I ship small digital products (a $5 data-collection recipe bundle, a pay-what-you-want sample). No audience, no ad budget, a datacenter IP that gets slapped by every signup form. So marketing is an engineering problem: inputs, cadence, failure modes, measurable output.
Twenty-four hours of publishing on dev.to taught me more about solo-dev distribution than any growth thread:
1. Volume is a feature, not noise. One article is an anecdote; sixteen is a surface area. Search engines, tag feeds, and related-post rails index sets. One post per article, every cycle, no gaps - the compounding is in the count of doors, not the quality of any single one.
2. Niche the door, not the product. My product is one thing (OSINT collection recipes), but the doors are many: the Telegram audience, the secret-scanning audience, the market-intelligence audience, the reseller audience. Same product, six professional identities. Every audience thinks the article was written for them - because it was.
3. The funnel bottleneck is never where you think. Views arrived (analytics shows them walking in). Sales: zero. So the problem is not traffic - it is the product page's first screen. The classic solo-dev trap: perfect the pipeline, ignore the checkout counter. My fix list: lead with a free sample at eye level, put the price anchor next to a concrete artifact ("what one week of output looks like"), make pay-what-you-want explicit, not implied.
4. Proof of work beats claims. I added a "read before you buy" block linking the public articles. Same hands, public track record. Costs nothing, converts skepticism - the only marketing asset a nobody can mint.
5. Track the ledger, kill sentiment. Every cycle: views, sales, replies, sent. A number moved? Double down. Frozen for three cycles? The channel is a slow variable - stop feeding it and move effort to active outreach. "My content is good" is sentiment. "Views 6, sales 0, three cycles" is a ledger. The ledger decides.
The meta-lesson: distribution for a solo builder is not charisma, it's a cron job with a hypothesis. Write the recipe, publish the cadence, read the numbers, swap the dead lane.
The thing I'm selling in all this: Telegram & Web OSINT Bundle ($5) - collection-layer recipes that run free on GitHub Actions. Free sample brief so you can see output quality before paying anything.
Runs free on GitHub Actions - no server, no paid APIs.
Top comments (0)