Field Note / day-32-screenshotone
From $7 to $20K MRR: How Dmytro Krasun Turned a “Boring” Screenshot API into a Compounding Solo-Founder Engine
ScreenshotOne is a developer-first API that programmatically converts any URL, HTML, or Markdown into a clean,...
Answer Engine Brief
This case study is part of Jesse's 100-day founder marathon for Solo Unicorn Club: stories of solo or near-solo founders who reached meaningful revenue gravity and left reusable lessons about product, distribution, AI leverage, and one-person company design.
Real Reason to Study This Business
ScreenshotOne is a developer-first API that programmatically converts any URL, HTML, or Markdown into a clean, banner-free screenshot. Its core users are engineers who would rather pay for reliable rendering than wrestle with headless-browser edge cases. What makes ScreenshotOne worth dissecting isn’t a splashy valuation; it’s how one engineer, with no outside capital and competing against larger teams, created an asset that now throws off roughly $240K ARR and scales almost entirely on code and search traffic. For solo AI-native builders, the playbook shows how to:
- Exploit “infrastructure jobs” big firms ignore.
- Compound SEO through tiny, free side-tools instead of content farms.
- Trade headcount for serverless primitives—then buy back time with ruthless automation.
YouTube homepage showing trending categories and featured videos including a music video from Madeline the Person and trending content like BTS and Apex Legends.
What the Founder Did Differently
- Started with an unglamorous pain – Developers hate screenshot drift, cookie banners, and flaky Puppeteer scripts. Solving a “maintenance tax” problem produced high urgency and low price sensitivity.
- Constraint-driven stack – From day one Krasun chose Cloudflare Workers as a gateway and Google Cloud Run for autoscaling so he could ignore ops and focus on feature velocity.
- Built in public, but with purpose – Daily Twitter/X threads exposed road-mapping, revenue, and failures, creating social proof plus free backlinks that quietly juiced domain authority.
- Delayed everything that doesn’t compound – No mobile app, no open-ended consulting, no “nice-to-have” integrations until core API reliability hit six-nines.
Dmytro Krasun, image source.
The Growth Flywheel: Step-by-Step
| Stage | Strategic Move | Irreversible Gain |
|---|---|---|
| 0 → $7 MRR (Jul 2022) | Launch bare-bones “/take-screenshot” endpoint on Indie Hackers + Reddit drops | First pay-wall validation; Stripe history for social proof |
| $7 → $2.2K MRR (Feb 2023) | Added dark-mode, ad-blocking & device emulation—features SEO keywords “clean screenshot” & “retina” rank for | Organic leads > paid ads; support tickets reveal ICP |
| $2.2K → $5.6K MRR (Mar 2024) | Free micro-tools (Open Graph checker, PDF-to-PNG) earn backlinks; raises pricing for new cohorts | DA 40+, higher ARPU without churn spike |
| $5.6K → $13.6K MRR (Jan 2025) | Community SDKs + no-code n8n/Make connectors shift support to self-serve | Requests triple while support time stays flat |
| $13.6K → $20K MRR (Jun–Jul 2025) | Iterative pricing: paid overage (“charging extra”) auto-bills power users | Revenue scales faster than signup growth |
ScreenshotOne’s MRR climbed steadily from $7 in July 2022 to $20K by mid-2025.
Strategic Leverage & Business Model
1. Code Leverage – One containerized renderer handles hundreds of screenshots per minute; Cloud Run scales instances on-demand, so compute costs remain <$400/mo even at 28 M API calls. 2. Marketing Leverage – 70% of new customers originate from high-intent Google queries like “screenshot API” thanks to 200+ backlinks earned by free utilities and open documentation. 3. Pricing Mechanics – Three self-serve tiers ($17 / $79 / $259) plus metered overage at $0.004 per shot create negative churn as successful customers grow usage. 4. What He Didn’t Do
- No VC, no payroll.
- No long-form blog grind; SEO flywheel is product-led.
- No enterprise custom work—API only.
Can You Replicate This in 2025?
Easier Today
- GPT-4o can scaffold a TypeScript + Playwright renderer and write docs in hours.
- Vision models auto-detect bad screenshots, shrinking manual QA.
- Zapier/N8N connectors are now drag-and-drop; no bespoke SDKs required. Still Hard
- Six-nines uptime isn’t trivial; you must budget for observability tooling and on-call alerts.
- Winning “screenshot API” SERP now requires differentiated long-tail niches (e.g., VR page capture or AI-summary endpoints).
- Browser vendors change breakpoints monthly—someone has to chase those regressions. If Starting from Scratch
- Run Chromium on AWS Lambda SnapStart to skip cold starts.
- Ship usage-based pricing first; annual plans later.
- Bundle AI post-processing (text extraction, color-palette analysis) at higher ARPU.
Takeaways: Think Like Dmytro Krasun
- Pick revenue-critical grunt work—painkillers outlast vitamins.
- SEO beats social when you own an unambiguous keyword. Start building backlinks before shipping every feature.
- Serverless ≠ buzzword; it’s your fractional DevOps hire.
- Ship one killer option, then layer “nice” toggles that map to new keywords.
- Automate support before marketing. Every saved hour can be reinvested in features that resell themselves.
- Raise prices when rankings rise; trust the data, not imposter syndrome. Part of the 100 Days, 100 Solo Startups series.