From a $250k exit to a two-week build
Bhanu Teja Pachipulusu belongs to a new wave of founders who treat software businesses like a craft. After leaving a conventional software engineering job in 2019, he started building small SaaS products solo. His first notable win was Feather.so, a tool that turned Notion pages into polished blogs, which grew to about $6,000 in monthly recurring revenue before he sold it in 2024 for $250,000.
That exit did two things: it validated the bootstrapped, product-first approach, and it funded the next experiment. In early 2023, Bhanu's timeline was full of AI. He wondered whether the technology could help his Feather customers — every one of them had a blog, so why not let visitors chat with it? He built a prototype in two intense weeks: scraping pipeline, bot training, chat UI, and an embeddable widget.
Then came the realization that changed everything. The feature was too big to be a feature. Anyone with a website — not just Feather bloggers — could use an AI chatbot trained on their own content. So he launched it as a standalone product called SiteGPT, with a paywall from day one and a demo bot trained on the SiteGPT website itself.
A viral launch, then the real work
SiteGPT's launch was the kind of viral success founders dream about. A single tweet took off, the product spent hours on the Hacker News front page the next day, and then it claimed the top spot on Product Hunt. Within a month, SiteGPT was doing over $10,000 in monthly recurring revenue — Bhanu's year-end goal, hit in weeks.
The Starter Story interview he gave afterward is strikingly honest about the mechanics. There was no growth hack and no ad budget. The product was simple, the demo was instantly legible — chat with any website — and the timing was perfect: every website owner in 2023 was asking what AI could do for them. SiteGPT answered with a working product instead of a pitch deck.
The first two weeks after launch were brutal in the way all viral launches are: a flurry of bugs, constant user engagement, daily firefighting. Bhanu did it all solo. But the revenue curve justified the pain, climbing toward $15,000 MRR within months and eventually past $300,000 in annual recurring revenue by mid-2024, according to the Latka SaaS database.
What made the launch convert — not just go viral — was the paywall. Bhanu never offered a free plan, so every signup was a buying decision. The demo bot on the homepage did the selling: visitors could literally chat with the product before paying, which collapsed the trust gap that kills most new SaaS launches.
50 free tools as a marketing team
The most instructive part of the SiteGPT story is what happened after the launch spike faded. Bhanu did not become a content marketer or buy ads. Instead, he kept building — dozens of small, free, search-focused tools: PDF converters, chatbot name generators, sitemap validators, FAQ generators. Each one targeted a real keyword with low competition, and each one carried a call to action back to SiteGPT.
The numbers behind this engineering-as-marketing machine are remarkable. SiteGPT now gets around 50,000 monthly visitors, and roughly 90% of its Google traffic comes from those free tools. The funnel converts methodically: 50,000 visitors become about 200 leads, 60 trials, and with 25–40% trial conversion at roughly $100 average customer value, the math lands at five-figure MRR — all organic, all free.
Bhanu has said that with his existing tools as templates, spinning up a new free tool takes less than five minutes using AI coding assistants. The marginal cost of each new traffic asset approaches zero, which means the strategy compounds indefinitely. It is product-led growth taken literally: the product portfolio is the marketing department.
The quiet economics of a solo SaaS
Zoom out and SiteGPT is a case study in profitable indie SaaS. Total revenue crossed $500,000 across roughly two years of operation. Monthly expenses run only $4,000 to $5,000 — including about $1,000 a month in OpenAI API costs — against five-figure monthly revenue. Zero employees, zero external funding, zero marketing budget.
The Latka database valued the bootstrapped business at around $900,000, which arguably understates a profitable, growing asset with diversified organic traffic. But valuation was never the point. Bhanu's model is to build cash-flowing software businesses, sell or hold them, and fund the next one — the same loop that turned Feather.so into SiteGPT.
For founders, the uncomfortable lesson is that SiteGPT's success looks like luck — a viral tweet, perfect AI timing — but was actually the third or fourth iteration of a repeatable system: build small, launch with a paywall, distribute through product, keep costs near zero. The tweet went viral because the product was ready when the wave arrived. Most founders are still waiting for the wave with nothing built.
The discipline extends to the portfolio level. Bhanu treats each product as an experiment with a clear success metric, and the $250k Feather.so exit proved the model: build to profitability, hold or sell, redeploy the capital and the lessons into the next one. It is indie hacking as an asset class.
The playbook
- Let exits fund experiments — Sell the first product (Feather.so, $250k exit) to fund the second — every win buys the freedom to take the next swing.
- Spin out the feature with the bigger market — SiteGPT started as a weekend feature for Feather's blogging customers; Bhanu spun it out when he saw the market was everyone with a website.
- Build free tools as the marketing team — 50+ free tools targeting low-competition keywords drive 50k monthly visitors, 90% of Google traffic — product work as distribution.
- Launch loud, charge from day one — Launch with a paywall and a demo bot trained on your own site — one viral tweet, a Hacker News front page, and Product Hunt #1 followed.
- Stay solo and keep the burn near zero — $4–5k monthly expenses against five-figure MRR, zero employees, zero funding — profitability is the growth strategy.