How to market a fitness or health app on TikTok and Instagram
By Thomas, founder of UserClimbUpdated 16 min read
Show the progress in the numbers, never in the body. Talk to the viewer's situation, not a demographic. Lead with myths and form fixes, why you fall off and the one number to watch, and keep most posts about the viewer's problem rather than your app. Bring the app in as the way they get the result, usually late in the video or in the caption, and ask for the download in only about 20% of posts.
Who you're actually talking to
Someone who has started and stopped more than once. They lift, run or want to, they half-track their food, and they know roughly what to do but not how to keep doing it. They follow gym and running accounts for permission to start again, and they are tired of being sold a body.
What they want
- A simple way to see I'm actually getting stronger or faster
- To stop guessing what to eat and how much
- To not have to think about my workout, just be told what to do
- To stay consistent without the Monday restart
- To know why I'm tired and what to do about it
- Logging that takes seconds, not minutes
- Coaching without paying a coach
What frustrates them before they find an app like yours
- Logging is tedious, so they stop after a fortnight
- Apps shame them with red numbers when they go over
- Progress charts and basic analytics sit behind a paywall
- Programmes assume they already know what they are doing
- Two or three apps that each see one piece of the picture
- Advice online contradicts itself and most of it is selling something
- A sleep score tells them they slept badly, then offers nothing to do about it
- They get hurt, stop, and lose the habit
Words they actually use
Use their words in your hooks and captions, not yours: "I kept forgetting to log", "I've bounced off every tracker", "I don't want another subscription", "I've never been this consistent with any other app", "I don't have to think about my workout", "red numbers", "basic analytics behind a paywall", "stuck at the same numbers on my lifts", "a fraction of the cost of a real coach", "sleep debt", "groggy", "quitting feels like giving up", "how I'm cutting for the summer", "make lifting feel fun again", "I thought I needed eight to nine hours".
What makes them stop scrolling
The one rule for fitness and health apps: show the progress in the numbers, never in the body.
They stop for
- A number they recognise from their own life (stuck at 100 kg on bench, 2,000 steps on a desk day)
- Being told the thing they do every session is wrong
- A lift, a route or a plate they can picture themselves doing
- A chart that moves in the first two seconds
- Gym humour that proves the account actually trains
- Permission: it's not you, it's the plan
- A protocol with numbers and a timeframe
They scroll past
- Abs, scales and before-and-after bodies; reads as an ad or as diet culture
- 'Today I'll show you the best exercise for' openers
- Motivational quotes over gym footage
- A talking head explaining features
- Anything that smells of a supplement or a coaching funnel
- Perfect kitchens, perfect bodies, perfect weeks
- Generic tips with no exercise, food or number named
What to post: the content mix
Split your posts roughly like this: myths and form fixes (25%), why you fall off (20%), the one number to watch (20%), inside the log (15%), gym and run culture (12%), built by someone who trains (8%). Each kind of post has a different job, so you're not asking for the download in every video.
Myths and form fixes (about 25% of posts)
The reach engine for this niche. Takes one thing the viewer does every session and shows a cleaner way, or kills a belief they picked up from a bigger account. Worth watching if the app never existed; the app earns a half-second as the place the founder spotted the pattern.
- Hook: Stop doing [exercise] like this if you want your [muscle] to grow
- Hook: Most people's [lift] is stuck for one reason, and it isn't strength
- Hook: [Common belief] is why your [pace / bench / sleep] hasn't moved in a year
- Hook: You don't need 10,000 steps. Here's the number the research actually supports
Example: "Stop doing [exercise] like this if you want your [muscle] to grow". Camera. 0 to 2s: founder says the line to camera, hard cut. 2 to 8s: 'what most people do' in a red-outlined card while the founder mimes the mistake. 8 to 20s: the fix, one cue, said twice, punch-in on the cue word. 20 to 30s: the test for next session (a rep count, a weight, a feeling). 30 to 35s: 'I keep the cue in my exercise notes so I don't forget it' with half a second of the app's note field. Cold stop. It ends on: save this for your next [muscle] day and send it to the mate who does it wrong.
Why you fall off (about 20% of posts)
The pain nearly every viewer shares: they start, they stop, they restart on a Monday. This pillar makes falling off normal and gives one small mechanic for staying in. It builds the audience that later converts, because a log is a consistency tool before it is anything else.
- Hook: I quit [tracking food / the gym / running] four times. Here's what finally stuck
- Hook: Week 3 is where everyone falls off. Here's why, and what to do that week
- Hook: You don't have a motivation problem, you have a Monday problem
- Hook: The low-energy day is the only day that matters
Example: "Week 3 is where everyone falls off". Motion. 0 to 2s: a calendar with 14 ticks then blanks, narrator: 'this is the shape of almost every new routine.' 2 to 10s: why week three: novelty gone, results not visible yet, one busy week. 10 to 22s: the rule: shrink the session, don't skip it (one set, a 10-minute run, one meal logged), a card for each. 22 to 30s: the app's calendar with the small sessions still logged, narrator: 'a short session logged keeps the streak honest.' End card. It ends on: send this to whoever started with you in January.
The one number to watch (about 20% of posts)
Teaches the viewer to read their own data: weekly volume, expenditure, trend weight, sleep debt, easy pace, protein per kilo. The app's charts appear here naturally as the worked example, and the founder earns trust as someone who understands the data rather than the feature.
- Hook: Your scale lies to you daily. This line doesn't
- Hook: Stop tracking calories eaten. Track this instead
- Hook: Weekly volume is the strength number most people never look at
- Hook: Your 'easy' runs are too fast. Here's how to find your real easy pace
Example: "Your scale lies to you daily. This line doesn't". Motion. 0 to 2s: a jagged daily weight line draws across the screen, narrator: 'a normal fortnight for one person.' 2 to 10s: why it jumps (water, salt, glycogen, a big meal), each cause a small label on a spike. 10 to 20s: a smooth trend line draws through it, narrator: 'this is the number that tells the truth.' 20 to 28s: rule card: weigh daily, read weekly, never react to one day. 28 to 33s: the app's trend screen as the example, end card. It ends on: save this for the next time the scale ruins your morning.
Inside the log (about 15% of posts)
The app reels. They still open on the pain (a slow log, a guessed weight, a plan that doesn't adapt) and then show the app removing it, in real time, on a real screen. Brand accounts end with a direct ask. Founder accounts run the same shape as 'what I built' with a soft end.
- Hook: How I log a full gym session in under 60 seconds
- Hook: My app told me to deload this week. Here's what it saw
- Hook: What a week of my training actually looks like inside the app
- Hook: I stopped planning workouts. This plans them from my last session
Example: "My app told me to deload this week". Screen. 0 to 3s: 'I didn't want to deload. The app disagreed.' 3 to 12s: the readiness or fatigue screen, narrated: what it saw (three sessions under last week's numbers, poor sleep logged). 12 to 25s: what it changed in the plan, old and new weights side by side. 25 to 35s: 'this is the stuff I used to ignore until I got hurt.' Brand account: direct ask. Founder account: 'anyway, deload week.' It ends on: if you train hard and never deload, this is the feature. [App], link in bio.
Gym and run culture (about 12% of posts)
Relatable observational humour about the life around the app: the gym, the run club, meal-prep Sunday, the 5am alarm. It proves the account is made by someone who actually trains and brings in people who would never search for a tracker. Humour is the strongest-performing emotion in this niche's cold start.
- Hook: Things every gym person knows but nobody says
- Hook: POV: you're the only one at your gym who writes down their sets
- Hook: Run club etiquette nobody wrote down
- Hook: The 5 stages of forgetting to log your lunch
Example: "POV: you're the only one at your gym who writes down their sets". Camera. 0 to 2s: founder with phone, caption, deadpan. 2 to 20s: four fast beats: the look from the guy curling in the rack; checking last week's weight mid-set; the 'what app is that' question; the rest timer going off. Each beat a sound effect and a caption. 20 to 25s: 'yes I know exactly what I lifted on 14 March' with half a second of the app's history. Cold stop. It ends on: tag the friend who thinks they'll remember.
Built by someone who trains (about 8% of posts)
The founder's story: why the existing trackers annoyed them, what users asked for, the feature shipped because of one comment. It gives the brand a person and a reason to trust the data. Mostly for founder accounts; brand accounts use it sparingly.
- Hook: I built a gym app because the one I used had too many taps
- Hook: A one-star review changed my app. Here's what it said
- Hook: Why I refuse to put red numbers in my nutrition app
- Hook: The feature I built after I got injured
Example: "Why I refuse to put red numbers in my nutrition app". Camera. 0 to 3s: the line. 3 to 15s: the memory of going over by 200 and the old app turning red, and what that did to the next day. 15 to 28s: the decision: targets adjust, nothing flashes, going over is data. Screen of an over day in the app, shown plainly. 28 to 35s: 'if a chart ever made you feel like a failure, that was a design choice, not you.' It ends on: follow for the next design decision: why there's no cheat day.
Hooks that work, and hooks that flop
The first second decides everything. These are the patterns that hold this audience, and the ones that lose them.
Works
- Stop doing [X] if you want [Y]: contrarian openers outperform 'today I'll show you' several times over in fitness short-form; the viewer checks whether they do X.
- Identity call: 'If you've trained 6 months and look the same...': names the viewer's situation without naming their body; they feel seen, not judged.
- A number everyone knows that's wrong: 10,000 steps, 8 hours, 1,200 calories: killing a familiar number is instant tension and earns saves.
- Confession: 'I quit tracking four times': gives permission to the viewer who fell off; trust without claiming authority.
- 'My app told me to [deload / rest / eat more]': curiosity plus a natural demo beat; the app is a character, not a pitch.
- A protocol with numbers and a time: '3 sets, twice a week, 6 weeks': specificity reads as real; vague tips read as generated copy.
- 'What my watch thinks I did vs what I did': humour performs best in fitness reels, and this one sets up the log as the honest record.
- 'Week 3 is where everyone falls off': names a moment the viewer recognises from their own history.
- 'Your [easy runs / rest days / scale] are lying to you': a truth about something they already do; curiosity without a body claim.
Flops
- Today I'll show you the best exercise for [muscle]: the default teacher opener; the fitness reel data puts it far behind contrarian and curiosity hooks.
- Transform your body in 30 days: reads as an ad, trips platform and ASA rules on weight-loss claims, and the audience has seen a thousand.
- Introducing [feature] / New update: nobody follows a tracker for its changelog; the pain has to open.
- Motivation Monday / No excuses: the viewer who fell off is the audience; shame makes them scroll.
- What I eat in a day, totals on screen: saturated, invites calorie policing in the comments, and drifts toward diet-culture flags.
- Trust me, I'm a [trainer / founder]: trust-led openers are the weakest emotion in this niche's cold start; show the number instead.
- Rate my physique / flex check: body-first content is exactly what this playbook avoids, and zoomed body shots breach Meta's health rules.
Where the app shows up
Most posts shouldn't sell. A good split is about 50% for reaching new people, 30% for building trust and 20% that ask for the download.
Reach reels: the app is half a second of a real screen as 'where I saw this', or absent, with the name only in the caption. Trust reels: the app's chart or log is the worked example midway, named once. Convert reels, brand accounts only: the screen is the whole middle and the ask is the last 3 seconds, spoken and captioned, pointing at the pain just solved. Founder accounts never ask for a download; they ask for a follow, a comment or a share.
Calls to action that fit
- Save this for your next [muscle] day.
- Send this to whoever started the gym with you in January.
- Try one genuinely easy run this week and tell me how slow it felt.
- If the scale ruined your morning, share this with someone it'll help.
- The free version does everything in this video. Search [app] on the App Store.
- If you've bounced off every tracker, give [app] one week. Link in bio.
Asks that backfire
- 'Comment [word] for the link'; banned in UserClimb and reads as a funnel
- 'Start your transformation' or any ask tied to a body outcome
- Discount codes in a weight-loss context: Instagram removes 'miraculous claim plus code' posts and hides priced weight-loss posts from under-18s
- 'Download now' with no pain named
- Asking for a follow before giving anything
TikTok, Instagram, YouTube and X
- TikTok: Gym humour, POV, reply-to-comment videos and 'my app told me' demos do best; fast cuts, deadpan captions, a brand sound you reuse. Anything framed as rapid weight loss or 'muscle gain product' marketing is age-gated and kept out of the For You feed, so never frame a reel that way.
- Instagram: Reels are the main surface for lifters and runners; saves and shares drive reach, so myth vs fact, number explainers and 'send to your run club' lists work. Keep the look clean: the app's chart card, white graphics. Zoomed bodies, scales and before/after images trip Meta's health and appearance rules.
- YouTube Shorts: Shorts behave like search: 'how to find your easy pace', 'what is sleep debt', 'how to log a superset in [app]'. Educational openers survive better here than on TikTok. 45 to 60 seconds is fine; point to the full tutorial.
- X: Lifting and running discourse lives here (the evidence-based crowd, run data nerds, indie app founders). Post the chart with a one-line take, a build-in-public note, or the 'why no red numbers' opinion as native video; link in the first reply.
Accounts that grew this way
Real accounts worth studying for fitness and health apps, and what they did that worked.
- Hevy on Instagram: Bootstrapped gym tracker that grew on word of mouth and in-app social features, with very little ever spent on ads; Instagram and TikTok are its core social channels, run for Gen Z lifters alongside an ambassador programme. instagram.com/hevyapp
- Hevy on TikTok: The same brand on TikTok, hiring for 'regular viral hits' and measuring social by downloads as well as engagement; gym-culture content for young lifters rather than feature tours. tiktok.com/@hevyapp
- Guillem Ros (Hevy co-founder) on X: Talks openly about Hevy's growth (viral loops from the social feed, low pricing so users keep recommending it, organic over paid); a model for the founder account in this niche. x.com/theguiros
- MacroFactor on YouTube: Education-led nutrition tracker from Stronger by Science: tutorials, evidence articles and an 'adherence-neutral, no red numbers, no shaming' stance; growth credited to word of mouth from a community that trusts the data. youtube.com/@MacroFactor
- Runna on Instagram: Running plan app that grew from runners telling runners, then an ambassador and affiliate network of real runners and coaches posting their own training; the promise is a race goal, never a body. instagram.com/runna
- Ben Parker (Runna co-founder and head coach) on Instagram: The coach as the face: plain-language run advice and his own training, with the app inside the story. instagram.com/benparkerfitness
- JuggernautAI on Instagram: Strength programme app that sells 'you don't have to think about your workout' through readiness scores and auto-adjusting loads, built on a long-running coaching and education brand. instagram.com/juggernauttraining
- Rise on TikTok: Sleep app that reframed the category as 'sleep debt, not sleep score' and 'sleep more is a scold'; shows how to teach one number per reel without a wearable. tiktok.com/@risescience
- Boostcamp on Instagram: Free programme library ('the free workout app lifters actually use'); content leads with coach-designed programmes and progress charts rather than the logger itself. instagram.com/trainwithboostcamp
If you're the founder
A founder of a fitness app shows up as someone who trains and got annoyed enough to build the tool: their own log with the gaps left in, the design decisions (why no red numbers, why the plan adapts to a bad night's sleep), the one-star review that was right, what shipped this week. The audience follows them for honest training content and a builder's view of a crowded category, so the app appears inside the story and the asks are for follows, comments and shares. Never a download ask, never revenue screenshots as the point, and never their body as the proof.
Lines not to cross
- No before-and-after body photos, zoomed abs or anyone on a scale, in any reel. Meta's health and appearance ad rules name all three as non-compliant, Instagram hides priced weight-loss posts from under-18s, and the audience reads them as diet culture.
- Never state an amount of weight lost in a time ('10 lb in 10 days'), a rate above about 1 kg (2 lb) a week, or fat loss from one body part. UK CAP Code section 13 bans each; the US FTC treats a claim implied by an image the same as a spoken one.
- Testimonials and reviews are not proof. A real user quote can appear as a quote, never as 'this works'; a 'results not typical' line does not cure the claim.
- No rapid or drastic weight-loss or muscle-gain framing, no extreme fasting, no 'what you eat in private shows up in public' tough-love lines. TikTok age-gates that content and keeps it out of the For You feed; SkinnyTok is the cautionary tale.
- The app estimates and suggests; it never diagnoses, treats or cures. No 'fixes insomnia', 'heals your back', 'reverses' anything. Physio and recovery reels end with 'see a physio if it persists'.
- No body-fat or muscle-mass figures from a photo or an 'AI scan' presented as measurement, and no 'clinically proven' unless the founder holds a trial on the app itself.
- Ambassadors, affiliates and gifted users say so on screen; the FTC and the ASA treat an undisclosed material connection as deception.
- No numbers invented for the reel: every stat is the founder's own [fill-in] or a named published range. A creator build with a fake perfect streak is off the table.
- Keep good food, bad food, cheat days and 'earn your food' language out; the accounts that win in this niche are explicitly non-shaming.
- No footage of other gym-goers without consent, and no AI-generated bodies or faces standing in for a demo.
Making these without a marketing team
You can film these yourself on your phone, record your screen, or build them without filming at all. Demos, the founder's real log, data explainers, the 'my app told me' beat.
If you'd rather not start from a blank page, UserClimb turns your fitness or health app's link into motion graphic videos built on hooks like these: a hook on your viewer's problem, a quick lesson, then your app as the answer, voiced, captioned and set to music. It posts them to TikTok, Instagram and YouTube for you. Your first video is free, plus 6 free posts, no card.
Questions
- What should I post to market a fitness or health app?
- A mix of myths and form fixes (25%), why you fall off (20%), the one number to watch (20%), inside the log (15%), gym and run culture (12%), built by someone who trains (8%). Most posts should help or entertain the viewer; only a few should ask for the download.
- What hooks work for fitness and health apps?
- Stop doing [X] if you want [Y]. Identity call: 'If you've trained 6 months and look the same...'. A number everyone knows that's wrong: 10,000 steps, 8 hours, 1,200 calories.
- Should I show the app in every video?
- No. Reach reels: the app is half a second of a real screen as 'where I saw this', or absent, with the name only in the caption. Trust reels: the app's chart or log is the worked example midway, named once.
- Is TikTok or Instagram better for fitness and health apps?
- Both work, differently. Gym humour, POV, reply-to-comment videos and 'my app told me' demos do best; fast cuts, deadpan captions, a brand sound you reuse. Reels are the main surface for lifters and runners; saves and shares drive reach, so myth vs fact, number explainers and 'send to your run club' lists work.
Keep going
Rather not work out what to post? Paste your app's link.
UserClimb makes motion graphic videos for your app: a hook on your customer's problem, a quick lesson, then your app as the answer, voiced, captioned and set to music. Then it posts them to TikTok, Instagram and YouTube. Your first video is free, plus 6 free posts, no card.