Why a lean operator pays the highest price for the wrong lens

If you run a company largely on your own, with AI agents doing the legwork, the stakes on this choice are higher, not lower. A large team can afford to half-optimise four stages at once and average out to progress. You cannot. You get one stage of real focus, so the single highest-leverage decision you make is which stage that is, and the model is the thing that names it. Choose the lens, and your scarce attention lands where the growth actually is rather than where the loudest dashboard happens to be.

Consider a worked example. A solo founder selling a 4,000 EUR-a-year B2B tool spends a quarter and a few thousand euros on Google Ads because the acquisition dashboard is the one they look at every morning. Leads climb. Sign-ups climb. Revenue does not, because seven in ten new accounts never reach the moment the product becomes useful and churn out inside sixty days. That is not a pessimistic figure: across 62 B2B SaaS products, Userpilot's 2024 benchmark put the median user activation rate at roughly 37 per cent, meaning fewer than four in ten sign-ups ever experience the thing the product is supposed to do for them. The acquisition lens told this founder to buy more traffic; the business needed lead activation. The campaign was a success and the company stood still. The cost was not the ad spend. It was the quarter.

The deeper point for a one-person operation is that the model is also how you divide the work between yourself and your agents. A model you can act on beats a model you can only admire. When the lens names a single stage as this quarter's bottleneck, you know exactly which playbook to open and which standing automation to point at it, and you give yourself explicit permission to ignore the other three for now. That permission to ignore is the entire benefit. Without a chosen model, everything feels equally urgent, which is the same as nothing being prioritised at all.