.
Joymatch is a dating app that recommends matches on who people are, not only on how they look: their skills, profession, interests, background and where they live. Appther built the iOS and Android apps, an encrypted backend and an admin platform, and took it live in six weeks.
Joymatch is a dating app for iOS and Android that recommends compatible people using five profile signals: skills, profession, interests, ethnicity and location. Members choose which of those matter to them, the app filters and scores profiles against those choices, and the best fits appear first. Sensitive details are encrypted and an admin platform lets the client moderate profiles and run the service.
Most dating apps rank people on photos and distance, then leave members to work out compatibility in the chat. Johnathan Peffer wanted Joymatch to do that work up front: recommend the people a member is most likely to get on with, based on what they do, what they care about, their background and where they live.
That puts sensitive personal data at the centre of the product. Profession, ethnicity and location say a lot about a person, so the brief asked for an encrypted, secure platform from the first release, not as a later hardening phase. It also asked for both mobile platforms and a way to run the service, inside six weeks.
Each of these decides whether members trust the recommendations, and whether they trust the app with their data.
Five different signals have to combine into one ranked list that feels right to the member, not like a random shuffle.
What goes wrong: matches that ignore stated preferences, and members stop opening the app.
Ethnicity, profession and location need protecting in storage, in transit and on screen, with members in control of what they share.
What goes wrong: an exact location or a private detail visible to the wrong person.
Fake profiles and abusive members drive genuine people away, so reporting, blocking and review have to exist on day one.
What goes wrong: one bad experience in the first week becomes a one-star review.
Two mobile apps, a backend and an admin platform, designed, tested and approved by both app stores inside six weeks.
What goes wrong: scope creeps, store review slips, and the launch date moves.
Every recommendation is built from five signals on a member's profile. Some are firm preferences that decide who is shown at all; the rest decide the order.
Four screens carry the core experience on both iOS and Android.
Photos, skills, profession and interests in a short guided setup, with sensitive fields clearly optional.
Distance, and which signals matter most, so recommendations reflect what the member is looking for.
Ranked recommendations with the shared interests and approximate distance shown on each profile.
When two members like each other, the app confirms the match and notifies both of them.
A dating app holds some of the most personal data a person shares online. Joymatch protects it in storage, in transit and on screen, and gives members control over what others see.
Every request between the apps, the admin platform and the backend travels over TLS.
The database and photo storage are encrypted with managed keys, and sensitive profile fields get an extra layer.
Other members see a rounded distance, never coordinates, so nobody can be pinpointed from their profile.
Photos sit in private storage and load through short-lived links, so a copied URL stops working.
Sensitive fields are optional and can be hidden, and members can block, report or delete their account at any time.
Admin staff see only what their role needs, and every admin action is written to an audit log.
Appther holds ISO/IEC 27001 certification for information security management. Read more about how we approach secure mobile app development.
A browser-based admin platform shares the same backend as the apps, so the client's team can manage members and keep the community safe on their own.
A current, widely supported stack chosen for speed of delivery: one mobile codebase for both stores, and a backend that handles location queries natively.
Why Flutter: a single codebase let one team ship identical iOS and Android apps inside the six weeks. See our Flutter app development and Node.js development services.
A fixed scope agreed in week one is what made the date hold. Anything outside it went on a list for the next release.
Matching rules, data model and screen designs agreed and signed off.
Apps, backend, matching service and admin platform built in parallel, with a test build every week.
End-to-end journeys on real devices, matching checks and a security review of data handling.
App Store and Google Play submission, approval and go-live, with monitoring in place.
“Appther exceeded every expectation, on time, on budget, and the final product has been instrumental in scaling our business. A rare engineering partner.”
Joymatch is a dating app for iOS and Android that recommends matches using five profile signals: skills, profession, interests, ethnicity and location. Appther built the two mobile apps, the backend and an admin platform for a client in Pennsylvania, USA, and the app is live.
Six weeks from kickoff to a live release, covering design, the iOS and Android apps from one Flutter codebase, the backend, the admin platform, testing and store submission.
In two steps. Hard preferences such as distance and any ethnicity preference the member has chosen filter the pool first. The remaining profiles are then scored on shared interests and on how well skills and profession fit, and the highest-scoring profiles are shown first.
Joymatch encrypts data in transit with TLS and at rest with managed keys, stores photos in private storage behind expiring links, shows other members an approximate distance rather than a location, and limits admin access by role with every action logged.
A focused dating MVP with matching, profiles and an admin panel typically falls between $6,000 and $20,000, and a full platform with messaging, video and subscriptions between $20,000 and $110,000. The exact figure depends on features and platforms, and Appther gives a fixed quote after a short discovery call.
Yes. Share your audience, the matching criteria you have in mind and your target launch date through the form on this page, and Appther will reply within one business day with a scope, a timeline and a fixed price.
Tell us who your app is for and how you want people matched. We will come back within one business day with a scope, a timeline and a fixed price.