The underperforming ads check flagged the ad that brings in our best cases

Purely click-through based, and you have found the honest limitation of it rather than a bug. The check on /dashboard/wasted-spend flags an ad when it has at least 500 impressions and its click-through rate is below half the average for its ad group, and it needs at least two ads in the group to have an average worth comparing against. There is no conversion component in that test at all, which is exactly why your ad is on the list. The reasoning behind leaving conversions out: on most accounts an individual ad does not accumulate enough conversions to compare fairly, so a conversion-based test would either stay silent or fire on noise. Click-through rate is the metric that's statistically usable at the ad level. That makes the check a good detector of ads nobody wants to click, and a poor judge of ads that filter deliberately - which is what yours is doing. Your ad is doing something valuable and unusual: repelling the wrong clicks. In a vertical at your CPCs that is worth more than click-through rate, and no CTR-based test will ever see it. There is no per-ad mute today. That's a reasonable request and worth its own post in feature requests so it does not get lost in this thread - the estimated saving on that row is also wrong for you, and both would be fixed by the same thing. For now: the saving figure on underperforming ads is deliberately conservative for related reasons, so treat that whole section as a prompt to look rather than a recommendation to act. The zero-conversion keyword and search term sections are far more clear-cut.