Loading page…
A scraped directory of TikTok influencers, built when no tool existed for the platform — sold for $55K cash seven months after launch, on purpose
MRR grew about $500/mo consistently · LinkedIn outreach alone added roughly $1K MRR at a 5% paid conversion rate · ~1,000 organic visitors a day by the time he sold
Every buyer emailed the same three questions, so he answered them once in a Google Doc and sent everyone the same link — and because Google Docs shows you who else is reading right now, the bidders could watch “anonymous badger” circling the same deal.
Fewer bars = easier, cheaper, or faster for an AI-assisted solo builder. Editorial judgments based on the case details.
Andy Cloke started working as a developer in 2017 and built side projects alongside contract work. The first was a football quiz site — build a better Sporcle, figure out money and marketing later. People used it and did not come back. His entire marketing plan was spamming subreddits, and as he puts it, once you get banned from all the subreddits the marketing plan is over. It carried banner ads that took up most of the interface and nobody clicked. It made nothing. Then in November 2019 he went looking for a project the deliberate way: through Exploding Topics, a site that surfaces trends on the way up. He found nano influencers — people with one to ten thousand followers — and while researching them noticed the much bigger trend underneath, TikTok. This was pre-lockdown, before TikTok was a household word. There were plenty of influencer tools for Instagram, because Instagram had run a friendly API for years. For TikTok there was nothing. He spent about a month working out how to scrape TikTok and a couple of weeks building a front end over the database, with two phones propped up in his bedroom playing TikTok all day — his brother christened them Tick and Tock. Influence Grid launched in December 2019 as a searchable directory of TikTok influencers for marketing agencies, priced by how many influencers you could access. MRR grew about $500 a month, pretty consistently, through cold email, then LinkedIn outreach, then programmatic SEO. By mid-2020, seven months after launch, he sold it to an American company for $55,000 in cash. He sold not because it was failing but because the whole thing sat on a scrape, and TikTok could switch it off any day it liked.
He emailed the journalists.
The quick-reference card — acquisition channels, replication playbook, risks & traps, and our verdict — is members-only.
Join the Unicorn Club — $5/moCancel anytime, in one clickPaste into Claude Code / Codex and get a working version of this product end-to-end.
Two production-grade prompts per idea — a full build spec and an SEO growth plan — plus 10 tailored build specs a week. Members only.
Join the Unicorn Club — $5/moCancel anytime, in one clickEvery claim traces back to a listed source — open any of them and check the number yourself. The exact words are quoted below, with a timestamp that jumps to the second each figure is said.
“…the past his most recent one influence grid was sold for 55k back in mid 2020 having only started it seven months before throughout this episode we talk about his…”
0:23 · hear it in the videoData credibility: Podcast interview with the founder on Indie Bites; every figure is self-reported and unaudited. The $55,000 sale price, the seven-month timeline, the roughly $500/mo MRR growth, the 5% LinkedIn conversion rate, the additional ~$1K MRR from that channel and the ~1,000 organic visitors a day at sale all come from Andy Cloke alone, with no dashboard shown or third-party data cited. He is explicitly approximate in several places (“another grand or something in MRR”, “a couple of those would convert”). The episode is titled “Bootstrapping two $3k MRR projects, selling one for $55k”, which implies Influence Grid was around $3K MRR at exit, but he never states an exit MRR in the conversation itself, so we have not carried that number into the revenue line. The valuation guidance (three to five times annual revenue for a non-strategic buyer) is secondhand from Flippa and FE International as he recalls it. The same episode also covers Data Fetcher, his next business, which is already in the index under a separate entry.
The pricing was set before launch and it was the obvious one for a database product: tiered plans based on how many influencers you could access.
Two channels stacked, and only one of them compounded. The first was LinkedIn, mined out of his own database: influencer bios list an agency email, the email's domain leads to a website, and those websites turn out to be small influencer marketing and talent agencies.
There were two flat stretches and he handles them differently.
A searchable directory of TikTok influencers aimed at marketing agencies, sitting on top of a database he built by scraping TikTok himself. The technical core was the scrape: about a month of work figuring out how to extract creator data from a platform with no friendly public API, and physically two phones in his bedroom running the app on an endless feed all day to keep the pipeline fed. Everything the customer touched — the search, the filters, the tiered access — was a front end over that database, built in a couple of weeks. Our read: the entire product thesis is a timing arbitrage, and Andy states the rule he was following explicitly. Find a platform growing massively, then look at a similar platform that is a bit older and more mature and see what tools exist there. Instagram had a rich ecosystem of influencer discovery tools precisely because its API had been open for years. TikTok had nothing, not because nobody wanted the tools but because the data was hard to get. The month he spent on the scrape was not incidental work around the product — it was the product, and it was the only barrier keeping competitors out. That is also exactly why he sold: a moat made of scraping is a moat the platform can drain whenever it notices.
Subscription plans tiered by how many influencers a customer could access. MRR grew about $500 a month, consistently rather than in spikes, from a mix of the one blog article, Product Hunt, LinkedIn outreach at 5% paid conversion, and eventually organic search at around a thousand visitors a day converting a couple of them. Andy's own post-mortem on the business model is sharper than anything about the growth. Influence Grid had no expansion revenue. Customers did not upgrade much, because there was no usage metric that rose with value — nothing a customer did more of that made a bigger plan feel necessary. He contrasts this directly with his next product, where the billable unit is API requests, a number that naturally grows with a customer's usage and therefore drives expansion revenue and offsets churn. Our read: this is the single most portable lesson in the episode and it is worth applying before you build. A directory priced by access count is a flat business — every customer lands on a tier and stays there until they cancel, so all growth has to come from new logos. Pick a billing unit that grows on its own, or accept that you are running a treadmill where the marketing never gets easier.
He checked the marketplaces first and rejected them all for concrete reasons. Flippa wanted its 10 or 20% even if the sale closed outside the platform during the listing period. FE International wanted a full year of revenue history before it would list him, and he did not have one. He did list on MicroAcquire and got no serious offers. What worked was posting it himself — LinkedIn, Twitter, Indie Hackers, Hacker News — a short note saying he was selling, here is the MRR, here is my email. Then the trick worth stealing: every enquiry asked the same three questions, so he wrote a full brain dump into a Google Doc, covering tech stack, how he had grown it, and why he was selling, and sent everyone the same link. Because he had been open about his metrics all along, the doc cost him nothing in disclosure. And because Google Docs shows concurrent viewers, each buyer could see anonymous animals reading the same document at the same time, which manufactured competitive pressure he never had to mention. On valuation he was realistic: brokers told him a non-strategic multiple is typically three to five times annual revenue, people online told him he should get $250,000, and he privately expected $40,000 to $100,000. Most offers were 10 to 20% up front plus a revenue share the following year, which defeats the entire purpose of selling to de-risk, so he took a lower all-cash number instead. The buyer was an American company already running a similar tool for Instagram, found through Twitter or LinkedIn. No broker, no escrow, straight bank transfer, $55,000.
Copy the idea-sourcing method, which is mechanical and repeatable: find a platform growing fast, find an older platform in the same shape, and build the tool that already exists over there and not yet over here. Copy the launch outreach — going to the journalists who already wrote the thesis article is the cheapest possible qualified audience, and asking for feedback rather than coverage is why it worked. Copy the mining of your own dataset for prospects; his customers were literally extractable from the database he was selling access to, which is a pattern that applies to almost any B2B directory. Copy the programmatic SEO discipline of shipping it early precisely because it takes three or four months to do anything. And copy the sale process wholesale: the public post, the one shared Google Doc, the all-cash preference. Change the billing unit, which is the change he makes himself on the very next project — find a usage metric that grows with the customer instead of an access tier that does not. Our read: also change your honesty about the timeline. Seven months to $55,000 sounds like a rocket, and it was preceded by a failed football quiz site, two years of contract work, and a month spent solving a scraping problem most people would have abandoned. Reuse the method, not the calendar.
The founder already named the fatal one and acted on it. A business built on scraping a hostile platform can be switched off without warning, and the risk gets worse as you succeed: the bigger the directory, the more likely TikTok notices and blocks it. Andy says explicitly that it always felt precarious and that this is why he started looking for a buyer. Alongside the technical risk sits a legal one he does not discuss at all — scraping creator data at scale has since attracted far more attention from platforms and regulators than it did in 2019, and anyone copying this in 2026 inherits that exposure rather than the 2019 version of it. Then there is the category clock. The window here was a platform that had exploded before its tooling ecosystem caught up, and that window closed on TikTok specifically within roughly a year; the buyer was already running the Instagram equivalent, which tells you how fast consolidation arrives. The business model risk is the flat one described above: no expansion revenue means permanent new-logo dependency, and the LinkedIn channel proved that finite audiences run out — he exhausted every agency on the platform and the channel simply stopped. Our read: the durable version of this is to treat the directory as a deliberately time-boxed asset. Andy did. He built it knowing what it was, grew it for seven months, and sold it to someone with a strategic reason to absorb the risk. Building this expecting to hold it for five years is the actual mistake.
The four founder playbooks — first customers, first dollar, the flywheel, the silent stretch — are in the Unicorn Club, on this idea and on all 1,435 others. The story, the numbers, the sources and every later section stay open to everyone.
Join the Unicorn Club — $5/moCancel anytime, in one click