One example rarely covers a whole practice. When you cast a verdict, use objective applicability (covered / partially / not) so the tally reflects which AOs still need another pattern.
Disagree here if you think the platform should force per-AO verdicts instead of one overall label + applicability map.