ScreenshotOne — How an ex-engineer’s “boring” URL-to-image API compounded to ~$200K–$300K ARR in 3 years on build-in-public and SEO alone

Background & product

ScreenshotOne is a developer-facing screenshot API: pass a URL, get back an image (or PDF/HTML). It powers automated OG images, website thumbnails, monitoring/archiving, and sales-doc automation. Its users aren’t consumers but other developers and companies — it’s a “sell the picks and shovels” business.

It was built by Dmytro Krasun. After ~10 years as a server-side developer he went solo and launched on May 29, 2022 (the idea is logged in Jan 2022 as “Project 0002/1001,” the second entry in his self-set goal to build 1001 projects). Far from viral, he runs what he calls a “calm business” — bootstrapped and solo.

It was not smooth from the start. About a year after launch, in November 2022, MRR was just $141, and he wrote that he questioned his life choices. He kept shipping features and docs anyway, climbing $2.5K → $10K MRR. In April 2025 the official blog announced “$200K ARR, 400+ paying customers.” By August 2025 it was ~$21.6K MRR, with ~$300K ARR / 650 customers cited more recently (this upside is secondary).

Technically, the core value is absorbing the pain of running headless browsers at scale — reproducible rendering, killing ads/cookie banners, and so on. Pricing scales with usage: a free tier (100/mo), Basic $17 (2,000), Growth $79 (10,000), and a $259 (50,000) plan above that.

The repeatable playbook

1. 1If you can’t go viral, pick a structurally sturdy boring problem (a screenshot API — everyone builds it once and tires of maintaining it — low churn)
2. 2Compounding is flattest early: design to survive year one’s $141/mo (runway, expectations), and keep shipping features and docs without quitting
3. 3Use build-in-public as a discipline device (public revenue = deadline + trust + backlinks in dev circles; deliberately toggle posting on/off)
4. 4Harvest developers’ search intent with SEO (mass-produce per-language guides, comparisons, use-cases → funnel into official docs; zero ad spend)
5. 5Usage-based subscriptions (free tier → $17–$259) that favor adoption and expansion over ARPU, so a small paying base still compounds ARR
6. 6Solo means optimizing for sustainability (a calm business) over speed: no VC, no hires, designed to run by one person

Deep dive

【Deep dive】Most of the 40+ cases in this database are consumer apps that ignite on TikTok/Instagram. ScreenshotOne is the opposite — a “boring API sold to other developers” that compounded over three years on just two legs: build-in-public and SEO. It’s arguably the most reproducible archetype here for a working engineer. Let’s break down this “win by being boring.”

■ 1) Pick a boring problem that is structurally sturdy, not viral. A screenshot API is neither new nor sexy, and it has competitors. But Dmytro chose a structural need: “every developer builds this once, gets tired of maintaining it, and wants to outsource it.” That matters — demand doesn’t ride a trend, so churn is low and, once embedded, an API is hard to rip out (it’s buried inside the customer’s product). You can’t go viral, but you also can’t flame out. That contrasts with the fragility of many consumer apps here: when virality stops, they stall.

■ 2) Year one is flat compounding — “not giving up” is the real skill. Thirteen months in, November 2022, MRR was just $141, and he admits he questioned his life choices. Fittingly, his podcast episode is literally titled “The Art of Not Giving Up.” Compounding looks flattest at the start — backlinks, domain authority, and build-in-public trust don’t show up in the numbers early. That’s where most indie hackers quit. He didn’t; he kept shipping features and docs. Surviving year one was the precondition for the later curve to $10K → $20K+.

■ 3) Growth leg one: use build-in-public as a discipline device. Dmytro posts revenue and progress continuously on X and Threads. Build-in-public pays off three ways: (1) publishing is a deadline you set for yourself — discipline; (2) a person who shows real numbers earns trust, generating word-of-mouth and backlinks in developer circles; (3) prospective customers (other developers) are literally your audience. Tellingly, he once posted “I quit Twitter till I reach $5K+ MRR” — when broadcasting was stalling the work, he cut off posting to focus on the product. Build-in-public isn’t attention-seeking; it’s an on/off tool he designs deliberately.

■ 4) Growth leg two: harvest developers’ search intent with SEO. The other leg is technical SEO. Developers always search when they hit a problem — “take screenshot with Go,” “puppeteer screenshot API,” “website thumbnail API.” Dmytro wrote a large body of per-language guides (Go/Python/Node…), competitor comparisons, and use-case posts, capturing these long-tail, high-intent keywords one by one, funneling demand from his bylined technical articles into the official docs. This drives CAC toward zero with no ad spend — boring, but it compounds. It’s the same family as stagetimer’s (#82) doc-driven SEO, but where that targets one feature in the events industry, this goes after the exact search terms developers type while writing code.

■ 5) Solo, calm business: optimize for sustainability first. Dmytro took no VC, hired no one, and runs it alone — coding at midnight while raising a newborn (Medium, “Midnight Code”). Because “keep it calm and durable” is an explicit goal rather than hyper-growth, pricing favors easy adoption and headroom over ARPU: a free tier scaling into usage-based subscriptions ($17–$259/mo). A B2B API bills more as each customer uses it more, so even a small base (400–650 companies) supports $200K–$300K ARR. It’s slow compounding over three years — the inverse of an 18-month consumer rocket.

■ Takeaways for indie builders. The lessons are clean. (a) If you can’t chase virality, pick a structurally sturdy boring problem. (b) Compounding is flattest early, so design for surviving the $141 of year one (runway, expectations) up front. (c) Drive CAC to zero on two legs — build-in-public (discipline + trust + backlinks) and SEO (harvest search intent). (d) If you’re solo, optimize for sustainability over speed. More than the flashy English-language exits, this is a template most engineers can start copying today.