Skip to the shop floor

AI tool that ranks headlines, titles and subject lines head-to-head to predict which will perform best, trained on thousands of real A/B tests.

About this launch

Screenshot of the Click Oracle homepage
clickoracle.trylumegrid.com, as visitors see it

Click Oracle ranks your headline options head-to-head to predict which will get the most clicks or opens before you test them live. You paste up to 200 variants or let Claude write them from your brief. The tool compares each pair in both orders, combines results with a Bradley-Terry model, and shows you win odds with a verdict: Clear (82% accurate), Lean (67% accurate), or Too close to call (59% accurate). It works for article headlines, YouTube titles, newsletter subject lines, ads and video hooks. You can upload your past test results to a brand profile so Click Oracle measures and calibrates its accuracy to your specific audience.

What it does

  • Head-to-head comparison of up to 200 variants
  • Bradley-Terry model ranking with calibrated odds
  • Honest uncertainty verdicts: Clear, Lean, Too close to call
  • Brand profiles to calibrate accuracy on your audience
  • Works for YouTube titles, subject lines, ad headlines, and article headlines
  • Accuracy measured on 9,841 held-out A/B tests

Pricing

Free: $0/month with 5 rankings and 60 Claude-written options monthly. Creator: $19/month with 100 rankings and 2,500 options. Pro: $49/month with 600 rankings and 6,000 options. Agency: $149/month with 3,000 rankings and 20,000 options.

Shopper questions

Does this replace A/B testing?

No. Click Oracle decides what is worth testing by ranking your options. Drop the weak options, then test the top tier live. On held-out tests, Clear head-to-head calls picked the better of two versions 82% of the time. When the whole ranking's verdict was Clear, the #1 pick was the single best version in the test 51% of the time versus 24% by chance.

How does it predict a test it has not run?

A decision model judges each pair of options in both orders to determine which one a typical person would click or open. Those head-to-head comparisons are combined into one ranking using a Bradley-Terry model, then calibrated against thousands of real headline A/B outcomes so the odds accurately reflect what those calls mean for that format of test.

Will it work for my audience?

You can measure it on your own past tests. Upload your past test results as CSV to a brand profile and you will see how often Click Oracle would have picked your real winners by accuracy band, and your rankings will use calibration fitted to your data.

What does "too close to call" mean?

The top options are within the range where real tests go either way. On held-out data, calls in that band were right 59% of the time, so Click Oracle recommends you pick on brand fit or test them live.

Can Claude write the options?

Yes. Describe the content and the audience, and Claude writes distinct options across different angles. It is instructed to use only facts from your brief and to avoid clickbait the content cannot deliver. Check every option before you publish. The Free plan includes 60 written options a month.

More on this profile

Click Oracle has a free listing. The owner can unlock the full profile with a paid launch for $5:

  • Founder story (locked)
  • Milestones (locked)
  • Demo video (locked)
  • Dofollow website link (locked)

Shoppers also viewed

Support