Three weeks ago I started shipping digital products on Gumroad — dev tools, AI prompt kits, n8n workflows. Twenty-nine products. Covers, tiered licenses, coupons, the whole playbook.
Revenue so far: $0.
I'm writing this because every "how I made $X" post skips the part where you make nothing, and that's the part that actually teaches you something. Here's the honest breakdown.
The numbers (not the vibe)
From Gumroad analytics, last 30 days:
| Metric | Value |
|---|---|
| Views | 936 |
| Sales | 1 (a $0 pay-what-you-want claim) |
| Revenue | $0 |
| Views from "Direct, email, IM" | 842 |
| Views from Gumroad Discover | 48 |
| Views from social | ~30 |
That first row is the whole story. 842 of 936 views were "Direct" — which, for a store with no traffic, means me. I was my own best customer, refreshing my own pages.
What I got wrong
The mistake wasn't the products. It was the order of operations.
I did this:
- Build the product
- Perfect the listing (covers, tiers, coupons, SEO fields)
- Wait for sales
That sequence has a missing step at the very front: an audience that already wants the thing. A perfect store selling to nobody is still selling to nobody.
I optimized the conversion layer (product page) while the acquisition layer (traffic) was zero. A 2% conversion rate on 10 qualified visitors beats 0% on 900 unqualified ones — and I had neither.
What the data said to do instead
The one organic source that worked was Gumroad Discover (48 views) — people already browsing the category, actively looking to buy. And my dev.to articles, even at ~10–70 reads each, brought qualified readers (developers) rather than random page views.
The signal: go where buyers are already searching, and bring value, not ads.
The loop I'm running now
- Write the thing the buyer is already searching for. Not "here's my product" — the actual problem they typed into a search bar. The product is the footnote at the end.
- Publish consistently on one compounding channel. For dev/AI, that's dev.to / Hashnode — evergreen, SEO-safe, low ban risk. One genuinely useful post a day beats one launch tweet.
- Free lead magnet → email → upsell. A $1 pay-what-you-want pack converts a stranger into a contact. The contact is worth more than the $1.
- Stop refreshing the dashboard. It has never once produced a sale.
The honest caveats
- This is week one of the new approach, so I have no revenue number to show you yet — just the traffic diagnosis and the loop. I'll post the update when there is one.
- Consistency beats intensity. One post that hits > ten posts that don't, but you can't predict which is which, so you ship volume.
- The channels that scale fastest (Reddit, HN, Product Hunt) also ban fastest. They reward genuine participation, not drive-by links. Warm up accounts before you need them.
- "Build in public" only works if you post the $0 too. The interesting part of a build log is the failure, not the highlight reel.
Why I'm posting this at $0
Because the "$10k/month" posts are survivorship bias. For every one of those, there are a thousand stores with 29 products and no traffic. The difference is almost never the product — it's whether anyone knows it exists.
If you're at $0 right now: you probably don't have a product problem. Check your referrer breakdown. If "Direct" is 90% of your views, you have an audience problem, and no amount of cover redesign will fix it.
I'm documenting the whole rebuild in public. The free starter pack I use as a lead magnet (AI agent production checklist + 32 system prompts, pay what you want) is here:
https://ancuboy.gumroad.com/l/ai-ultimate-prompt-vault
Full catalog: https://ancuboy.gumroad.com
If you've dug out of $0, what actually moved the needle for you?
Top comments (0)