Run four checks in order: does the denominator match, does the panel match, is it the same statistic (median vs mean), and is it the same time basis? If a comparison fails any one of them, the gap you are looking at is measurement, not performance. Most comparisons do fail, usually on the first or the third.
Add a fifth check on provenance: how many of the sources you are citing are ultimately quoting each other? Benchmark categories go circular quickly — a figure published without method gets repeated by a vendor blog, cited by an SEO page, and becomes "industry standard" within two years.
Every claim above comes from work LeanScale published. These are the sources.