Greg Stewart
CEO of Ladder, a strength training app
Where to hear and read about Greg Stewart1 episode or test
Greg Stewart is CEO of Ladder, a strength training app he built after Goldman Sachs tech banking and about 10 years of startups. The lesson most worth taking is to fix price, onboarding and creative one funnel step at a time, and to pair paid growth with a product that retains.
you have to be black belt at growth and product
How Ladder Nailed Product/Channel Fit on TikTok and Meta to Grow 2,000%
Greg Stewart: the story and the lessons
1Greg Stewart's path to building Ladder
Stewart went from Goldman Sachs tech banking to a decade of startups. He came from Goldman Sachs tech banking, then spent about 10 years in startups. Ladder is five years old.
A core group on the team has worked together for 10 years through several failed startups.
Ladder was built for serious lifters who wanted software to act as a trainer. Most fitness apps leaned toward cardio or home workouts, and few tried to feel like personal training. Ladder aimed to remove the mental load of planning and make each workout trustworthy.
Peers mostly put existing content into an app. Ladder wanted a Duolingo-style product that also keeps people motivated to return.
Being fitness outsiders gave Ladder a blank-sheet design. Gyms profit from members who pay and do not go. That felt wrong to the founders, who were fitness consumers themselves.
Stewart's real estate tech experience taught him that hiring smart problem-solvers beat hiring industry veterans.
Ladder launched into the pandemic with a gym plan that no longer fit. The strategy was locked at the end of 2019 for gym-goers with full equipment. Ladder rebuilt programming for people training at home with almost no equipment, guided by coaches and constant member surveys.
It bought $50,000 of kettlebells and dumbbells and offered shop access with a trial. Stewart says the pandemic was not a tailwind: launching an app was easy, YouTube was flooded and investors ignored another fitness app.
Early users came from coach followings, and the first two years were hard iteration. The first users came from coaches with 5,000 to 15,000 Instagram followers. Growth came after reopening, when members trained at the gym and at home.
The product worked in both settings.
2How Stewart's Facebook failure led to a pricing rebuild
Coach-led Instagram posts could not become a growth engine. Coach followings grew more slowly than Ladder drained them for users. Each new coach launch brought a flood of sign-ups that then faded.
Stewart calls it the opposite of an engine.
Facebook ads after the 2021 seed round failed within about three months. iOS 14 confused the agencies, and big fitness brands were bidding hard for people stuck at home. The math did not work, so Ladder stopped.
The deeper problem was the funnel: "we weren't really set up from an activation perspective to attract and move through the funnel a user with no or low context".
Stewart says hiring an agency was the wrong call. Ladder picked the wrong agency, and even a good one would have struggled in the iOS 14 environment. Stewart now runs TikTok Ads Manager himself every day.
His advice to founders: be strong at both product and growth, because paid ads only work when the product retains and you understand your user.
He paused growth and rebuilt pricing from research, not rivals' prices. The whole team stopped investing in growth to overhaul monetization. Ladder had launched at $60 a month, priced against a benchmark spreadsheet that ran from free YouTube workouts to one-to-one coaching.
Van Westendorp and conjoint studies, segmented by how much context users had, showed people compared Ladder to Spotify or Netflix at $9.99 to $15 and did not value the coach chat.
Cutting coach chat and moving to one $30 plan reset the product. Chat was the costliest feature and mattered for neither retention nor conversion. The new packaging was one concept at one price, with onboarding built to reach a first workout fast.
Ladder began the next year in a healthier place for acquisition, so the failed Facebook push was not wasted.
3How Ladder cracked TikTok
Stewart stopped building features and treated growth as an engineering problem. Runway was finite, and without a reliable engine the team would have had to pack up. They mapped growth loops where they had an edge and could learn without an agency, then dropped each as it was eliminated.
SEO came first, and hacky MVP courses showed a strategy existed but not one big enough. TikTok was second: coaches were strong on Instagram and nearly new to TikTok, but consumers were moving there.
No money went to TikTok until organic had line of sight. Ladder started a handle with one coach and dissected every video for what repeatedly got outlier views. The account went from zero to 250,000 followers in about 50 days.
Spend waited until organic showed creators could win, the right consumer was reached, people moved through the link in bio and it could scale. "we didn't think about spending a dollar on Tik Tok until we had line of sight on organic".
He read who was commenting, not just view counts. For a female coach, the team checked whether commenters were men talking about her outfit or women talking about the workout. That showed which content reached the intended persona.
It also guided the next batch of content.
Paid started at about $50 a day, copying organic winners. Early ads were brand ads from Ladder's own handle, then winning coach videos were replicated, sometimes with identical creative. That proved people would click out of TikTok and down into the funnel.
Stewart calls creative arguably the most important skill to master. Coaches are full-time teammates and creators, which he calls a superpower for the creative loop.
Fixing one funnel step at a time worked where fixing everything failed. On Facebook the team chased clicks, first workouts and conversion to pay together and could not go deep on any. On TikTok the order was creative, then the click, then the homepage, then the quiz.
"we mapped out the entire funnel and went piece by piece".
4Ladder's onboarding quiz and the white hot persona
The quiz began because sending people into the app gave no visibility. Early answers went into a Google sheet to check whether leads were getting closer to the right persona. The team spent about four months on the quiz alone.
An office room held every question printed out with live metrics, constantly moved around to push more people through. Marketing screens between questions reinforce the value prop based on answers.
Quiz answers define the persona, so each ad gets a read in 10 minutes. After many tests of conversion events, certain answers came to stand for the right persona. Stewart calls the ideal user the white hot persona and judges each ad on how well it reaches that person.
Because the quiz is on the web, data flows back to the ad networks, while a direct app install after ATT is a black box.
Ladder aims at strength-focused enthusiasts, not casual fitness. Casual fitness means cardio, an enormous market full of huge brands. Users going zero to one on motivation are expensive to acquire, and the product was not set up to convert them.
5Why Ladder builds for workouts completed, not payments
Every feature is anchored to workout completions. Stewart's north star for anything built is workout completions, which lead to long-term retention. Members get about eight or nine cheers per workout, and cheers received were correlated with retention before the feature was iterated.
Badges, streaks and friendly competition borrow from Duolingo, but none shipped all at once, and each was tailored to Ladder's customer and only built with insight it would add workouts.
Ladder takes no credit card and pushes users to first workouts. No competitor skips the credit card, but Ladder reasoned a better product benefits from more people in it. Product effort goes into the first couple of workouts, not forcing payment or an annual plan.
Stewart notes January resolution buyers inflate subscriber counts, and those who do not engage will not return the next year.
Small coach-led teams work like group chats. Members on the same persona and workout talk daily in team chats. Platform-wide topic chats cover supplements, equipment and playlists, and every major metro has a geo chat room.
The top unexpected benefit members name is making friends in real life, and some now meet up in person.
6How Stewart picks and vets Ladder's coaches
Quiz data tells Ladder which coach to recruit next. Millions of quiz responses reveal training styles, goals and equipment, and where the roster has holes. A surge of women wanting bodybuilding programs a couple of years earlier gave confidence to launch that team.
Ladder launched three coaches last year while the business grew over 2.5x. It is not a marketplace: "once we narrow down on a Persona that very clearly is in demand then we work backwards to find the right coach".
Every coach does a paid test week before joining. After multiple meetings with group heads, who screen for ego and long-term goals, the coach programs a week of workouts. The team does them and rates them quantitatively and qualitatively, and Ladder simulates engagement with intro and outro videos, voiceovers and chat.
Stewart compares it to an engineer's code test. Most coaches who reach test week pass, since the filtering happens in earlier calls.
Stewart does not judge coaches by follower count. Audience size proved a poor signal of programming quality or team fit, and the biggest followings are usually not the biggest names on the roster. Some coaches were not big influencers, so Ladder built their following with its own growth engine.
Coaches must first be world class at their programming style. With nearly 200,000 members and 18 programs, recruiting is easier now and coaches refer other top coaches.
7How Stewart sets the roadmap and manages platform risk
In-app surveys set the product roadmap. A platform-wide survey drew 5,000 replies with an average response time of 45 minutes. Members tracking workouts in spreadsheets, Apple Notes and paper notebooks led to the Ladder Journal.
An apparel store launches only after members asked for branded performance wear. It is nobody's full-time job yet, and nutrition is a candidate next.
Ladder spread its bets beyond TikTok before any ban landed. Ladder restarted work on Meta when the first ban passed Congress and the Senate. Reels serves content to new people based on what it is, so the same creators and strategy carried over, and it now sends leads on about every other day, more than TikTok.
Ladder is also adding PR and brand campaigns. Stewart's answer to investors was that it is working today and the habits carry beyond TikTok.
Stewart keeps a team of just over 20 and leans on AI. One person focuses only on AI, and no one, including Stewart, spends over half their time managing. The workout modifications feature swaps movements while keeping the coach's programming intact, learning across all teams.
He rereads Ladder's 60,000 App Store reviews to stay grounded.
How Ladder Nailed Product/Channel Fit on TikTok and Meta to Grow 2,000%
Ladder moved from a failed Meta and agency push to a reliable TikTok and Instagram engine by fixing price, onboarding and creative one funnel step at a time. That work, with a retention-first product, grew subscribers 2,000% in two years.
Stewart went from Goldman Sachs tech banking to a decade of startups. He came from Goldman Sachs tech banking, then spent about 10 years in startups. Ladder is five years old.
A core group on the team has worked together for 10 years through several failed startups.
Ladder was built for serious lifters who wanted software to act as a trainer. Most fitness apps leaned toward cardio or home workouts, and few tried to feel like personal training. Ladder aimed to remove the mental load of planning and make each workout trustworthy.
Peers mostly put existing content into an app. Ladder wanted a Duolingo-style product that also keeps people motivated to return.
Being fitness outsiders gave Ladder a blank-sheet design. Gyms profit from members who pay and do not go. That felt wrong to the founders, who were fitness consumers themselves.
Stewart's real estate tech experience taught him that hiring smart problem-solvers beat hiring industry veterans.
Ladder launched into the pandemic with a gym plan that no longer fit. The strategy was locked at the end of 2019 for gym-goers with full equipment. Ladder rebuilt programming for people training at home with almost no equipment, guided by coaches and constant member surveys.
It bought $50,000 of kettlebells and dumbbells and offered shop access with a trial. Stewart says the pandemic was not a tailwind: launching an app was easy, YouTube was flooded and investors ignored another fitness app.
Early users came from coach followings, and the first two years were hard iteration. The first users came from coaches with 5,000 to 15,000 Instagram followers. Growth came after reopening, when members trained at the gym and at home.
The product worked in both settings.
Coach-led Instagram posts could not become a growth engine. Coach followings grew more slowly than Ladder drained them for users. Each new coach launch brought a flood of sign-ups that then faded.
Stewart calls it the opposite of an engine.
Facebook ads after the 2021 seed round failed within about three months. iOS 14 confused the agencies, and big fitness brands were bidding hard for people stuck at home. The math did not work, so Ladder stopped.
The deeper problem was the funnel: "we weren't really set up from an activation perspective to attract and move through the funnel a user with no or low context".
Stewart says hiring an agency was the wrong call. Ladder picked the wrong agency, and even a good one would have struggled in the iOS 14 environment. Stewart now runs TikTok Ads Manager himself every day.
His advice to founders: be strong at both product and growth, because paid ads only work when the product retains and you understand your user.
He paused growth and rebuilt pricing from research, not rivals' prices. The whole team stopped investing in growth to overhaul monetization. Ladder had launched at $60 a month, priced against a benchmark spreadsheet that ran from free YouTube workouts to one-to-one coaching.
Van Westendorp and conjoint studies, segmented by how much context users had, showed people compared Ladder to Spotify or Netflix at $9.99 to $15 and did not value the coach chat.
Cutting coach chat and moving to one $30 plan reset the product. Chat was the costliest feature and mattered for neither retention nor conversion. The new packaging was one concept at one price, with onboarding built to reach a first workout fast.
Ladder began the next year in a healthier place for acquisition, so the failed Facebook push was not wasted.
Stewart stopped building features and treated growth as an engineering problem. Runway was finite, and without a reliable engine the team would have had to pack up. They mapped growth loops where they had an edge and could learn without an agency, then dropped each as it was eliminated.
SEO came first, and hacky MVP courses showed a strategy existed but not one big enough. TikTok was second: coaches were strong on Instagram and nearly new to TikTok, but consumers were moving there.
No money went to TikTok until organic had line of sight. Ladder started a handle with one coach and dissected every video for what repeatedly got outlier views. The account went from zero to 250,000 followers in about 50 days.
Spend waited until organic showed creators could win, the right consumer was reached, people moved through the link in bio and it could scale. "we didn't think about spending a dollar on Tik Tok until we had line of sight on organic".
He read who was commenting, not just view counts. For a female coach, the team checked whether commenters were men talking about her outfit or women talking about the workout. That showed which content reached the intended persona.
It also guided the next batch of content.
Paid started at about $50 a day, copying organic winners. Early ads were brand ads from Ladder's own handle, then winning coach videos were replicated, sometimes with identical creative. That proved people would click out of TikTok and down into the funnel.
Stewart calls creative arguably the most important skill to master. Coaches are full-time teammates and creators, which he calls a superpower for the creative loop.
Fixing one funnel step at a time worked where fixing everything failed. On Facebook the team chased clicks, first workouts and conversion to pay together and could not go deep on any. On TikTok the order was creative, then the click, then the homepage, then the quiz.
"we mapped out the entire funnel and went piece by piece".
The quiz began because sending people into the app gave no visibility. Early answers went into a Google sheet to check whether leads were getting closer to the right persona. The team spent about four months on the quiz alone.
An office room held every question printed out with live metrics, constantly moved around to push more people through. Marketing screens between questions reinforce the value prop based on answers.
Quiz answers define the persona, so each ad gets a read in 10 minutes. After many tests of conversion events, certain answers came to stand for the right persona. Stewart calls the ideal user the white hot persona and judges each ad on how well it reaches that person.
Because the quiz is on the web, data flows back to the ad networks, while a direct app install after ATT is a black box.
Ladder aims at strength-focused enthusiasts, not casual fitness. Casual fitness means cardio, an enormous market full of huge brands. Users going zero to one on motivation are expensive to acquire, and the product was not set up to convert them.
Every feature is anchored to workout completions. Stewart's north star for anything built is workout completions, which lead to long-term retention. Members get about eight or nine cheers per workout, and cheers received were correlated with retention before the feature was iterated.
Badges, streaks and friendly competition borrow from Duolingo, but none shipped all at once, and each was tailored to Ladder's customer and only built with insight it would add workouts.
Ladder takes no credit card and pushes users to first workouts. No competitor skips the credit card, but Ladder reasoned a better product benefits from more people in it. Product effort goes into the first couple of workouts, not forcing payment or an annual plan.
Stewart notes January resolution buyers inflate subscriber counts, and those who do not engage will not return the next year.
Small coach-led teams work like group chats. Members on the same persona and workout talk daily in team chats. Platform-wide topic chats cover supplements, equipment and playlists, and every major metro has a geo chat room.
The top unexpected benefit members name is making friends in real life, and some now meet up in person.
Quiz data tells Ladder which coach to recruit next. Millions of quiz responses reveal training styles, goals and equipment, and where the roster has holes. A surge of women wanting bodybuilding programs a couple of years earlier gave confidence to launch that team.
Ladder launched three coaches last year while the business grew over 2.5x. It is not a marketplace: "once we narrow down on a Persona that very clearly is in demand then we work backwards to find the right coach".
Every coach does a paid test week before joining. After multiple meetings with group heads, who screen for ego and long-term goals, the coach programs a week of workouts. The team does them and rates them quantitatively and qualitatively, and Ladder simulates engagement with intro and outro videos, voiceovers and chat.
Stewart compares it to an engineer's code test. Most coaches who reach test week pass, since the filtering happens in earlier calls.
Stewart does not judge coaches by follower count. Audience size proved a poor signal of programming quality or team fit, and the biggest followings are usually not the biggest names on the roster. Some coaches were not big influencers, so Ladder built their following with its own growth engine.
Coaches must first be world class at their programming style. With nearly 200,000 members and 18 programs, recruiting is easier now and coaches refer other top coaches.
In-app surveys set the product roadmap. A platform-wide survey drew 5,000 replies with an average response time of 45 minutes. Members tracking workouts in spreadsheets, Apple Notes and paper notebooks led to the Ladder Journal.
An apparel store launches only after members asked for branded performance wear. It is nobody's full-time job yet, and nutrition is a candidate next.
Ladder spread its bets beyond TikTok before any ban landed. Ladder restarted work on Meta when the first ban passed Congress and the Senate. Reels serves content to new people based on what it is, so the same creators and strategy carried over, and it now sends leads on about every other day, more than TikTok.
Ladder is also adding PR and brand campaigns. Stewart's answer to investors was that it is working today and the habits carry beyond TikTok.
Stewart keeps a team of just over 20 and leans on AI. One person focuses only on AI, and no one, including Stewart, spends over half their time managing. The workout modifications feature swaps movements while keeping the coach's programming intact, learning across all teams.
He rereads Ladder's 60,000 App Store reviews to stay grounded.
Ladder