How KBMH grows

DM every single new Instagram follower: 'Thanks for the follow, <name>. Are you looking for a program or just enjoying the content?' Interested people got a Zoom call where he sold the monthly subscription and created their account by hand, because the app had no login screen yet.

Idea picked up from an App Masters video. New user creates an account, finishes onboarding and skips the paywall; the app 'thinks' for a couple of seconds, then shows 50% off the yearly plan with no free trial. Shown only once; trying to exit triggers an 'Are you sure?' dialog saying they won't see it again. Revenue tracked separately in RevenueCat.

Testing creatives on purchase optimization is unaffordable at ~15 EUR CPA (needs 100-200 EUR/day per creative). Each week, run 5 new creatives in an install-optimized campaign; copy the best into the main purchase-optimized campaign, funded with at least 200 EUR/day (less never exits Meta's 50-event learning phase).

Turned off all default Meta app events and send a single purchase event worth 25 EUR whenever a user starts a free trial, buys monthly, or takes the 50% offer, with no 7-day wait. Value set below the ~40 EUR blended LTV per converting user. Each week, check that actual trial -> paid matches the projection in the P&L sheet. Advised for small budgets; big spenders can optimize with delayed events.

Daily spreadsheet: ad spend; new yearly free trials x 35 EUR projected value (replaced with actual conversions after 7 days); 50% offer revenue; new monthly revenue; total after store cut; P&L minus ad spend; and P&L minus ad spend minus a 900 EUR/day organic baseline (a deliberately high estimate of what social brings without ads). Above 0 means ads work. Needed because ATT consent is under 10% and SKAN data lags, and traffic comes from many sources (Instagram, affiliates, YouTube, Reddit, Facebook, ads). Uses RevenueCat filters per offer.

Started with launch (Sep 2024): at least one kettlebell workout reel on his personal Instagram every day, aiming for 100 reels as fast as possible, expecting nothing, then evaluate. Took ~3 months. Learned editing, hooks, which workouts get views and which exercises reach. Has not skipped a day since.

Jan-Mar 2025, with traffic but no revenue: tried one-time program purchases, monthly and yearly subscriptions; monthly at $50, then $25, $10, $25, $10; a coupon code with $15 list / $10 with coupon; login placement at start, end, before and after the paywall; free trial vs none; very long vs very short onboarding; hard vs skippable paywall; free content vs none; at least 30 paywall designs (emotional, straight to the point, pain points). Settled on email-only login (random password), later Apple/Google sign-in.

Paywall kept unchanged for a long time: a straight-to-the-point feature list (structured programs, individual workouts, members-only chat group, workout history and insights, workout and rest timers, demo videos). Yearly $100 with a 7-day free trial; monthly $10 with no trial; a coupon code field.

A WhatsApp group on Vadim's personal number, listed on the paywall as 'members-only chat group'. After purchase a welcome pop-up links to it. Members ask questions, send technique-check videos and get fast support; experienced members help beginners. Used to spot UX problems (e.g. people can't find saved workouts), catch crashing releases instantly, and run polls on missing features, with much higher response than email.

Whenever a member posts a positive message in the community group about enjoying the app, reply directly with a link to review it on the App Store.

Happened by mistake in December 2025: the offer meant for new users was shown to every free user on app open, including lapsed users with a past trial/subscription who were not eligible and were charged full yearly price (the system price dialog showed it, but people missed it). Many complaints and refunds.

In-app affiliate area where anyone generates a deep link and a coupon code giving 25% off; affiliates earn 50% of the initial purchase. Target creators were workout creators, who nearly all sell their own programs.

Like creators' posts, DM compliments and engage with their content until they engage back and follow; after ~3 months of back-and-forth, invite them. Four affiliates now (two ~150K followers, two smaller) post workouts as Instagram collabs with the KBMH account. Switched from commission to an upfront monthly retainer for one piece of content per week. Track via coupon code plus Branch.io deep links, cross-checked in-app.

Early Meta ads followed online advice: abs, promises, pain points ('don't wait for January'). None worked. After testing many creatives, a plain white banner with text in the middle (no video) became one of the first hits; two other simple ads also worked. Audience of 35+ men wants straight-to-the-point messaging, not 'TikTok nonsense'.

Tested: lookalike audience from buyers' emails (didn't work), men 35+ matching the user base (didn't work), interests in kettlebells/kettlebell equipment (kind of fine), then broad with no targeting. Onboarding data showed half of new users are beginners, so the real audience is busy people training at home, not kettlebell fans.

In January 2025, watched Brendan Kane's episode on Chris Do's podcast, downloaded his free ebook/workbook (assignments, research tasks, checklists, spreadsheets) from the video description, and did exactly what it said. Combined with small boosts of winning reels (t3).

Strategically boosted successful reels on Instagram, a couple of hundred dollars per month in total.

Built the app for himself first, then made it usable for his wife. Learned Flutter to cover iOS and Android and launched in 2-3 weeks with one program, a timer, no videos, no login screen, and many bugs. Put effort into impressions, downloads and conversion before retention features. He also advises trying a landing page with screenshots and a waitlist button before building, but did not do that for KBMH.

After authentication, a few swipeable slides show the app's features (programs, workouts, timers), because users forget the store screenshots by the time they open the app. Then questions (name, source, training days per week, level, kettlebell weight), a confirmation screen, a few seconds of 'building your plan', then the paywall.

The last onboarding feature slide showed the rate-us prompt. Works because most new users come from Vadim's Instagram and already know and like his workouts before using the app. Recently replaced with the ATT prompt for paid ads, which he plans to move back to the splash screen where it worked better.

Took reels that performed very well organically and ran them as Meta ads, per common online advice.

During Q1 2025 tested different App Store screenshots, app names and ASO. Current screenshots are straight to the point (features, no pain-point copy) for a 35+ male audience.

Around mid-2025 at ~20K/month, Vadim (PM, designer, marketer and developer) was shipping unvalidated features whenever something went wrong. Hired Sasha as mobile developer / de facto CTO; everything goes through him, Vadim learned Figma, designs and documentation come first, and Sasha converts Figma designs to Flutter. Sasha challenges ideas before they are built.

App Store/Google Play pay out 30-60 days after revenue, so scaling profitable ads can still run out of cash. Keeps a detailed cash-flow spreadsheet. Planning a web-to-app onboarding to get paid sooner, and considering a bank loan (slow in Cyprus, ~2 months just to check eligibility).