← All distillations
channel·one-page markdown

Starter Story

Pat Walls

The repeated pattern across dozens of Starter Story founder interviews: how people with no capital and no audience got a business to its first real money, and the six questions they answer in flatly opposite ways.

Chat with this corpus

This is not one person's method. It is a library of founders explaining exactly how a specific business went from nothing to money, with the numbers attached, so the rules below are the pattern that survives across them rather than any single founder's opinion. What they converge on is that demand has to be proven with something cheaper than the finished product; where they contradict each other, which is on market choice, validation, audience, headcount, pricing and paid spend, the contradiction is written down in its own section rather than averaged into a recommendation nobody actually made.

Do

  • Sell it before you build it. Marc Lou spent a full year on a solo app with no business plan and scrapped it in twenty-four hours, then built a lead-generation tool for escape rooms only after a marketer friend told him to sell it first; a single cold call produced the first paying customer before any code existed. [Source: "The Underdog: From $10/hr to $1.5M/Year"]
  • Test the idea with one post before you test it with one product. Alex Lieberman validated a ghostwriting agency with a single tweet asking who would pay $5,000 to $10,000 a year for it, got twenty-five DMs, then personally connected the first clients to two ghostwriters over Slack. Validate in cheap incremental steps: a tweet, then a thread, then a newsletter, then a build. [Source: "The Multipreneur: He's Building A $10M Portfolio of Income Streams"]
  • Fulfil the service by hand before you write software for it. ListKit ran as a manual, form-plus-24-hour-turnaround service using an existing agency's processes, which produced real revenue and real feedback at zero development cost. The software relaunch then hit $1M ARR in 87 days, largely because the audience built during the manual phase was already waiting. [Source: "I Built A $1M SaaS In 87 Days"]
  • Spend days studying what already works in your niche before you make anything. Steven Cravotta scrolled and logged videos for seven straight days before posting a single PuffCount video, recording the hook, the storyline and the call to action of each top performer in a spreadsheet. The research is what told him which product to build, not just how to market it. [Source: "The $1M Mobile App Playbook"]
  • Build a painkiller, not a vitamin. Rob Hallam shipped five products over two and a half years that made a combined zero, and the lesson he drew was that distribution matters, monetization starts on day one, and the product has to solve a problem the buyer already knows they have. [Source: "How I Used Twitter to Hit $10K/Month"]
  • Narrow the offer until the buyer is unmistakable. Scott Millard went from video, to short video, to short video for online brands, to short video ads for funded B2B SaaS companies specifically, and the narrower offer converted faster on a smaller audience and let him raise prices, from a $500 first sale to packages at $6,000 to $12,000. [Source: "He Built A $600,000 One Person Business (with video editing)"]
  • Pick one channel and get obsessive about it. Miquel Palet took Late to $40K MRR in about seven months on high-intent search alone, running the same keyword set through in-depth organic guides and Google Search ads, and explicitly rejects being mediocre across ten channels. [Source: "Zero to $40K/Month With One Marketing Channel (No Social Media)"]
  • Answer your own support tickets for the first month. Thomas Frank staffed customer support himself roughly eight hours a day for the first month after launch specifically to find out firsthand where users got stuck, and names it as something every founder should do at least once. [Source: "I Made $2.5M Selling A Digital Product"]
  • Add the paid tier early, because payment is the only real validation. Florin Pop's playbook is to ship something small and useful in days, put it publicly where the audience already is, gather usage feedback, then attach a price and see whether anyone crosses the line. After a few weeks or months, decide to double down or drop it without guilt. [Source: "I Made $500K From 8 Different Income Streams"]

Don't

  • Don't build for months without a buyer. Connor Burd's first app, a social app, took six to eight months and failed; his second took under two weeks and reached $20,000 a month within fifty days. The failure taught him that utility apps scale for a solo builder in a way social apps do not, and that the build itself is rarely the constraint. [Source: "I vibe coded a $20K/month mobile app in 14 days"]
  • Don't count views as progress. One PuffCount video hit 11.6 million views and drove almost no installs, because it had no call to action and addressed no pain point. Track conversions per video rather than reach, and pin the highest-converting ones to the top of the profile. [Source: "The $1M Mobile App Playbook"]
  • Don't let one platform or one customer segment carry all the revenue. Marc Lou's escape-room tool reached about $4,000 a month in recurring revenue and went to zero in twenty-four hours when COVID shut down the physical businesses it served. Concentration is invisible until the day it is not. [Source: "The Underdog: From $10/hr to $1.5M/Year"]
  • Don't add features instead of doing distribution. Diego Roshardt shipped an MVP in about two weeks and started marketing immediately, reaching $17K MRR and more than a thousand paying customers in four months on $0 of marketing spend, and names prioritizing unrequested features over distribution as the mistake to avoid. [Source: "How I Used Reddit to Hit $17K MRR (With ZERO Audience)"]
  • Don't quit the income until the new thing has passed it. Dickie Bush ran a writing business from 4:30am around a sixty-hour trading job and only quit once the side income had doubled his salary. Karthik Shanadi worked an overnight Amazon shift and then a full day job while building his apparel business every night until midnight for about two and a half years. [Source: "He Quit His Job And Makes $10M/Year Writing Online"] [Source: "He Turned $500 Into $10M"]
  • Don't stay the only person who knows how the work gets done. Tim Stoddart names realizing he was personally the biggest bottleneck as his largest management shift, and the fix is documenting the process, teaching it, then getting out of the way. Ian Myers hit the same wall between five and eight employees, when all the institutional knowledge still lived in his head. [Source: "I Make $5M/Year With 3 Businesses"] [Source: "I Quit My Job & Accidentally Built A $10M Business"]
  • Don't start the second thing before the first one works. Alex Lieberman refuses to launch a new portfolio business until the previous one has reached product-market fit, measured concretely as at least half of current clients willing to recommend it and average retention around ten months. He names dilution as the single biggest risk of running several businesses. [Source: "The Multipreneur: He's Building A $10M Portfolio of Income Streams"]
  • Don't sell an asset into a decline, or to the highest bidder. Lotts Ezeike builds apps to roughly $10-20K MRR, shows three months of upward trend before listing, and expects around 2 to 4 times annual profit. He deliberately offers the deal to the third-highest bidder who can close in three or four days, because top-price deals tend to drag out and collapse. [Source: "I flipped 4 apps and made $500,000"]

Where they disagree

  • Should you enter a crowded market or an empty one? Guillaume Moubeche picked a red ocean on purpose after a blue-ocean personalization startup was killed by a single platform change, on the logic that competitors already taking money prove the demand exists; he then won lemlist on one dimension, personalized outreach tied to meetings booked, and runs it near $30M ARR. Thomas Smale did the opposite, building a brokerage for small online businesses precisely because no such firm existed, which handed him zero direct competition. What separates them is where the edge lives: Moubeche's was execution on a job buyers already pay for, Smale's was being the only person offering the service at all. Neither picked by default, and that is the part that transfers. [Source: "The Underdog: He Turned His Last $1,000 Into $150M"] [Source: "Zero to $1.1B from Flipping Websites"]
  • Does money have to change hands before you build? Marc Lou closed a paying escape-room customer by cold call before writing code, and treats that as the turning point after five years of failure. Against that, Connor Burd validates by reading TikTok and Instagram comments for people already asking how to solve the problem, skips extensive upfront research, and shipped Payout in under two weeks; Dawson Botsford built nfy in four or five hours during a hackathon and let one launch tweet do the testing, reaching 10,000 email signups in 48 hours. Marc Lou himself sits on both sides: ShipFast, his largest product, was built in under a week with no validation and no revenue expectation and made $500 in its first two hours. What decides it is the cost of being wrong. When the build is days, shipping is the cheapest available test; when it is months, presell or do not start. [Source: "The Underdog: From $10/hr to $1.5M/Year"] [Source: "I vibe coded a $20K/month mobile app in 14 days"] [Source: "I Built A $1M App In 5 Hours"]
  • Do you need an audience before the product? Andre Haykal Jr. credits ListKit's 87 days to $1M ARR to an audience kept warm through years of coaching, agency work and public tweets about the struggle, so buyers rushed in the moment it launched. Dickie Bush takes a third position: do not build your own audience and do not go without one, borrow somebody else's by ghostwriting, which he says produces $5,000 to $10,000 a month without needing tens of thousands of followers first. Against both, Diego Roshardt reached $17K MRR and over a thousand paying customers with under 100 followers and no other social presence, using value-first posts across a wide list of subreddits, and Miquel Palet built to $40K MRR with no social presence at all, arguing organic social reach is oversaturated while search intent still converts. The dividing line is whether your buyer has to trust you before purchasing or is already typing the problem into a search box. [Source: "I Built A $1M SaaS In 87 Days"] [Source: "He Quit His Job And Makes $10M/Year Writing Online"] [Source: "How I Used Reddit to Hit $17K MRR (With ZERO Audience)"] [Source: "Zero to $40K/Month With One Marketing Channel (No Social Media)"]
  • Stay solo, or build a team? Marc Lou's rules after being laid off were explicit: never raise money, never hire employees, never spend more than a short window on any one product, and he shipped six apps in seven months under those rules. Andre Haykal Jr. says the opposite outright, naming a five-person co-founder team with complementary skills (marketing, sales, technical execution, a team leader) as the single biggest driver of ListKit's success, on top of roughly forty contractors. Tim Stoddart states the constraint that actually decides it: an agency adds workload with every client and roughly every two to four clients forces another hire, while a product does not. [Source: "The Underdog: From $10/hr to $1.5M/Year"] [Source: "I Built A $1M SaaS In 87 Days"] [Source: "I Make $5M/Year With 3 Businesses"]
  • Free trial, money-back guarantee, or no way out at all? Andre Haykal Jr. skipped free trials entirely for a money-back guarantee, arguing trials mainly attract freebie seekers while a guarantee still takes the cash up front. Rob Hallam runs SuperX at $29 a month with roughly a 30% trial rate, and churn around 30% that he was actively working down. Umberto Mezzadra used neither: a strictly no-refund lifetime deal in three tiers, capped at five to seven days and a fixed number of spots, which took in about $120,000 in twenty-four hours from 500 to 600 buyers, and which he argues produced better bug reports than subscribers who can leave any time. All three worked and the corpus does not settle it. What it does show is that the mechanic follows the specific buyer, not the product category. [Source: "I Built A $1M SaaS In 87 Days"] [Source: "How I Used Twitter to Hit $10K/Month"] [Source: "My App Made $120K in 24 Hours"]
  • When do you turn paid spend on? Steven Cravotta's order is market research, then organic volume, then paid, and he says never the reverse; PuffCount only absorbed $82,000 of ad spend after organic content had already identified which creatives converted, taking it from $3K to $44K MRR. Miquel Palet launches Google Search ads in the same first week as his first ten to fifteen articles, spending about $8K a month against a $120 target CPA and getting roughly $2.50 back per dollar. The difference is what the money is buying. Palet intercepts people already searching for the thing, so an ad works on day one; Cravotta has to manufacture attention, where paid only amplifies a creative that already earned attention on its own. [Source: "The $1M Mobile App Playbook"] [Source: "Zero to $40K/Month With One Marketing Channel (No Social Media)"]

The one line

Find the cheapest possible proof that somebody will pay, then pick the single channel that reaches them and stay on it until the revenue is real.