The fact that the impact is generic, and you can not be sure about that the candidate's actions were the true cause?
In this case would a more specific claim with a more solid reasoning help?
Or are you dismissing all claims of potential positive impacts?
One objection is of course that 93% is very achievable if the numbers involved are very small, but in that case the statement might just be code for "selling the office sofa" which is often seen as less impressive.
We had to re-earn their business.
In this situation I was actively evaluating support issues, customer reported problems and new feature requests to triage what was worked and when. I’m comfortable suggesting that if somebody else was making those decisions at the time, the entire company would have folded. It’s hard to expand further on that without going into a lot more detail though.
Your use of the non-statistical meaning of “accurate” as “free from error” is spot on.
I'm on a team that rebuild a sales/onboarding funnel. I was told it resulted in 5 million in additional revenue generated this year. I'll use that number on my resume at some point. But mostly because I don't have a more useful number to use.
If you ask me I can share all about how and why we build it which is probably what you actually care about, right? 5 mill might be a lot at a small company or almost nothing at a large company so $ and % are pretty much irrelevant. It's the process of building that you would actually care about, no?
Maybe I've read it too much but doing the actual analysis of "what have I done and how was the impact" is a lot more than simply taking revenue beforehand and revenue afterwards and calling it a success. Also if I solve a bug that was discovered by QA, specified by PM that was hogging a lot of resources, did I really increase revenue? This trend in CVs is really annoying me as most of the time it's complete BS and only there because people watched a YT video on how to build your resume.