I looked into similar projects and the business model seems to be that you pay for certification, they do the test, and they certify your product if it passes.
So if they publish poor performers, then it kills their business model because nobody would want to pay for certification if they risk getting revealed as a poor performer.
BTW ICP-MS lab tests are like $100/pop which is probably why nobody has a table of real data.
If it’s such an important mission for them, then why not publish the offenders?