Early on, we said yes to nearly everything that came in the door. It felt like the responsible thing to do — revenue is revenue, and turning down a paying client seemed like a problem only businesses with the luxury of choice got to have. It took a few overextended quarters to notice the actual constraint wasn't sales. It was senior attention, and there's a fixed amount of it no matter how many contracts get signed.
A few patterns showed up reliably in the projects worth declining: no single person who could actually make a decision, a scope that kept expanding in the kickoff call before any contract was signed, or a budget that assumed a much smaller ambition than what was being described. None of these are disqualifying on their own. Together, they're a reliable predictor of a project that consumes far more attention than its size suggests.
The cost of taking on a bad-fit project isn't just the hours it burns. It's the better-fit project that gets a worse version of us because our attention is split, and it's the reputational risk of shipping something mediocre with our name attached to it — which, per the previous problem with referrals, is the thing that actually generates the next project.
The heuristic we use now is deliberately simple: if we can't describe what 'done and good' looks like in one sentence before the work starts, we don't start. Not because ambiguity is always bad — some of the best projects begin genuinely open-ended — but because there's a difference between open-ended and undefined, and only one of those is a project we can actually deliver well.
Turning down work never feels like a strategy in the moment. It feels like leaving money on the table. It's only in hindsight, looking at which clients came back and which projects we'd rather not have our name on, that it's obviously the better trade.