Stringers
Mobile App · iOS · written 11-14
핏츠 PITZ - AI 헬스 트레이너, 운동기록 식단 PT · yongbin lee

Every one of the week's 28 good reviews describes the same workflow: PITZ, the Korean AI fitness coach, hasn't broken out yet. Its cold-start and gym-entry plays are worth studying

What it is A Korean-only AI personal trainer app: after each set you tap how hard it felt, and it tells you the weight and reps for the next set. You can also chat with an AI coach of your chosen personality and snap photos of meals. Subscriptions run about $7 to $9 a month.

Why it's getting attention It isn't really. In early November a batch of review-team participants wrote 28 good reviews from a fixed script, pushing the rating from single digits to 35. It also landed on the Korean Health & Fitness Top Grossing chart at #110, where an app with only a couple of ratings can still show up.

CHow you could follow it Don't build another generic AI fitness coach. Borrow the gym-entry play: give physical gyms a free membership system so members check in through your app. Target niches the big apps ignore, like Pilates, boxing gyms, and climbing. For cold start, run the sequence: insider creators → micro-influencer experience posts → review tasks.

Grade C: the product has no moat. It's up against Planfit, a Korean rival with ₩5B in revenue, and a row of older apps with tens of thousands of ratings. PITZ itself has only a few hundred users. The cold-start playbook and the gym-entry idea are worth taking, but the gym play only launched in December and isn't proven yet, so copying it now isn't a good bet. Almost nobody knows about it

Est. monthly revenue (App Store) under $1KWe estimated this from its chart ranks and rating growth across storefronts, calibrated against verified revenue; typical error is 2–3x. Read it as a tier, not an exact number.

Data (updated daily) · First launched 2026-09-25 · Korean rating 35 · 11-15 charted in 1 stores
iOS · best category chart rank · by store
11211411611811-1011-15118
Data 2026-11-10 → 2026-11-15 (6 days, 2 observations). Lower rank is higher; gaps mean it wasn't charted that day
iOS · rating count · Korea
01020304011-0611-1011-1535
Data 2026-11-06 → 2026-11-15 (10 days, 3 observations). Apple doesn't publish download counts; rating growth is a proxy for downloads.

PITZ (핏츠) is an AI fitness coach app from a 6-person team in Seoul, released on 25 September. In November its Korean App Store rating went from 11 to 35 in a week, and it slipped onto the Health & Fitness Top Grossing chart, which looks like momentum. Look closer: the reviews were stamped out from a script, the chart bar is low enough to mean little, and the team itself says it had only 400+ users after launch. Don't copy the product. Two things are worth taking: a Korean-style cold-start playbook that costs little, and an unproven way to acquire users by giving gyms a free system and getting members to install the app.

Nov 12: seven nearly identical good reviews

On November 2026 12, PITZ got 7 written reviews on the Korean App Store. Three of them:

"I did dumbbells at the gym, logged them, tapped done, picked how hard it was. Afterward the coach gave feedback and the weight and reps for the next set, and I did the next set at the suggested weight." (까메오코코, 5 stars)
"I used to only log weight and reps after barbell work. In 핏츠 I log it, tap done, pick the intensity, read the feedback, check the next set's recommendation…" (준호지민아, 5 stars)
"I'm used to logging dumbbells, but picking how hard it felt was a bit annoying at first. Still… once I saw the feedback and the next set's weight and reps, I got why you fill it in." (기로실, 4 stars)

These three follow the same route as the other reviews that week: log → 세트 완료 (complete the set) → pick how hard it felt → read the coach's feedback → check next set's weight and reps → do the next set at the suggested weight. From November 3 to 13, the Korean storefront got 28 written reviews: 24 at 5 stars and 4 at 4 stars. Of those, 18 followed the training route above. Others followed a diet route ("pick the diet coach → photo-log a meal → get feedback in chat → ask what to eat for dinner") or a chat route ("pick a coach during rest → ask one question → follow up twice"). Many end with some version of "a bit of a hassle, but I'll keep using it." The author of review 11-08 goes by 그놈의 친구초대, roughly "that damn friend invite" (App Store Korea reviews, 11-03 to 11-13).

Before that, over the prior month or so, the app had only 2 written reviews in total, all from mid to late October.

This is the pattern of Korean "앱 리뷰 체험단" (app review experience groups): participants download the app, walk through the core features on instructions, write a detailed review, and get a reward. Which platform they used and what each review paid aren't publicly known. The rating curve fits: it was 11-06 at 8, 11-10 at 11, 11-15 at 35, then 24 more in the final five days (App Store Korea).

What it sells

A Korean-only AI personal trainer. The first line of the store page is two thoughts: "Is this weight right for me?" and "Training alone, I always keep putting it off." The four core features:

  • Real-time weight adjustment per set: after each set you tap how hard it felt (RPE), the app rates it as "under / fine / overload," and then gives the weight and reps for the next set.
  • A coach you can chat with between sets: pick appearance and personality. It remembers your body data, training history, and past chats, and you can also talk about weekend plans on rest days.
  • Photo meal logging: recognizes the food, calculates calories and macros, and the coach comments on the meal.
  • Monthly InBody report: enter body composition and weight, and get a monthly report on your body changes.

Subscriptions come in two tiers: Pro at ₩9,900 a month or ₩59,000 a year, Pro+ at ₩12,900 a month or ₩77,000 a year. That works out to roughly $7–$9 a month, with a 7-day free trial (App Store Korea in-app purchase items).

On a top chart, but still can't make payroll

In November 10 it sat at #113 on the Korean Health & Fitness Top Grossing chart, and by November 15 at #118. It's not on the overall chart (App Store Korea).

That chart bar is very low. Around November 15, the apps in the same band: #106 had only 2 ratings, #115 had 13, #121 had 1, #123 had 5, and the one with the most ratings in that band, #105, had only 29. At that level, a handful of subscriptions or trial conversions a week is enough to chart here (estimate). So being on the chart doesn't prove it's reliably earning.

The team's own account is more direct. In November 5, operator 빈즈파이렛츠 (Beanspirates) posted on Korean developer community 인프런 looking for a product designer. The post said that after the official launch it "gained 400+ users." The team has 6 people: a CEO, a PM, 1 frontend, 2 backend, and 1 on AI. Pay was listed as "salary and formal hiring discussed after revenue or outside investment." The same post lists its grants: the 2024 "생애최초청년창업" program, the 2024 "학생 유망 창업팀 300" program, and the 2026 "재도전성공패키지," all government or university startup funding. On Google Play it shows 100+ downloads (Play store listing, November).

In other words: an early team propped up by government grants, with a few hundred users and close to zero revenue (estimate).

Where users came from: four steps, each cheap

TimelineWhat they didDetail
September 3–13Recruited 300 beta testers through the university activity platform LinkareerRewards were small gifts and transport money, prioritizing "creators / active on social media." The team says it ran the beta with 100+ people in the end.
Mid-OctoberInstagram micro-influencer experience postsA batch of lifestyle and marketing accounts posted with #광고 (ad) tags, all using the same @pitz.official and #핏츠운동기록 hashtags. Several captions were nearly word-for-word identical, opening with "그래서 사용해 본 운동 기록 앱 핏츠(PITZ)" ("So I tried the workout logging app 핏츠 (PITZ)"). Clearly the same script.
Around Oct 16University partnership press releaseIncheon National University brokered a co-marketing deal with skincare brand Better Habit. Local outlets including 라이브뉴스, 국제뉴스, and 중도일보 ran the same story.
November 3–13App Store review experience groupThe 28 reviews above, all on the same route, moved the store page from "8 ratings" to "35 ratings, 4.9 stars."

Two design choices stand out. First, every step targets people who like to post: beta testers are creators, the experience groups recruit micro-accounts, and the review group is paid content itself. Each step builds material for the next. Second, the review script hardcodes the core path. Everyone paid to write a review was asked to walk through "log → complete → pick intensity → read the recommendation." That produced detailed five-star reviews for the store page, and it also got those users through the new-user activation flow.

What I didn't find: no entry in Meta's public ad library; no mentions on X, TikTok, or Reddit in the last two months; the official Instagram has only 264 followers.

The truly different step: a free gym system

The most interesting part of PITZ isn't in the app. It's 「핏츠 매니저」 (PITZ Manager, biz.pitz.co.kr), launching in Dec 1.

It's a membership management system for gyms, with both the sign-up fee and the monthly fee set at ₩0. The team visits gyms to install two things: a QR check-in code at the front desk and NFC stickers on the equipment. Members scan the QR with the PITZ app to enter, and the gym's back office sees their attendance and training. When members buy things like protein inside the app's store, the gym takes 5%.

Run the numbers: the gym gets a free front-desk and membership system plus a small cut of sales. PITZ gets an install channel, with the gym's front desk getting members to install the app every day, at zero paid-acquisition cost. Other fitness apps rely on ads and influencers. If this works, acquisition costs are nearly zero, and it's tied to physical locations, which makes it hard for rivals to copy.

But right now it's only a blueprint: it has signed 3 gyms in the Seoul metro area, with roughly 2,400 members in total (라이브뉴스, 10-16). The 인프런 post from Nov 5 just says it has "signed several." Whether front desks will actually push it, and whether members will pay after installing, won't be clear until December.

Who it's competing with

CompetitorsLaunchedScale (Korean App Store, November)How PITZ differs
플랜핏 Planfit2020Health & Fitness Top Grossing #1, about 30k ratings; ₩5B in revenue in 2025, 40% from overseas (머니투데이, 2026-03)Also pitches an AI coach, real-time guidance, and logging, but earlier and more fully
짐워크 Gymwork2021#2,3,557 on the Top Grossing chart; the same company has two other apps in the top 20A logging tool with no personified coach
필라이즈 / 인아웃2021Top Grossing #3 / #414k / 58k ratingsAn AI coach focused on diet and weight loss. PITZ's photo meal logging goes head-to-head with it
플릭 Fleek2021Top Grossing #1.11M ratingPure training log
번핏 Burnfit2020Top Grossing #24, about 7,100 ratingsLogging plus an AI training plan, also uses Instagram experience groups to build volume
회원님Lite2026Just launched on Google PlayThe closest match: a chat-style virtual trainer with two named coaches and set-by-set logging
Alpha Progression, Fitbod2019 / 2015Top global playersHas long recommended next-set weights based on how much effort you have left

PITZ's own three selling points, and the first two aren't new. "Set-by-set weight adjustment based on RPE" is already in Planfit and Alpha Progression; "selectable personality coaches with memory" is also in 회원님Lite, and anyone can build that. Plus many people just ask ChatGPT what to train and eat, and it ranks #1 on the Korea Top Grossing chart. Only the third, the gym entry point, is something nobody else has.

If it were you

C: learn something from it, but following it along isn't worth the money. Building another Korean-language AI personal trainer app means facing Planfit, which reports ₩5B in revenue, and a crowd of older apps with tens of thousands of ratings. PITZ itself has only 400 users. What's worth taking is its cold-start sequence and the gym entry idea, which is still unproven.

Who it suits: people who already have relationships in an offline "membership + front desk" setting, such as owners of Pilates studios, boxing gyms, climbing gyms, or yoga studios, or who run one themselves; and small teams launching an app in markets like Korea or Japan, where experience-group marketing is mature, who need to build a store page from zero ratings.

Not for: Anyone going head-on in the general AI fitness coach space: the features are easy to copy, so it comes down to paid acquisition budget and brand. Anyone without offline relationships or unwilling to go on-site to install equipment: a 2B2C path is mostly done in the store. Anyone hoping to stand out just with a "personality AI coach": the barrier to entry is too low.

How to do it: don't build a big all-in-one fitness app. Pick a gym type that top apps don't serve, and give the gym a genuinely useful free tool (check-in, class booking, membership expiry reminders), on the condition that members check in through your app. Inside the app, build only the one or two things that type of exercise needs most, like post-class notes for Pilates or route logging for climbing. Money comes from membership subscriptions and a cut of in-store sales, with a share passed to the gym.

How to get users: two legs. The gym leg comes first: sign 3 to 5 gyms willing to cooperate, prove that "you need the app to check in" works, then use real data to pitch the next batch. The store page leg follows PITZ's order: first recruit a group of beta users who like posting about things, then find small creators for review posts, and finally use review tasks to script the core activation flow so that people doing the task walk new users through it. This costs little in Korea. Other markets first need to confirm that this kind of experience-group ecosystem exists, and you have to watch the store's rules on incentivized reviews.

When to do it: no rush. PITZ's gym system launched in Dec 1, and by February 2027 you'll see whether gyms actually push it. That's the time to decide whether to copy it. If you only borrow the cold-start playbook, get the store page up before the January New Year fitness season.

What would mean you got it wrong: ① by March 2027, PITZ's signed gyms grow from 3 to dozens and the Korean rating climbs into the hundreds, with new reviews no longer following the same script. That would show the gym entry really drives volume, and this read underestimated it, so raise the grade; ② it raises funding, or enters the Health & Fitness Top Grossing top 30; ③ the reverse: if a major player like Planfit or 짐워크 launches a free gym system first, the entry is taken by big companies and small teams should stay away.

iOS store page ↗

Research notes: competitor table, acquisition evidence, full research report
Competitors
NameWhereLaunchedScaleHow it differs from this one
플랜핏 PlanfitiOS2020-05Health & Fitness Top Grossing #1 in Korea, Korea iOS 29,814 ratings, 7 days +170; 2025 revenue ₩5B, claims 4M users worldwideAlso AI coach, real-time guidance, and logging, and earlier and more complete; PITZ only adds RPE-based weight adjustment per set, selectable personality coaches, and chat that remembers you
짐워크 GymworkiOS2021-08Top Grossing #2,3,557 by ratings; the same company's 칼로그 and Condi also sit in the top 20Workout log tool + one company runs several apps, no personality coach
플릭 FleekiOS2021-08Top Grossing #11,13,861 by ratingsPure workout log, no AI chat coach
번핏 BurnfitiOS2020-03Top Grossing #24,7,108 by ratingsLog plus AI training plans; also builds volume through Instagram #광고 experience groups
필라이즈 / 인아웃iOS2021-12Top Grossing #3 / #4,14,304 / 57,831 by ratingsAI coach focused on diet and weight loss; PITZ's photo-based food recognition competes directly with them
회원님Lite(HashBrown)Android2026Google Play 0+ downloadsClosest rival: a chat-style virtual PT with two named coaches and per-set logging, showing the bar for personality AI PT is very low
Alpha Progression / Gravl / FitbodiOS2019 / 2023 / 2015US ratings 2,304 / 6,432 / 290kAlready recommends next-set weights by RIR and generates AI training plans; no Korean language, no companion-style personality
ChatGPTBig-company features2023Korea iOS Top Grossing overall #1General Q&A, not tied to workout logging
Where users come from
  • Verified Instagram micro-influencer experience group (#광고): in mid-October a batch of lifestyle and marketing micro accounts (beencong_11, hatt.zip, lx.yujin, team_d.shot_yuni, etc.) posted #광고 @pitz.official, all with the #핏츠운동기록 tag, and several captions were word-for-word identical
  • Verified App Store task-based reviews (review experience groups / reward missions): 11-03 to 11-13 in the Korean store, 31 reviews in total, almost all following three scripts: "log → set complete → pick effort level → read next weight/reps" and similar. In one week the rating went from 8 to 35
  • Verified Linkareer Beta recruitment: 2026-09-03 to 09-13, recruiting 300 people, with gifts and transport fees as incentives, priority to bloggers and SNS users; the team says it has run a beta with over 100 people
  • Inferred Gym B2B2C (핏츠 매니저 free CRM): free CRM, members enter via QR code; mall commission splits 5% to the gym; 3 gyms signed, about 2,400 members, launched Dec 1
  • Verified Government and university support, local media press releases: 재도전성공패키지, 생애최초청년창업; the Incheon University agreement appears in 라이브뉴스, 국제뉴스, and others
  • Inferred Paid acquisition (Meta etc.): not found in the metaads library (the library only tracks advertisers on its list)

핏츠 PITZ - AI 헬스 트레이너 (iOS 6802202356) research report

Research date: 2026-11-07 (data as of the library's 2026-11-15 records). Channel: iOS, Korean store; developer account yongbin lee (personal account), operator 빈즈파이렛츠 Beanspirates (near Gasan Digital Complex, Geumcheon-gu, Seoul)

Bottom line

It hasn't actually taken off yet. This is an early team of 6 people who can't yet pay salaries, with about 400 users after the official launch. The two signals for its rebound are both unreliable:

  • Ratings jumped from 11 to 35 (+24 in a day), almost entirely from scripted task-based reviews;
  • It ranked #113–118 in the Korea Health & Fitness Top Grossing chart, a very low bar in that range, where neighboring apps have only 1–29 ratings.

The growth tactics actually in use are three, all small-spend volume plus government support: ① October Instagram lifestyle micro-influencer #광고 experience groups; ② September recruiting 300 Beta testers on Linkareer; ③ preparing to use the free gym CRM "핏츠 매니저" (launched Dec 1) as a B2B2C acquisition channel.

1. What it is

A Korean-language-only AI fitness coach app, launched on the store in Sep 25. Four features:

  • Tap RPE after each set (how hard it felt); the AI judges whether the set was "too light / right / overload" and gives the next set's weight and reps;
  • Chat with the AI coach between sets; the coach has 4 selectable personalities and looks, and remembers your logs and past chats;
  • Photo-based food recognition, calculating calories and the three macronutrients;
  • Monthly body report based on InBody and weight.

Subscription: PITZ Pro ₩9,900/month or ₩59,000/year, Pro+ ₩12,900/month or ₩77,000/year, both with a 7-day free trial (from data card recall info). Website www.pitz.co.kr/.

2. Data: how many people, really

MetricValueSource
iOS rating count in Korea11-06:8 → 11-10:11 → 11-15:35 (+24), average 4.89App Store database app_day; iTunes lookup
Korea Health & Fitness Top Grossing11-10 #113,11-15 #118; not on the overall chartappstore library, chart_rank
Neighbors in the same ranking band#105 GOWOD 29 ratings, #106 Coherence 2 ratings, #115 Seed 13 ratings, #121 CGX 1 ratings, #123 고러닝 5 ratingsappstore library v_chart, 2026-11-15
Google PlayDownloads 100+, 42 ratingsPlay store page (read via Jina)
Self-reported users"Gaining 400+ users after the official launch"Inflearn recruitment post, 2026-11-05
Self-reported retentionThe 7-day return rate in September's Beta is about 2.8 times the global Health & Fitness app average라이브뉴스,2026-10-16
Team6 people (CEO, PM, frontend 1, backend 2, AX 1); designer recruitment states "salary talks after revenue or after investment"Inflearn recruitment post
Instagram official account264 followersinstagram.com/pitz.official

Conclusion:

  • The team itself said in Nov 5 it had "400+ users" and would only pay salaries once there's revenue, which suggests subscription income is near zero.
  • In the Korea Health & Fitness Top Grossing 110–130 band, apps with 1–29 ratings also rank there, so a week or so of a few subscriptions or trial conversions could get it onto the chart. Its chart position doesn't prove that subscriptions are really bringing in money.

3. The truth behind the rating surge: task-based reviews

Of the 33 most recent text reviews in the Korean App Store, only 2 were from October (10-18, 10-20), and 11-03 to 11-13 has 31, almost all following three fixed scripts (source: iTunes Korea review RSS):

  1. Workout script (about 19): "log a dumbbell/barbell exercise → set complete → pick effort level → read coaching → check the next weight/reps → (tried the recommended weight)". Almost every review walks through these same steps in order, with only the opening impression changed.
  2. Diet script (about 7): "choose a diet coach → take a photo and log a meal → check feedback in the chat → ask about dinner ideas".
  3. Chat script (about 5): "during a break, pick a trainer and chat → ask a question → read the answer and ask two more times".

These reviews also share a few common traits:

  • Every one has a "small flaw + I'd still use it" line ("입력이 좀 번거롭지만…"), with the star ratings clustered around 4–5;
  • The writing style strongly resembles LLM-rewritten text;
  • It has a nickname: "그놈의 친구초대" (that damn friend-invite).

This is the classic shape of Korea's "app review experience group / reward mission" setup: download the app, go through the core features along a set flow, write a review, and get a reward once it's done. I didn't find which platform was used (unverified).

So the 24 ratings between 11-10 and 11-15 came from people completing tasks, not from organic downloads. This is a legitimate cold-start tactic: it fills the store page with reviews to lift conversion, and it gets people to try the core features along the way.

4. Where users come from (growth tactics)

ChannelEvidenceConfirmed?
Instagram micro-influencer trial group (#광고)In mid-October (the posts show roughly 4 weeks ago), a batch of lifestyle and marketing side accounts posted #광고 @pitz.official with the same set of tags #핏츠 #핏츠운동기록 #PITZ #ai헬스트레이너. Examples: beencong_11 ("브랜드 마케터의 취향 셀렉트"), hatt.zip, lx.yujin, team_d.shot_yuni, plus one account named "아이쓰". Several captions are nearly word-for-word identical ("그래서 사용해 본 운동 기록 앱 핏츠(PITZ) 💪 인바디 기록을 바탕으로 운동 목표를 설정하고…"), which points to scripts supplied by an agency.Confirmed
App Store task-based reviewsSee section 3: 11-03 through 11-13, 31 reviews total, all following the same scriptPattern confirmed; platform not verified
Linkareer beta recruitment2026-09-03 to 09-13 recruiting 300 people, with perks including gifts and transport fees, prioritizing "power bloggers/SNS" (linkareer.com/activity/348319). The team says it has run "100+ person betas".Confirmed
Gym B2B2C: PITZ ManagerA free membership CRM for gyms (biz.pitz.co.kr): the setup fee and monthly fee are both 0. Members buy things in the app, and 5% of that goes to the gym. The team visits in person to install QR entry badges and NFC stickers on equipment. Members need PITZ to get into the gym and to be visible to the CRM, so the gym effectively does user acquisition for it. It has signed 3 gyms in the Seoul metro area with about 2,400 members (LiveNews 2026-10-16). Launched Dec 1 (Inflearn).Confirmed, but not yet launched; no traffic to show for it so far
Government and university support2024 생애최초청년창업, Student Startup Team 300, 2026 재도전성공패키지 (Incheon National University). Incheon National University brokered a co-marketing deal with skincare brand Better Habit, with local press coverage (LiveNews, 국제뉴스, 중도일보, 인천뉴스).Confirmed
Paid acquisitionNot in the Meta Ad Library (metaads lookup found nothing); the library only tracks advertisers on its list, so a miss doesn't mean no ads were run.Not found
X / TikTok / communitiesx-radar found 0 posts mentioning "핏츠" in the last 60 days. No Reddit, HN, or major media coverage found.Not found

Where the money is supposed to come from: by design, member subscriptions (9,900–12,900 KRW/month), plus a cut from in-app shop sales (protein foods and the like) with 5% going to the gym. Neither stream has picked up yet. For now the team runs on government grants and is pitching investors (Inflearn).

5. Competitors

ProductWhereLaunchedScaleHow it differs from PITZ
플랜핏 PlanfitiOS/Android2020-05Ranked #1 on Korea's health & fitness Top Grossing chart, with 29,814 iOS ratings and +170 in 7 days. Revenue for 2025 was ₩5B, doubling year over year, with 40% from overseas (Money Today 2026-03-10). It claims 4M users globally.It's also pitched as "AI coach + real-time guidance + logging," and it already offers real-time AI guidance and a Planfit GPT. PITZ's differences are only per-set RPE weight adjustment, an optional personality coach, and a chat that remembers you.
짐워크 GymworkiOS/Android2021-08Top Grossing #2,3,557 rated; same company also has 칼로그 (#16) and Condi (#17)A tracking tool plus one company running several apps; no personified coach.
플릭 FleekiOS/Android2021-08Top Grossing #11,13,861 by ratingsPure workout log, no AI coach chat.
번핏 BurnfitiOS/Android2020-03Top Grossing #24,7,108 rating; also using #광고 on Instagram for the experience groupA workout log that adds AI-recommended routines; it uses the same influencer-driven playbook as PITZ.
필라이즈 / 인아웃iOS/Android2021-12Top Grossing #3 / #4,14,304 / 57,831 by ratingsAn AI coach focused on diet and weight loss; PITZ's food-photo feature competes with it directly.
라이즈 RISE / 애슬로직iOS2023-11 / 2026-01Top Grossing #35 / 36,542 / 108 ratingsA small team's workout-plus-diet app from the same generation, noticeably bigger than PITZ.
회원님Lite(HashBrown)Android20260+ downloadsThe closest rival: a KakaoTalk-style chat with a virtual PT and two named coaches (HIIT coach 김유미, hypertrophy coach 이준호), with per-set logging. It shows that a "personified AI PT" is something anyone can build.
Alpha Progression / Gravl / FitbodOverseas iOS (also available in Korea)2019 / 2023 / 2015US ratings 2,304 / 6,432 / 290kAlpha Progression already recommends the next set's weight by RIR, and Gravl and Fitbod use AI to build routines. PITZ's "RPE-based weight adjustment per set" isn't new.
General-purpose large models like ChatGPTBig-company features—Korea Top Grossing overall #1Many people just ask ChatGPT how to train or what to eat. PITZ's value is in tying the conversation to the logging.

I found no product that's clearly copying PITZ. It's too small to be copied yet.

6. What's the differentiation, and is it why it broke out?

PITZ's own stated differentiation has three layers:

  1. Auto-adjusting the next set by RPE: it works, but Planfit, Alpha Progression, and the RP family of apps already do this, so it's not distinctive.
  2. An optional-personality, memory-keeping AI coach: a companion-plus-character-chat idea that works whether or not you're working out, aimed at raising open frequency. Korean users are already used to character chat (there are lots of AI girlfriend apps), but 회원님Lite is already doing the same thing, so the bar is low.
  3. Trading a free gym CRM for a member entry point (B2B2C): this is the most distinctive layer. Most consumer fitness apps rely on paid ads or influencers. PITZ wants the gym front desk (QR entry) to be the install funnel, then shares shop revenue back with the gym. If that works, acquisition costs almost nothing and the product is tied to an offline setting that competitors can't easily copy. But it only launched in Dec 1, has signed 3 gyms, and has about 2,400 members.

So differentiation isn't why it got recalled this time. The signal came from task-based reviews, plus a low-bar category Top Grossing rank.

7. Can you see it? (visibility)

Hardly anyone knows about it. Evidence:

  • Revenue isn't public, and the team itself says it hasn't earned revenue yet;
  • The only coverage is local press releases about the Incheon university startup program, with no tech media coverage;
  • Nobody on X mentions it;
  • The official Instagram has 264 followers;
  • No copycats.

8. What others can learn

  • A cold-start combo on the store page: Linkareer beta recruiting (university students, with gifts) → Instagram micro-influencer trial group (shared tags, shared talking points) → task-based reviews that put up 30 detailed positive reviews within a week, taking the rating from single digits to 4.9 stars across 35 reviews. It's cheap, and it suits a market like Korea where the trial-group ecosystem is mature.
  • Build the core feature into the task script: every reviewer was asked to go through "log → complete → pick RPE → see recommendation." Users get reviews, and new users actually walk through the activation path (the aha moment).
  • Use a free B2B tool to get a foothold in C: give gyms a free CRM, then make money from the member app's subscriptions and shop commissions on the B side. The same play works for yoga studios, Pilates, boxing gyms, and study rooms: anywhere that runs on memberships and a front desk.
  • The bar for Korea's health & fitness Top Grossing top 100 is low, so small products can chart easily. Discount chart data when you look at other apps.

9. Still unclear

  • Which platform the task-based reviews ran on and what each one cost (unverified); whether the scripts were LLM-generated is also just a guess.
  • Actual paid subscriber count: you can't work backward from Top Grossing rank #118 to a number of transactions.
  • The September Beta's claim of "7-day retention 2.8 times higher" has an unclear basis, and the sample is only about 100 people.
  • After PITZ Manager launches: will gyms actually push members to install the app, and will those who install it pay?
  • The relationship with its predecessor app "Moobean" (a camera-counting AI workout coach): same team, but unverified whether Moobean was pulled from the store or whether its data carried over.
  • The total size of the Instagram experience group (how many posts in all, how much was spent) could not be fully counted.

Sources

This article was written by AI from public data and the data we collect daily. Numbers are sourced, and inferences are labeled.