Audit tools, including Ancil, run a set of checks against a portal and turn the results into a score so admins and leaders can see progress over time. The useful part is not the number itself but the trend and the list of issues behind it. A score should always link back to the specific findings that moved it.
A good health score weights by business impact. Five extra super admins or tickets breaching their SLA matter more than a few lists with similar names. If a score simply counts issues, a large portal with many minor items looks worse than a small portal with one serious security gap, and teams end up fixing what is easy instead of what matters.
A health score should not measure things that are a matter of preference, such as naming style, or things HubSpot behavior makes unavoidable. It should not punish a portal for size, and it should not treat a missing optional feature as a failure if the business has no use for it. It also cannot judge whether your process is right for your market. It can only show whether the setup is consistent and maintained.
Use the score as a conversation starter. Agree which findings you will act on, which you accept as intentional, and who owns each one. A score that rises because issues were fixed is useful. A score that rises because checks were switched off is not.