Founder research
Starter Story founder analysis: how 10 founders got customers.
I wanted to test a simple idea: founder-story videos are more useful if you study how the founders actually got customers.
I took the latest 10 non-Shorts videos from Starter Story, split them by chapter, and analyzed each founder story case by case.
I wanted to know what actually made these founders work: how they found the first users, what channel moved, and why that channel fit the product.
The obvious takeaway is that AI lets people build faster. Fine. The more useful question is how the product got in front of buyers.
So I pulled out the useful details: channels, numbers, validation steps, and how each founder got people to try or buy the product.
Short version
This is a case-by-case analysis of 10 Starter Story founder videos. The useful pattern is how each founder got customers, not just what they built.
The winners found proof that people already cared, built for one narrow buyer, tested with behavior, and stayed with the channel that moved.
The 10 cases
$30K/month app
Benji Chen / Snag
He had a repeatable app-building system. Snag had a simple economic promise, and he tested videos from creators before putting money behind the ones that worked.
How they got customers
- Paid creators to make simple short videos about the app, then turned the best videos into Facebook and Instagram ads.
- Interviewed creators first. He said roughly 9-10 out of every 100 interviewed creators were good enough to test.
- Looked for videos over 50,000 views, strong click-through, or ads that made money.
- Started paid tests around $50/day, then raised spend on the videos that kept working.
- One creator video got 240,000 views. In the video, Benji says every 100,000 views usually produced about $1,000-$2,000 in subscription profit.
Watch the video here
$50K+ monthly gross volume AI app builder
David and Daniel / Shipper
They entered a huge validated category and used competitor complaints to decide what to build differently: less technical intimidation and more output types.
How they got customers
- They said they had not used paid channels.
- Launched the MVP on Product Hunt in week one and made the first $50 in monthly recurring revenue from that launch.
- Used Reddit for early traction. They said they were regularly getting around 400 upvotes, which pushed them from $50 to $1K in monthly recurring revenue.
- Built Google search pages for competitor terms like alternatives and pricing.
- Around day 50, X started working. They made about $20K in monthly recurring revenue in one or two weeks by building in public and using their own product publicly.
Watch the video here
$30K+ monthly revenue app
Cedric Roberge / Pep AI
He spotted peptides rising through social signals, saw weak existing apps, read reviews, built fast, and recruited niche influencers before launch.
How they got customers
- Found demand through a roommate, TikTok, social media, App Store search, and weak existing peptide apps.
- Read reviews from existing apps to learn what users wanted before building.
- Got niche influencers posting before launch to test whether people cared.
- Made Reddit posts and built a waitlist of around 300 people before launch.
- One Instagram story post drove about $1,000 in revenue. One later post got 50,000 views, drove about $4,000 that day, and over $10,000 total by their estimate.
Watch the video here
$1.7M/year boring business
Bohdan Drozdov / Savvy Nomad
He chose an unsexy workflow where customers were already stressed, confused, and paying professionals.
How they got customers
- The video gives less channel-by-channel detail here. The clear move was turning a paid workflow into software: state domicile change for Americans abroad.
- They sold a concrete financial outcome. The product shows likely tax savings and then presents plans, making the buying decision easier.
- The founder says customers were choosing between scattered blog posts, Reddit threads, or professionals charging thousands of dollars.
- The stack included Framer for the marketing site, Ghost for publishing, Ahrefs for Google search research, Customer.io for email, and SavvyCal for sales calls.
- At the time of the interview they were adding roughly 120-150 new customers per month.
Watch the video here
multiple $150K/month consumer apps
Nicole / Glam Up and Sprout
She understood that consumer app design and getting customers are connected. The app had to convert, but the videos had to spread.
How they got customers
- Designed app screens and onboarding so the product could be shown clearly in short social videos.
- For Glam Up, the video cites 40-50M monthly TikTok views and around 200M Instagram views at peak.
- For Sprout, it cites 100M monthly views and one seven-day period with 400M views.
- Managed around 200 active creators during the strongest scaling period.
- Her playbook was to build a creator operation: sourcing, interviews, creator courses, content examples, quizzes, Discord or feedback calls, referral systems, and dashboards.
Watch the video here
$20K/month event app
Brian Shin / Once
He validated with a real commitment metric: event hosts actually using the disposable-camera product at real events before he rebuilt the final app.
How they got customers
- Started with a real Halloween party using a rough web app and printed invitation codes.
- Validated through actual event usage, even though the first version broke during the party.
- Worked through his personal network on X, LinkedIn, Instagram, and Facebook to find people with upcoming events.
- Searched Instagram hashtags like #wedding and #birthdayparty, listed 250-300 prospects, sent short cold messages, got around 15 replies, and fixed 12 events for that month.
- His goal before serious buildout was 10 events with real dates and real commitment to use the product.
Watch the video here
$77K/month solopreneur portfolio
Marc Lou / Marc Lou's portfolio
He treats shipped products as repeated bets. His edge is protected deep work, shipping volume, buy-button validation, and low attachment to weak projects.
How they got customers
- Launches small products publicly and quickly instead of waiting for a large campaign.
- For one macOS app, he launched with a tweet and also shared on LinkedIn, Threads, and Reddit.
- That launch produced first customers and about $1,000 in first-day revenue.
- He validates ideas by shipping with a buy button.
- Every launch also grows the audience a little because people discover the work and users post about it.
Watch the video here
$3M/year SaaS and exit
Jeremy Redman / TaskMagic
He had one main SaaS, then built smaller products around the same customer's next problem. Those smaller products brought in search traffic and sent buyers back to the main product.
How they got customers
- Used TaskMagic as the main product, then built side products that served the same customer.
- TaskMagic's original customers were business owners, agencies, freelancers, and people who needed sales.
- Built Mail Lead as a simple outbound email product for that same buyer.
- Mail Lead got its own Google traffic because it targeted more specific searches than broad automation terms.
- Mail Lead brought in almost seven figures alone, then created upgrade paths back into TaskMagic. LeadQuest added another product in the same ecosystem.
Watch the video here
$1.5M lifetime revenue
Jordan Rejaud / Parakeet Chat
He served an ignored closed ecosystem: incarcerated people and their families. The product fit a serious communication and legal-help pain, so word of mouth carried it.
How they got customers
- The video says the business had zero ads and no content.
- Because prison is a closed ecosystem, validation and MVP were the same step: he had to build it and get it in front of people inside.
- He talked to contacts in prison, got feedback, and those users told other people.
- Within one month he had 200 paying users.
- Growth was word of mouth, helped by an internal recruitment system that gave credits when one customer recruited another paying customer.
Watch the video here
$20K/month niche sports app
Ethan / CutCoach
He had lived the problem as a combat-sport athlete, built for an urgent seasonal pain, pivoted from coaches to athletes, and used small niche creators with high conversion.
How they got customers
- Started with organic posts adapted from social media pages in the same niche.
- Early posts only got 200-500 views, but still drove around 10-15 downloads per day because viewers were high intent.
- Then moved to influencer marketing, starting with small creators because there were not many big influencers in the category.
- Found creators by scrolling TikTok and Instagram Reels, then DMing them. He said most creators he DM'd said yes.
- Started with creators getting about 1,000-10,000 views per video, later moved to 20,000+ view creators, then put influencer videos into paid ads.
Watch the video here
What made the idea worth trying?
The clue changed by case. That matters. There was no one correct channel. The good founders noticed where people were already searching, paying, complaining, sharing, or asking for help.
- a painful niche problem the founder personally understood
- a social trend people were already talking about
- competitor reviews full of complaints
- a boring workflow people already paid professionals to handle
- a closed community with bad existing tools
- a large category where one small product difference was enough
- an existing customer base with obvious next problems
How they got customers
The interesting part is how specific the channels were. These were not generic launch plans. Each founder picked a route that matched the buyer and the product.
Consumer apps used TikTok, Instagram, creators, and paid social. Software products used Product Hunt, Reddit, Google search, X, and smaller products that sent buyers back to the main product. Closed markets used insiders and word of mouth.
Benji watched views, clicks, and whether ads made money. Brian wanted 10 committed events. Nicole watched creator output. Ethan watched downloads from tiny-view posts.
Creators were interviewed. Influencers were DM'd. Event hosts were listed by hand. Prison contacts were spoken to directly. Search pages and side products were built one by one.
Snag had a simple savings story. Once created shareable events. CutCoach hit athletes during season. TaskMagic side products created their own search demand and upgrade paths.
The repeatable structure
Pain, search, social obsession, existing spend, ignored communities, or repeated complaints.
The winners were not broad platforms at first. They solved one job for one identifiable user.
Payment, usage, referrals, creator posts, real events, and conversions mattered more than compliments.
Once a channel moved, they kept improving it instead of jumping to a new tactic every week.
Getting customers was the real difference
The product was often simple. The customer pain was specific. The hard part was getting the product into the right channel and staying there long enough to learn.
The general conclusion
The repeatable formula is: find evidence people care, build a narrow product, validate with behavior, then push the channel that actually moves.
AI compresses the build. It does not replace taste, timing, choosing the right customer, or getting the product in front of them.
That makes channels like Starter Story more useful than they look at first. If you treat them as entertainment, you get inspiration. If you treat them as data, you get a map of how real founders got strangers to care.
FAQ
What do successful Starter Story founders have in common?
The strongest pattern was not the product category. The founders picked a specific customer, found evidence that customer already cared, built a narrow product, and kept pushing the channel that produced real behavior.
How did these founders get their first customers?
The first customers usually came from manual work: creator outreach, Reddit posts, Product Hunt launches, personal networks, Instagram cold messages, niche influencers, Google search pages, and word of mouth inside closed communities.
Which customer channels appeared most often?
Consumer apps leaned on TikTok, Instagram, creators, influencers, and paid social. Software products leaned on Product Hunt, Reddit, X, Google search, and smaller related products. Closed communities grew through insiders and referrals.
Why I am doing this
I build AI systems for founder research, distribution, and market intelligence. This was a small example of the workflow I care about: take messy public information, preserve the source, extract the pattern, and turn it into something useful for deciding what to build or where to look next.
If you want to compare notes, I am @yorkartur on X.