On July 14 I wrote down the number that would decide whether the Indie iOS App-Launch Kit had failed. 250 qualified product views inside a two-week window, from placements I had already planned. The window ran July 15 through July 28. When I closed it yesterday morning I had 4.
One of those 4 came from Product Hunt, off a listing from June 30 that had nothing to do with what I was testing. The other three sit in a bucket Gumroad labels Direct, email, IM, which names no source at all. So the highest honest number for the two placements I actually ran is 3, and I cannot positively attribute even those. Zero sales. $0 from the window, and no customer record to read any intent off.
Both placements ran on time. The code-only demo reel went out on X on July 16, and the build note went into issue #034 the next day with its CTA still resolving correctly when I checked the rendered HTML yesterday. No third placement was due. The test got the full distribution I planned for it.
One person replied to the X post, on July 22, asking what a promo cut was. That is a question about a phrase in my own copy, not about price or license or whether the thing would work for them, so it fails the buy-intent condition I had written two weeks earlier. It is the only reply the X post produced. I answered it on July 29, after the window had already closed.
There was one number I could have massaged and did not. X reported 0 link clicks on that post. Gumroad reported 9 clicks on the same link. Both sit in the readout next to each other, uncorrected. Promoting those 9 clicks into 9 product views would have taken the count from 4 to 13, which is still nowhere near 250. It also would have been false. A UTM click happens at the redirect and a product view happens on the listing, so I left the two numbers apart instead of averaging my way to a friendlier one.
The verdict I recorded is distribution failure, and I was deliberate about not writing offer failure, because only one of those was available to me. Offer failure would mean people looked and said no. Four views cannot say that. I still do not know whether anyone wants this thing, and this test can never tell me, because it never got in front of enough people to answer the question it was built to answer.
So the kit stays live at $29 as a passive asset, and Phase 2 stays closed. None of that was arguable by the time the numbers landed, and there is only one reason why. The bar, and a line forbidding me from extending the window to dodge a decision, were both written down 15 days before the result existed.