METRhttps://metr.org/

Your landing page’s weakest score is CTA.

How METR ranks

  1. 1.ARC Prize Foundation76
  2. 2.Epoch AI (FrontierMath / Benchmarks)76
  3. 3.METR73
  4. 4.Kaggle Competitions72
  5. 5.LMArena66
  6. 6.Artificial Analysis65
Fix CTA
73Growth scoreupdated

category scores

CTA64Information hierarchy72
category scoresShowHide
CTA64Information hierarchy72

CTA

64Fair

Copy, size, verb, cost-feel — does the primary CTA earn the click?

METR
LMArenain-product action
ARC Prize Foundation
Epoch AI (FrontierMath / Benchmarks)
Kaggle Competitions

Problems

  • No action verb — reads as a label
  • Weaker than card CTAs ("Read Report")
  • Inconsistent casing: Report vs report

Information hierarchy

72Fair

Does the design help or inhibit what the page is trying to say?

Information hierarchy annotations on the desktop capture123
  1. Read Report cards — Two saturated dark-green report panels carry more visual weight than the headline and 'Our research' CTA, splitting the fold's focus.
  2. Read Report link — Same-weight 'Read Report' exits on both cards compete directly with the primary 'Our research' button; the duplicate below repeats the leak.
  3. Truncated chart section — A cut-off chart with 'Measurements above 16 hrs are unreliable' creeps into the fold as a stray, context-free fragment.

Value props

72Fair

Title, subtitle, hero — do they sell the product's true value?

METR’s headlineModel Evaluation & Threat ResearchWe inform the public about the risks and capabilities of advanced AI systems.

Working

Could improve

Trust

76Fair

Proof that the product is real, competent, and safe to try.

METR desktop capture for Trust

How METR compares

Trust read

Mobile web

82Strong

The phone experience as a first-class citizen.

METR mobile capture for Mobile web

What works

What to improve