Prioritizing Experiments: Effort, Success Rate, Upside

Score every project on three factors: effort required, likelihood of success, and upside if it works. Upside is the most powerful lever and the most misjudged.

Classic anti-pattern: celebrating a 50% improvement on a feature that touches 0.01% of users, while ignoring a 5% improvement that touches 50% of active users. Reach dominates.

Teams also overestimate success rates. Until you have resources for a portfolio approach, favor many small low-effort bets over a few big swings.

Upside = Reach × Impact. Impact (how much the metric moves) usually ranges 5 to 50%. Reach (how many users are affected) is the most underutilized lever.