Elston launched Tiny Host in 2019 as a barebones side project — there was no login, no paywall, just a WeTransfer-style widget. His first users came entirely from Reddit, where he posted honestly about what he built, asked for feedback, and was not spammy. He described it as: "Hey, I created this app, I think it's useful, what do you guys think?" He also posted on Slack communities and Twitter. This got him early feedback and helped him understand what features people wanted. From that early feedback loop, he identified features people were willing to pay for — most notably custom domains — and placed them behind a paywall. He did not build the full back-end for some features initially (e.g., password protection was faked manually at first), only committing engineering time once demand was confirmed. This lightweight, user-driven approach validated the product before he invested heavily in it.
Tiny Host
Simple drag-and-drop web hosting for non-technical users to share files online
8 moves, in order
- Pre launch / Launch (2019)Reddit community posts
Posted honestly on Reddit (and Slack/Twitter) saying 'I built this, what do you think?' — not spamming, just asking for feedback. Used responses to identify which features people valued enough to pay for.
First users and product validation; identified custom domains as a paid feature - Early monetization (6 7 months post launch)Product Hunt launch
Launched on Product Hunt roughly 6-7 months after the initial release. Iterated on pricing downward (from ~$18) until the first customer converted at ~$12.
First paying customers; first Stripe SaaS revenue - First 6 months post monetizationYoutube tutorials
Created faceless tutorial videos targeting how-to queries (e.g., 'how to upload a PDF online', 'how to upload a React app'). Focused on high-quality production vs. the low-quality competition. Videos were evergreen and took up to a year to rank organically.
Some videos hit 100,000 views; YouTube became an early organic traffic driver and trust-builder - Growth phase (side project era, ~2019–2021)Seo content and landing pages
Used Ahrefs/Semrush to find low-difficulty, high-volume keywords around specific use cases (e.g., 'upload HTML file', 'share PDF as link', 'restaurant menu QR code'). Built a combination of bottom-of-funnel landing pages and top-of-funnel blog posts for each use case. Focused on technical SEO (page speed, Core Web Vitals, Google Lighthouse compliance) alongside content.
70,000+ users/month on autopilot; eventually 100,000+ monthly organic visitors - Scaling (quit job at ~$8K MRR, ~2.5 years in)User driven product development
Observed that many users were uploading PDFs converted to HTML. Built native PDF upload support in a weekend. New use case unlocked a new customer segment and SEO content angle (PDF hosting), which took off 3-4 months later.
PDF hosting became one of the top use cases; unlocked a new audience segment - Mid growth (post $8K MRR)Design overhaul conversion
Hired a professional designer and did a major landing page/product redesign to move away from the 'indie project' look and build trust with non-technical users.
Increased conversion confidence; positioned product more credibly against established tools - Scaling SEO (team build out)Seo team scaled
Hired dedicated SEO roles: a link-builder, a keyword research/SEO analyst, and a content writer. Made SEO a structured team function rather than a solo founder activity, cementing it as the core acquisition foundation.
Sustained 100,000+ monthly organic visitors; SEO remained top traffic source even amid ChatGPT disruption - Vibe coding / AI trend (2024–present)Ai tool ecosystem positioning
Noticed a surge of non-technical users generating HTML via Claude/GPT who needed the simplest way to host it. Repositioned Tiny as part of the 'DIY stack' for vibe coders. A customer found Tiny through Claude directly. Leaned into this new audience via SEO content and product messaging without building new core features.
Accelerated growth; monthly growth rate jumped from 3-4% to 20-30%; reached $1M ARRMRR $83k
Software engineering background meant he could build and iterate the product solo with no technical co-founder needed. He was also embedded in the Indie London / Indie Hackers community, giving him access to peers slightly ahead of him who had figured out SEO and distribution channels.
SEO (organic search) — driving 70,000–100,000+ visitors/month and described as the foundational traffic channel that compounded over years
No explicit channel failures mentioned, but he implicitly discarded social/viral tactics as not core to his strategy. Earlier startups failed due to: spending 6 months building before validating, having too many co-founders, chasing YC/investors, and rigidly holding onto a product vision instead of adapting to market feedback. Multi-product / portfolio approach also explicitly called out as not a path to $1M ARR.
I Make $1M/Year Hosting PDFs on the Internet
Zeronel (Deal Neil / Zenevy)
Social media API aggregator letting SaaS apps add social features without direct platform integrations
ShipFast
Next.js boilerplate codebase for indie developers to ship startups faster
Late
Social media API that lets developers post/automate across platforms via one unified API