The Gaffer was going to be my grassroots football app: fixtures, availability, payments, comms for Sunday league clubs. I had the name, the folder, and a place for it on this site. What I did not have yet was a single line of code, which turned out to be the best thing about it.
Before building, I ran the idea through a pressure test I now apply to everything: who is the user, what do they use today, what would make them switch, and who already owns the relationship. The answers were brutal. Stripped to its viable core, The Gaffer was a worse, paid version of Spond, which is free, established, and good enough. No distribution wedge, no trust wedge, no pricing room.
I tried the pivot: reframe it as sponsor retention for the local businesses that fund grassroots clubs. That walks straight into SNAP Sponsorship, entrenched in exactly that niche at nine pounds a month. The pivot did not escape the problem, it just moved the competition from consumer to B2B, where switching costs are higher.
So it is parked. Total cost: a few hours of thinking and one markdown file recording why, so future me does not relitigate the decision from scratch. The note ends with a rule: come back with a distribution wedge, not a feature set.
The same process killed a second idea, a UK immigration navigator, for a different reason: the useful version of the product is a regulated activity that requires registration as an immigration adviser. The legal version just rebuilds gov.uk. That one is not parked, it is scrapped, and the note carries its own rule: if the value depends on telling someone what to do in a regulated domain, check the regulation before the first screen.
Ideas are cheap to kill before code and expensive to kill after. The discipline is having a process that runs before enthusiasm commits you, and writing the verdict down so it stays killed.