Skip to content
Product teardown · not shipped workStrava · 6 min ·

Strava sells the comparison that stings

Challenges are free and compare you to a number. Segment leaderboards compare you to named people on the road you actually ride, and those, plus every filter that makes the comparison bearable, are the paid half.

What this is built on

Public sources walked 19 August 2026: the iOS App Store listing, the marketing site and subscribe flow in a desktop browser, and the help centre. Re-checked 21 August 2026 after a first draft got the Challenges distinction wrong. No signed-in session, no account, no mobile app.

App Store listing
Read 19 Aug 2026. 4.8 from 368K ratings,
strava.com/subscribe
Read 19 Aug 2026 from India. 30-day trial, then ₹208.25/month billed annually.
Strava Help Centre, "Strava Subscription Features"
Read 19 Aug 2026. Strava’s own enumeration of what the subscription unlocks.

What I could reach

  • The App Store listing, its screenshots and its category ranking
  • The public marketing site and the subscribe flow up to the payment step
  • Strava’s own help-centre article listing every subscription feature

What I could not

  • The signed-in onboarding, the first-run experience and the empty feed. Reaching those requires creating an account, which I did not do.
  • Anything about how the free tier feels in use. Every claim below is bounded to what Strava’s own help centre states about which features the subscription gates.

What this product is trying to get you to do

Record an activity, publish it to people who know you, and come back tomorrow because they are watching.

Everything else, the maps, the training analysis, the equipment tracking, sits underneath that one sentence. Strava’s own App Store subtitle is “Track & share with friends”, and the second word is doing the work: a tracker you keep to yourself is a utility, and utilities get replaced by whatever is already on the phone.

Walking what I could reach

I read the App Store listing, the marketing site, the subscribe flow up to the payment step, and Strava’s help-centre article enumerating subscription features. I did not create an account, so I have not seen the signed-in onboarding, the empty first-run feed, or what a new user is shown on day one. What follows is built only on what is publicly readable.

Three things are visible from outside, and all three matter.

The store listing sells community, not measurement. The screenshot carousel leads with “Track your workouts”, then “Get motivated by the community” over a feed of activities carrying kudos and comments. The product is positioned as social before it is positioned as analytical.

The subscribe page names the paywall explicitly. It offers a 30-day trial and then ₹208.25 per month billed annually, and describes what unlocking gets you: routes, segment leaderboards, advanced training analysis and more.

The help centre is the honest version of that list, and it is longer than the pitch. Behind the subscription: overall segment leaderboards, filtered leaderboards, segment effort comparison, Live Segments, Group Challenges, custom goals, Relative Effort, Fitness & Freshness, the training log, route creation, suggested routes and personal heatmaps.

And a distinction that a first draft of this piece got wrong. Strava runs two different things called challenges. Group Challenges, private, invite-based, up to 199 people you follow, are subscriber-only, and are the ones named in the subscription pitch. Strava’s public Challenges are a separate product with their own gallery, are not listed among subscription features, and athletes are automatically entered into segment challenges by riding the segment. So the free tier is not without a time-boxed goal. It matters, and it is checkable in two clicks on the same help centre I was already citing.

Three friction points, and what each costs the business

1. The free tier’s motivation does not renew

A monthly distance challenge creates one obligation per month, and it is an obligation to a number you already agreed to. A rank you are slipping down creates one every time somebody else rides. Kudos, the other free social mechanic, is a reason to post rather than a reason to return: the reward arrives after the activity you had already decided to do.

What it costs the business: the free tier has to convert people into subscribers before they have felt the mechanic the subscription is actually for. That is a harder sale than it needs to be, and it leaves the 30-day trial carrying the entire job of demonstrating value.

2. The unfiltered leaderboard is the demoralising one, and the filters are the paid part

A segment leaderboard ranks against everyone who has ever ridden or run that stretch. The help centre lists filtering by followers, club, age and weight as subscription features, which means the harshest possible view of a comparison is the one on the free side of the line, if it is visible at all.

What it costs the business: the people the comparison lands worst on are the largest group and the most likely to lapse. I shipped this exact mechanic without the escape hatch and wrote up what it cost. The short version is that a cohort you cannot choose is a cohort you can only lose in. Strava already built the fix. It sells it.

3. Value is gated by feature, not by depth

The split is binary. You have segment leaderboards or you do not. There is no version where a free athlete gets three segments, or a leaderboard restricted to people they follow, or last month’s Fitness & Freshness.

What it costs the business: every free athlete is someone who has never experienced the loop, which makes upgrading an act of faith rather than the renewal of a habit.

One proposal per friction point, and what each would cost

Give the free tier one challenge at a time

Group Challenges move to free, capped at one active challenge. Not the leaderboards. The challenge, which is time-boxed, has a finish line, and does not require being fast to be satisfying.

The trade-off, and who objects. Challenges are named in the subscription pitch, so this removes a line from the paid list, and someone owning subscription revenue will say so immediately and be right to. The counter is that a challenge is a demonstration of the loop rather than the loop itself, and a capped one leaves the ceiling intact.

Make the followers-filtered leaderboard the free default

Invert the current split: the leaderboard restricted to people you follow becomes free, and the global leaderboard plus the age, weight and time filters stay paid.

The trade-off, and who objects. This is the smallest change and the one I would ship first, because it costs Strava almost nothing: a leaderboard among eight people you know is not a substitute for the global one that serious athletes pay for. Growth will like it. The objection will come from whoever owns the paid feature list, because “segment leaderboards” stops being a clean bullet and becomes a bullet with an asterisk.

Meter by depth, not by feature

Three saved segments free. One challenge free. Fitness & Freshness visible for the trailing 30 days rather than not at all.

The trade-off, and who objects. This is the most expensive of the three to build and the hardest to communicate. A feature list with limits in it is harder to put on a pricing page than a feature list with checkmarks. Marketing objects, correctly. I would put this third for exactly that reason.

What I would need to see before shipping any of it

None of the above is a recommendation yet. It is a hypothesis about a product I have read about from outside, and I would want four things before spending a sprint on it.

GuardrailsPlanned · not measured
Week-4 return rate, free vs trial
The whole premise. If free users already return at the same rate as trialists, the paywall is not the retention problem and everything above is a solution to nothing.
What would send me back to diagnosis: A gap smaller than a few points would send me back to diagnosis rather than to a proposal.
Trial-to-paid conversion, split by whether a challenge was completed
Whether the challenge actually drives the subscription, or whether people who would convert anyway are also the people who finish challenges.
What would send me back to diagnosis: No difference between the groups means giving the challenge away costs revenue and buys nothing.
Lapse rate against first-week leaderboard position
Friction point 2, directly. If people who land in the bottom quartile of their first segment lapse faster, the default view is the problem and the filter is the fix.
What would send me back to diagnosis: A flat curve means I have projected my own product experience onto a different user base.
Subscription revenue per cohort, not in aggregate
The obvious failure: giving away a paid feature and reading the resulting engagement lift as a win while the revenue line quietly moves the other way.
What would send me back to diagnosis: Any cohort-level decline that persists past one renewal cycle.

Not measured. I have no access to Strava’s data. This is the instrumentation I would want before treating any of the three proposals as a decision rather than an opinion.

I would also want a sample size that can detect a small effect, because the effect here is small by construction: one capped challenge is not going to double anything, and a test underpowered to see a two-point move will read as “no difference” and kill a real improvement.

What Strava does well that I would steal

The activity is the post. Most consumer health products ask you to do the thing and then separately tell them about it. Strava collapsed those into one action, which is why its feed has content in it without anyone having to be a content creator. Every product asking a user to log something manually should look at that and ask what it could capture instead.

Kudos costs nothing to give and is worth something to receive. It is a lower-effort like than a like, and it is the single cheapest social mechanic I have seen in a health product.

The paid tier is aimed at people who already identify as athletes, and the pricing follows the identity rather than the usage. ₹208.25 a month is not priced for a casual walker and is not trying to be. Knowing precisely who you are not for is worth more than most positioning exercises produce.

Where this stops

I have read this product from outside. I have not seen its onboarding, its retention numbers, its cohort structure, or the reasoning behind a split that the people who built it have certainly already argued about at more length than this piece.

What I would ask in the first meeting is not “why is the leaderboard paid”. It is: what happened the last time you moved something across that line, and what did it do to the renewal cohort. That answer is worth more than everything above it.

Next

Manual logging is the retention ceiling for Indian health apps

Work is what I shipped and can be held to. This page was not.