I spent 19 days marketing a free Android camera app with a $0 ad budget — here's the exact organic cadence that 2.5x'd my reads

I build Background Camera RemoteStream — a free, local-only Android app that turns an old phone into a screen-off security camera with no signup and no cloud. The product is the easy part. The hard part, for a solo dev, is getting anyone to read about it without paying for ads.

So I ran an experiment in public: pure organic distribution, zero ad spend, one writer (me), and a strict content cadence. Over 19 days the catalog went from 253 article reads to 644 — about +154% — and in the final week followers stepped up on three platforms at once (Bluesky 3→9, dev.to 5→10, LinkedIn 5→11).

This is the honest write-up: what worked, what got falsified, and the one mistake that nearly cost me an account. No "10x growth hacks" — just the cadence and the data.

The cadence that actually moved the numbers

The core loop is boring on purpose. Boring is repeatable.

1. Answer a real question first. Every topic starts as an answer to a specific question people actually ask ("can I use an old phone as a security camera without a subscription?"). If I can't write a genuinely useful answer, there's no article.
2. Let it breathe, then write the canonical. Seven days after the short-form answer, I publish the long-form canonical article on dev.to — the version with the full argument, the cost tables, the architecture diagram in prose. The 7-day gap (I call it T+7d) means the canonical lands on a topic that already has a warm audience instead of a cold start.
3. Cross-post, don't copy-paste. Each platform gets a natively rewritten version — never the same text. The X thread leads with the surprising claim; the Bluesky posts lead with the threat-model; LinkedIn leads with the one concrete metric.

The single biggest lever was step 2. When I tested publishing the canonical first (before warming the topic), it died — 0 reads at day one. When I warmed the topic first, the same archetype carried. That's the whole game: sequencing beats volume.

What got falsified (the part most write-ups skip)

I pre-registered a hypothesis: that leading with LinkedIn (a "LinkedIn-first" sequence) would give long-form pieces a better runway. I wrote down the success criterion before I ran it.

It failed. The LinkedIn-first canonical read 0 views at day one against the criterion, and the supporting data all pointed the same way (8 impressions on the native LinkedIn post vs. 34 on the previous cross-post). So I retired the idea instead of moving the goalposts.

If you're doing build-in-public, write the kill-criterion down first. It's the only thing that stops you from narrating noise as a win.

The mistake that nearly cost me an account

I was putting a Play Store link in nearly 100% of my answers on a Q&A platform. It worked right up until it didn't: a moderator flagged one answer as spam and deleted it. First strike. A second strike likely takes the whole account — and that account had 60+ answers of accumulated equity.

The fix wasn't "stop linking." It was link discipline: include the link only where the question genuinely asks for an app, keep it to one outbound link, and lead with value the reader gets even if they never click. Distribution that gets you banned isn't distribution.

The counter-intuitive winner: impressions came from format, not follower count

The single highest-reach post of the whole run wasn't a carefully written article — it was a short, native-format post that happened to ride a distribution mechanism the platform favored. It out-reached my best native post by 13x with the same tiny follower count.

The lesson I took: on most platforms, format and mechanism dominate follower count at small scale. You don't need an audience first. You need the post shaped the way the platform wants to distribute it.

What I'd tell another solo dev

• Sequence beats volume. Warm the topic, then publish the canonical. T+7d beat everything.
• Pre-register your kill-criteria. Falsify your own ideas on schedule or you'll fool yourself.
• Link discipline is survival. One platform strike is a warning; two is a funeral. Lead with value.
• Honest numbers travel. The post that got my first-ever comment was the one where I just showed the real trajectory — 253 to 644 — with no spin.

If you want to see the product the whole experiment was about, it's here: Background Camera RemoteStream on Google Play. And if you're weighing whether your own camera app's "free" tier is quietly shrinking, that's the companion read: Did My Wyze, Arlo, or Eufy Plan Just Get Worse in 2026?

Building in public as Super Funicular LLC. Everything above is real data from a single solo run — including the parts that didn't work.