Scenario · Marketplaces & consumer apps

Building an app like Strava for a running club: what to copy, what to skip, and what it cost

How to build an app like Strava, told through an Amsterdam running club: why not to rebuild GPS tracking first, Strava's API limits, map costs and a cost table.

By Akash Bansal, founding engineerUpdated 5 November 20267 min read

Illustrative scenario. The founder in this story is a composite, not a real client. The process, costs and decisions are drawn from builds I have done; names and business details are invented to show how a build like this typically goes.

People who search "how to build an app like Strava" usually want one specific thing that Strava does, not all of it. The first job is to find out which thing, because rebuilding Strava's core (accurate GPS recording on every phone and watch) is the most expensive part and the least likely to beat the original.

This is a composite story. The founder below is invented, and the build is typical of what this kind of project needs. My evidence for the location side is real but related, not the same: as head of technology at Urban Culture, an at-home salon marketplace with 500K registered users across six cities, I ran location-heavy apps for customers and partners and built a custom geolocation service that replaced Google's Distance Matrix and geocoding, cutting that cost by 80% while holding accuracy. That was routing professionals to homes, not recording runs. Some lessons carry over; I'll say where.

The founder and the starting point

Call him Joris. He runs a running club in Amsterdam with three weekly group runs, a beginners' programme and a few hundred paying members. The club lives on a WhatsApp community, a Strava club page, a Google Sheet for the monthly challenge and a payment link for memberships.

His brief: "an app like Strava, but for our club". Members would record runs in it, see a club leaderboard, sign up for group runs and pay their membership.

Do you need to rebuild Strava?

No, and Joris couldn't win if he tried. Strava reports over 180 million users in more than 185 countries (Strava Year in Sport 2025). His members already record runs on Strava, Garmin, Apple Watch or their phones. Asking them to switch recorders is asking them to give up their history.

The obvious shortcut, pulling everyone's runs from the Strava API, doesn't work for a club app either. Strava's API agreement says data from a user "can only be displayed or disclosed in your Developer Application to that user", and that data about other users may not be shown even if it is public (Strava API agreement). A club leaderboard is exactly that. The terms tightened in November 2024 and caused a stir among third-party apps (Cyclingnews).

So the scoping turned the brief around: don't record runs, collect them. Apple Health and Android's Health Connect are where most watches and running apps can write finished workouts, including the route. With a member's permission, the club app reads those workouts and builds its leaderboards from them.

Who built it, and why

Joris got a quote from an Amsterdam agency for a native iOS and Android app with its own GPS recorder, well above his budget. He looked at club-management tools, which handled memberships and events but not the challenge leaderboard members actually cared about. He chose a founding-engineer build: one senior engineer responsible for design, code, hosting and launch, on a fixed quote after scoping. On my builds the client owns the code, repositories, app store accounts and cloud accounts from day one, sees a working demo every week, and gets written design docs and a handover.

What got cut from v1

The build, week by week

1. Week 1: scope, data model, privacy. What gets read from health data (date, distance, duration, route), what never does (heart rate, other health records), and the privacy defaults.
2. Week 2: accounts and club structure. Sign-in, member profiles, pace groups, roles for coaches.
3. Weeks 3–4: workout import. Apple Health on iOS and Health Connect on Android, with clear permission screens. A single Flutter codebase covers both apps; Flutter vs React Native covers when the other choice is better.
4. Week 5: challenges and leaderboards. Monthly distance, streaks, beginners' programme progress, all visible only inside the club.
5. Week 6: group runs and routes. Weekly sessions with capacity per pace group, waitlists, and coach-drawn routes on a map.
6. Week 7: payments and admin. Memberships, a coach dashboard, attendance.
7. Week 8: store review and launch prep. App Store and Google Play submissions, privacy labels, a beta with the beginners' group.

Ten weeks is a more honest plan than eight if the club wants Android and iOS polished at the same time; store review adds days you don't control.

What the build and running costs look like

The build is a fixed quote after scoping. My MVP builds start at $15,000 over about 6–10 weeks; for this scope, two mobile apps plus an admin, my rough estimate is $20,000–$30,000, mostly depending on how polished both platforms need to be at launch and how many challenge types there are. If your club app has a different mix of platforms and features, the app cost calculator gives a rough range for it.

The map line is where my Urban Culture experience applies directly. Location APIs are cheap at small volume and grow with every user and every request. We learned that the hard way with Google's distance and geocoding calls, and the fix was a custom service built once we knew exactly which calls we needed. For a club app, the lesson is simpler: use static route maps, not per-request routing, and check the pricing tiers before you pick a map provider. For the full story, see how we cut the Maps bill by 80%.

What went wrong

• Duplicate runs. A member with a Garmin and a phone had both write the same run to Apple Health, so it counted twice. The fix was to treat workouts that overlap in time as one, keeping the one from the watch.
• Treadmill runs on the leaderboard. Indoor runs have no route and sometimes generous distances. The club decided they count for streaks but not for the distance challenge. That was a rules decision, not a code one, and it had to be written down first.
• Permission screens scared people. The first version asked for all health data at once. Asking only for workouts and routes, with a sentence explaining why, made more members say yes.
• Home addresses on the map. Imported routes start at members' front doors. In 2018, Strava's public heatmap exposed the layout of military bases around the world (Mail & Guardian). A club app is smaller but the risk is personal: routes now hide the first and last few hundred metres, and are visible only to club members.

When you do need your own GPS recorder

Some products can't avoid it: live race tracking, a running app for people without watches, or a fitness product whose recording is the differentiator. If that's you, budget for it as its own project. On iOS, recording in the background needs "Always" location permission and the location background mode, and the user sees an indicator while you track. On Android, Google Play requires a declaration and review for background location, and lists health and fitness among the accepted uses (Google Play policy). Then comes the actual work: battery drain, GPS noise between tall buildings and under bridges, and testing on many devices.

What launch looked like

Joris launched to the beginners' group first, because they were new to the club and had no habits to break. The monthly challenge moved from the Google Sheet into the app the following month, and group-run sign-ups moved over once coaches trusted the waitlists. The WhatsApp community stayed for chat. The numbers he watched were the share of members who connected their health data and the share who signed up for a run through the app each week, because those tell him whether the app replaced the sheet and the polls.

Checklist if you want to build an app like Strava

1. Write down the one thing Strava doesn't do for your audience. That is your product.
2. Read Strava's API agreement before designing any feature that shows one person's data to another.
3. Import from Apple Health and Health Connect before you build a recorder.
4. Default to private: club-only visibility and hidden route ends.
5. Pick a map provider by its pricing tiers at your expected user count, not only by the free tier.
6. Write the challenge rules (indoor runs, duplicates, late uploads) before the code.
7. Own the app store accounts, the repository and the cloud account yourself.

For more on cutting a v1 down to what matters, see what to cut from an MVP and the 2026 MVP cost breakdown. If you're building a fitness or community app and want one engineer accountable for all of it, here's how I work.