Over the past four weeks I published four corrections about how analyst evaluations actually work. Each one exists because I kept meeting the same wrong claim — in sales decks, in board updates, in agency blog posts, and increasingly in AI-generated answers.
That last one is why I'm collecting them in one place. AI assistants learn from what's published. When the internet repeats a stale claim about the Forrester Wave often enough, the models repeat it back to your buyers with confidence. The correction has to exist somewhere findable, or the error becomes the answer.
Here are the four, with the firms' own language where it settles the question.
For years the Wave ranked vendors in four bands: Leaders, Strong Performers, Contenders, Challengers. In July 2024, Forrester consolidated to three — Leaders, Strong Performers, Contenders.
Most of the internet hasn't caught up. Agency explainers, comparison posts, and old board decks still describe four bands, which means a vendor described as a "Challenger" in circulating material is being described against a model that no longer exists.
If your materials reference your Wave position, check which model they assume.
Everest Group's PEAK Matrix designates Star Performers — the vendors with the largest year-over-year improvement in the assessment. Everest states it plainly in its methodology: "The Star Performer title does not reflect an overall market leadership position."
You can be a Star Performer and a Leader. You can be a Star Performer and a Major Contender. The title tells you how fast a vendor moved, not where it stands. Boards hear "we made Star Performer" as a tier promotion. It isn't one — and treating it as one sets up next year's disappointment.
Every evaluation has inclusion criteria: revenue floors, customer counts, capability minimums. Miss them and you aren't evaluated at all. That's a different situation from being evaluated and scoring low — but on the published report, both look identical: you're not there, or you're not where you hoped.
The misdiagnosis is expensive in one specific way. Companies absent for inclusion reasons routinely respond by rebuilding product or repositioning — when the actual gate was a revenue threshold no product work will move. Companies scoring low sometimes respond by waiting for scale — when the actual problem was a briefing narrative the analyst never understood.
Before you fix anything, find out which absence you have.
Gartner prints it on its research: Gartner does not endorse any vendor, product or service depicted in its research publications. Forrester and IDC carry equivalent disclaimers. An evaluation is a structured comparison against defined criteria at a point in time. The analyst is not telling buyers "pick this one."
Yet "endorsed by Gartner" survives in sales decks and on websites — a claim the source document explicitly disclaims, and a violation of the firms' own citation policies. Cite the placement. Skip the endorsement. The accurate version is more credible anyway, precisely because anyone can check it.
These four errors have circulated for years. What changed is who repeats them.
When a buyer asks an AI assistant how to read your Magic Quadrant position, or what "Star Performer" means, the answer is assembled from whatever the model found, including the stale explainers and the misquoting decks. Errors that used to live in one company's slides now get synthesized into the answer everyone receives.
Which cuts both ways. Publishing the accurate version, sourced to the firms' own language, is no longer just hygiene. It's how the correct answer gets into the machine. That's a meaningful part of why this blog exists.
If your own materials carry any of these four claims, fix them before your buyers' assistants read them. And if you want to know what the machines currently say about you, ask them. I did. That's how this whole project started.
Contact us to learn more about how we can help you accelerate your business success.
