Is there an easy to tell when the "text matches" are pretty far? As in only 3 out 6 tokens were matched or something? I'd like to use this distance to show something like: "Can't find what you are looking for? Request here"

No there's no easy way to tell this. The match score is an aggregation of multiple scores but the components are not directly exposed. Can you please create an issue for this on Github? I think it will be great to expose the ranking information.

on it