Side-hustling for three years before going all-in and hitting $6.4k MRR

Work smarter every single day.

Why spend hours figuring out the perfect prompt when professionals have already tested what works?

Unlock 1,000+ ready-to-use Claude prompts for business, marketing, writing, research, and productivity, delivered free when you join Superhuman AI.

The clean version of an indie hacker story usually skips the first five attempts. Jonathan Geiger's story is mostly the first five attempts — and that is what makes it useful.

Jonathan has been a software developer at a small startup for nearly five years, and for three of those he built side projects on nights and weekends. The scorecard before his current products is a realistic picture of what building on the side actually looks like: he sold his first product, LectureKit, for $7k. WaitListKit got a single pre-sale, which he refunded because he decided he did not want to build it. He made $400 on NextUpKit with no marketing, then made it free and open source. And he grew CaptureKit to $127 MRR and sold it for $15k just two and a half months after launch.

This month, he finally left his full-time job to go all in. His focus is products five and six: SocialKit, a social media scraping API, and PostPeer, a unified social media posting API. Together they generate about $6.4k in monthly revenue.

SocialKit launched 13 months ago and grew to around $3.5k a month with 110 active subscriptions. PostPeer, his newest and fastest-growing product, launched in April 2026 and is already near $2.9k a month with 1,000 users.

Here is the framework underneath all of it — and it is one of the most systematic we have run.

Jonathan's early products failed in a way many builders will recognize: he built whatever he personally thought was cool, and it showed. The turning point was inverting that instinct entirely. He stopped trying to be original and started actively looking for competition.

His rule now is specific: at least two or three solid competitors, each doing roughly $20k to $80k per month, in a niche he understands. If there is no competition, he does not build — because in his experience, an empty field usually means there is no real demand, not that he found untouched territory.

That rule led him to CaptureKit, his biggest sale, and to his niche: API products in the scraping, data, and automation space. The logic is that developers and automation people pay for reliable APIs, and as a developer himself, he knows exactly what the buyer wants. When an idea passes the competition test and he understands the customer, he applies one more filter: can he differentiate slightly without reinventing anything? He does not try to out-feature anyone. With SocialKit and PostPeer, the wedge is speed of support and early adoption of the AI-agent use case. Small, real differences beat original ideas.

Financially, Jonathan spent almost nothing to build these products — domains, servers, and some AI subscriptions. The real cost was time: nights and weekends for three years alongside a full-time job.

He made that sustainable with a specific discipline: ship small, and market during development instead of after. By the time each product was "done," blog posts already ranked, free tools already brought traffic, and early users had already given feedback. The launch was not a starting line — it was a checkpoint on work that had been compounding for months.

His advice on this point is the emotional core of the whole story: build while you work if you can. It is slower, but it lets you take the leap from solid ground, so you never have to build from a place of desperation. He did not quit until his products were making more than $3k a month.

One of Jonathan's most valuable habits is treating each product as infrastructure for the next. When he built SocialKit, he reused CaptureKit's infrastructure wholesale — auth, billing, API key management, docs structure, even the landing page patterns. This compounding is now central to his approach: every product makes the next one cheaper to build.

PostPeer took longer, because social platform integrations are genuinely painful — TikTok's approval process alone is a running joke among developers who have been through it. It is also the first product he is not building alone; he works on it with a partner, which changes the pace. The partner splits platform integrations with him and keeps shipping when Jonathan is deep in SocialKit.

The stack is TypeScript everywhere:

Two candid notes from his experience. On payments, SocialKit runs on one processor he openly dislikes and PostPeer on another he currently likes — a reminder that tooling opinions are earned, not theoretical. And hosting has been a moving target: he started on Vercel for convenience, moved to Railway when API traffic made Vercel expensive, and is now considering self-hosting on a VPS as Railway's reliability wavers. The lesson is not which host to pick — it is that the right answer changes as you scale, and staying flexible matters more than committing early.

Both products use the same model: subscription tiers based on API usage credits, plus one-time credit packs for people who prefer not to subscribe. That one-time option matters more than Jonathan expected — it adds a consistent $500 to $700 a month per product, and developers who dislike subscriptions often start with a pack and later convert to a plan.

New users get 20 free credits to test the API with no credit card required. That is the entire free offering; after that, it is paid, and every product has paid plans from launch day. The first paying customer usually arrives embarrassingly early — $13 MRR in month one for SocialKit, $34 for PostPeer — and Jonathan argues those early payments teach you more about your pricing than any research could.

One small change that paid for itself: placing customer reviews at the top of the pricing page increased sales. His pricing advice distills to three moves — price against your competitors, since they already validated what the market pays; offer paid plans from day one; and always give people a way to pay without a subscription.

Jonathan's launches are deliberately anticlimactic. PostPeer's first customer arrived a week after a quiet launch, before any real announcement, because the SEO groundwork was already live. The classic launch platforms did little for him — SocialKit hit number one on Uneed and he used Product Hunt, but neither moved the needle. His conclusion is a principle worth adopting: compounding channels beat spiking channels.

SEO is the center of his system, and he starts it before and during product development, never after. The system is concrete: one or two relevant blog posts a week, a landing page for every API and feature, use-case pages, and competitor and alternative pages. His guiding trick is to watch search trends and search as his customers would, then create content for every relevant query — including YouTube videos targeting those same queries, which rank in both YouTube and Google.

He is emphatic about not skipping the basics: Google Search Console from day one, submitting your sitemap, internal linking, and some directory submissions for early backlinks. And he multiplies every piece of content by reshaping it. One YouTube video becomes a blog post with the video embedded, then a LinkedIn post, a Reddit post, shorts, and TikToks — the same content in different shapes reaching different audiences. Free tools do similar compounding work: a YouTube transcript extractor, for instance, ranks, brings steady traffic, and converts some of it.

A newer wrinkle: ChatGPT and Perplexity surface his competitor and alternative pages, and paying customers have found him through those LLM recommendations. The content he built for search engines is now being read back to prospects by AI assistants — the same asset, working a channel that did not exist when he started.

Beyond SEO, Jonathan builds in public, sharing real numbers on Reddit and LinkedIn. One recent post drew 27,000 views and a wave of signups — and posts like it have generated acquisition offers for his previous products. The numbers he was most nervous to publish turned out to bring both customers and buyers.

But the retention mechanism is simpler and more human than any growth tactic. He answers support in minutes and talks to users directly, even over WhatsApp. This shows up in nearly every review, and he names it as the one thing bigger competitors cannot copy. A large company cannot offer a personal, minutes-fast reply from the person who built the product. A solo developer can, and it becomes a genuine moat.

He is also honest about a piece of timing outside his control: the AI-agent wave arrived exactly when he had two APIs ready to plug into it. MCP did not exist when he started building, and now agents contribute a real share of his signups. He did not engineer that luck, but he was positioned to catch it — which is its own kind of skill.

Jonathan is generous about where his playbook came from. YouTube, he says, was his business school — he developed most of his approach (competition as validation, marketing before product, charging early) by absorbing free content while building at night.

The resources he credits directly:

He also credits Daniel Sternlicht, the CEO of the startup where he worked, as a significant influence — nearly five years working alongside him amounted to a masterclass in building products and shipping fast, much of which Jonathan applied to his own work. It is a good reminder that a day job, chosen well, can be part of the education rather than the thing standing in its way.

Jonathan is honest that he is not entirely sure what comes next. In the short term, he wants to grow PostPeer and SocialKit to $10k a month each — or sell one for a good deal. Beyond that, his goals are refreshingly plain: be happy, build things he enjoys building, and keep free time for the rest of life. And, of course, the dream still standing at the end of the list: to sell a product someday for over $1,000,000.

You can follow along on X, LinkedIn, and Reddit, or look at the products directly at postpeer.dev and socialkit.dev.

Claude can be your analyst, editor, and strategist.

But most professionals are using it to fix grammar.

These 1,000+ Claude prompts take it from grammar tool to your most powerful AI work assistant.

Sign up for Superhuman AI and get:

Claim your free prompts