Use extreme selection criteria so your good doesn’t crowd out your great

A small team kept saying yes to decent projects. The pipeline was full, the calendar packed, and morale oddly flat. They wrote a short list of minimum criteria and three ‘ideal’ criteria for new work. Then they added a single ruthless rule: if a prospect didn’t score 90 or above on their top criterion—strategic fit—they’d decline, even when the fee looked tempting. The first few nos hurt. But within a quarter, their weeks were quieter and profits higher. The projects in flight actually needed their unique skills, which meant fewer revisions and better results.

One afternoon, an email arrived with a logo they could design in their sleep. It scored 80 on money, 60 on fit. They passed and instead doubled down on a 95‑fit client, building a playbook that became sellable IP. Their coffee stayed warm more often because they weren’t context switching all day.

On a personal level, one team member set a 90% rule for networking: if a meeting didn’t score 90+ on ‘mutual curiosity and aligned work,’ he declined with thanks and suggested asynchronous exchange. He found himself leaving fewer conversations drained.

Psychologically, the 90% rule trades the pain of a few hard nos for the joy of a clean calendar and high‑energy yeses. It combats ambiguity aversion by forcing a clear threshold. It also exploits selection effects: when you consistently choose only high‑fit opportunities, you compound expertise and reputation in that lane, which in turn attracts more of the same. The hard part isn’t scoring, it’s the courage to hold the line.

Pick one decisive criterion that matters most this season and rate incoming options from 0 to 100 on that single factor. Treat anything below 90 as zero and decline cleanly. Write three minimum and three ideal criteria on a note where you triage requests so decisions are fast and visible. Keep a tiny log of each yes and what it produced to refine your filters. Try this on the next three opportunities that hit your inbox.