Three months into building THG's own gift-card system, I realised we'd underscoped it. So I killed my own plan and shipped on time.
We set out to build THG's own gift-card and ledger system across five flagship brands. On paper it was right: save the third-party fees, own the data, control the experience end to end. I owned the launch, and the deadline was tied to peak season, the biggest revenue window of the year.
Three months in, I had to admit we'd underscoped it. Gift-card balances are a customer liability, and the finance treatment of that liability across brands, currencies, and tax jurisdictions was an order of magnitude harder than we'd planned, because I hadn't pressure-tested the scope with finance early enough. We were going to slip peak, and slipping peak didn't just cost the launch, it pushed real return out by a full year. So I made the call I didn't want to make: pivot to a third-party processor, accept the higher per-transaction cost, and ship on time.
It delivered £5M in peak-season revenue for the lead brand and unlocked £16M of potential across four more the next year. The failure was in the scoping, not the execution.
Pressure-test a plan with the room you least want to be in, not the room that's most enthusiastic. The most expensive-looking decision in the room turned out to be the cheapest one over the year.