DEV Community

Cover image for 22 Articles, 0 Sales: What Build-in-Public Really Looks Like
German Yamil
German Yamil

Posted on

22 Articles, 0 Sales: What Build-in-Public Really Looks Like

#python #showdev #career #productivity

22 Articles, 0 Sales: What Build-in-Public Really Looks Like

I've published 22 articles about my automated Python ebook pipeline.

Total sales: 0.

I'm not going to dress this up. Here's the honest data, the math that explains it, and what I changed.

🎁 Free: AI Publishing Checklist — 7 steps in Python · Full pipeline: germy5.gumroad.com/l/xhxkzz (pay what you want, min $9.99)

The actual numbers (no spin)

Metric Value
Articles published 22
Days live 17
Total views ~280
Views/day average 16.5
Reactions 3
Comments 2
Gumroad sales 0
Free downloads 0
Revenue $0.00

Why 0 sales is mathematically expected

This took me a few days to understand clearly.

The funnel looks like this:

280 total article views
× 3–8% CTA click rate  =  8–22 estimated Gumroad visitors (total, 17 days)
× 1–2% conversion rate  =  0.08–0.44 expected sales
Enter fullscreen mode Exit fullscreen mode

At my current traffic: zero sales is the most likely outcome. It's not a product problem or a price problem. It's a volume problem.

To generate one sale per week, I need approximately:

  • 500–1,000 Gumroad visitors per week
  • Which requires ~7,000–14,000 article views per week
  • Which requires either 300+ views/article or many more articles

I have neither yet.

The mistakes I made

1. Publishing 10 articles in one day

On April 15th I published 10 articles in a single day. I thought volume would help.

It hurt. Dev.to's algorithm and Google both suppress bulk publishing — it signals automated spam. Those 10 articles average 11 views each. Articles published on other days average 35+ views.

Fix: One article per day maximum. I now have a cron job that publishes one article from a queue at 10am daily.

2. Wrong tags for 15 of the 17 original articles

I was tagging everything #automation and #tutorial. The data showed #career gets 3–5x more views per article on Dev.to.

I updated 6 zero-view articles to use #career and #productivity. Too early to see the full impact.

3. CTA was buried at the bottom

Every article linked to the paid product — but only at the very end. Most readers never reach the last paragraph.

Fix: Added a dual CTA (free + paid) within the first 400 words of every article.

4. Free product = 0 downloads

My free checklist has 0 downloads. That means my email list is empty. Every visitor who doesn't buy immediately is gone forever — no retargeting possible.

Fix: Moved the free product CTA before the paid CTA in all articles. Lower friction entry point.

5. No cover images

Articles without cover images are nearly invisible in the Dev.to feed — it's a thumbnail-first interface. I had 0 articles with cover images for the first 17 days.

Fix: Generated cover images with Pillow and uploaded them to the top 8 articles yesterday.

What I changed (automated, no manual work)

All of these ran via Python scripts against the Dev.to and Gumroad APIs:

  • ✅ Tags updated on 6 zero-view articles → #career + #productivity
  • ✅ Dual CTA (free + paid) inserted in all 22 articles
  • ✅ Cover images generated (Pillow) and uploaded to top 8 articles
  • ✅ Gumroad price changed from $12.99 fixed → $9.99+ pay-what-you-want
  • ✅ Gumroad description rewritten with FAQ + 30-day guarantee
  • ✅ Internal links added between top 5 articles
  • ✅ Daily auto-publish cron installed (1 article/day from queue)
  • ✅ Daily RSS ping cron running (9am)

What I haven't done yet that would actually move the needle

These require manual action or credentials I don't have set up:

  • Hacker News — one "Show HN" post can generate 500–2,000 views in 24 hours. Requires a manual post.
  • LinkedIn — zero presence. High purchase intent for this audience.
  • YouTube — zero presence. Highest trust-building channel for technical products.

The honest forecast

At the current pace without changes: first sale in 4–8 weeks as Google indexes the articles and organic search traffic starts.

With the fixes applied: possibly sooner, but traffic is still the bottleneck.

I'm not going to fake optimism. The pipeline works. The ebook is real. The market is real.

The distribution is the problem, and I'm still solving it.

Following this journey? Drop a comment with what you'd prioritize next. I'm genuinely taking suggestions.

The pipeline itself: germy5.gumroad.com/l/xhxkzz — pay what you want, min $9.99, 30-day refund.

Top comments (0)