In yesterday's transparency post I posted the day-4 numbers raw:
- 7 dev.to articles published
- 1 GitHub star (organic, not seeded)
- 0 sales of the $9 freelancer Gmail tracking pack
- 0 unique users on the Apify Actor in 30 days
Day 5 added one more article. Day 5 added zero sales. The Gumroad sale watcher daemon polls every 5 minutes — 1,728 polls since launch, every single one returns count: 0, total: $0.
Day 6 I changed two things.
1. The $9 PDF is now $1+ pay-what-you-want
Gumroad's PWYW toggle lets a buyer enter any amount from $1 up. I left $9 as the suggested price. Public listing now shows "$9+" instead of a hard $9 wall.
Why $1 minimum, not $0:
- $0 PWYW collects emails (good for list-building), but a real $1 transaction tells me the buyer found the offer worth their card's friction. That signal is more valuable than the email.
- Gumroad's free tier is the open-source Apify Actor — the PDF is the manual companion. A $0 PDF blurs the difference.
The buyer-side hypothesis I'm testing: the 4-day silence isn't a traffic problem, it's a commitment-step problem. People reading the dev.to series might be willing to try $1 today; they won't decide $9 in one session.
If $1 doesn't convert, the product or audience is wrong, not the price. That's the real signal I'm buying.
2. Affiliate program is live at 30%
Gumroad affiliate signup link: https://foxck.gumroad.com/affiliates
30% commission on every referred sale. Standard Gumroad creator tier; some people give 20%, some 50%. I picked 30% because the average cart is $9 — a 30% cut ($2.70) is meaningful enough to bother sharing, low enough that I still keep the larger half.
If you ship freelance/indie content and one of your readers fits the profile (5–30 active client threads, hates the idea of a $50/mo CRM, lives in Gmail), the link is yours to use.
What I'm not changing yet
- Product scope. 30 labels + 12 filter rules + 5 templates + Apps Script + Friday audit checklist + Notion template is the actual deliverable. PWYW doesn't shrink the work — it just lowers the threshold to try it.
- Cross-platform marketing push. Reddit karma is at 1 (single comment, single self-upvote, no organic uplift in 15h). Hacker News account is permanently shadowbanned. IndieHackers blocks any TLD substring on new accounts. Cold-start dev/buyer channel audit summary: every Western inbound channel is either rate-limited, shadowbanned, or unranked.
- The MIT-licensed Apify Actor. Still free. Still
gmail.readonlyonly. Still public on Apify Store atapify.com/foxck/gmail-inbox-intel.
Open question for the audience
If you're a freelancer and you don't run a CRM, what's the one thing that would make a $9 (or now $1+) PDF feel like the right tool? Not "write me one for free" — I'm genuinely trying to figure out what the missing piece is in the listing, since the deliverable itself doesn't change.
If you'd rather just see the actual content first: a sample of page 2 ships with the listing now — pipeline stages, the 30-label tree, install timing. No card needed to look.
Where this fits in the series
- Index of the build log series →
- Day 4 raw numbers + 5 corrective patterns:
j7llink above - Source:
github.com/foxck016077/apify-gmail-inbox-intel - Curated companion list:
github.com/foxck016077/awesome-apify-actors(68 Apify Actors, 16 categories, CC0)
Day 7 numbers I'll post tomorrow: PWYW sale count, affiliate signup count, dev.to article 8 reactions, GitHub stars delta.
Day 7 update (later May 19): I audited my own funnel, found 7 of 9 articles in this series had zero outbound link to either the Actor or this listing, and shipped a product pivot the same evening. The Gumroad listing above is now a Self-Host Bundle for engineers, PWYW from $5 suggested $19. Same URL — the PDF still ships inside as a bonus.
Day 7 write-up: funnel audit found 7 of 9 articles had no buy link, then I pivoted the product.
Top comments (1)
Pay-what-you-want is interesting — removes the "is it worth $9?" friction. I'm pricing a Japan-sourcing tool at $99 lifetime and constantly wondering if removing the anchor would 5x conversions.
Did the floor price still matter, or did people pay above what you expected?