Test ad copy with AI: what ChatGPT and Claude can tell you and what an audience adds
Testing ad copy with AI can mean three things: asking an assistant like ChatGPT or Claude, letting a simulated audience read it, or running variants live. Each answers a different question. Here is which, and how to get more out of each.
Last updated:
Three ways to test ad copy with AI
| Way | Answers | Does not answer | Needs |
|---|---|---|---|
| Ask an AI assistant (ChatGPT, Claude) | Is the copy clear? Which variants are there? | Whether your audience prefers it | Only the copy |
| Simulated audience | Which version the personas prefer and what holds them back | Whether the version gets more clicks in the market | Personas based on real customer voices |
| Live test with ads | Clicks and enquiries | Why a version wins | Budget, traffic and run time |
The ways complement each other: write with the assistant, shortlist with the audience, test the two or three strongest live. More on the ways compared under message testing.
How to get more out of ChatGPT or Claude
- Give context: audience, channel and where the reader comes from. Copy for a search ad has a different job than copy for a newsletter.
- Compare instead of scoring: give two versions side by side and ask about the difference. Scores from 1 to 10 say little.
- Ask for the strongest objection: what stops a sceptical reader from clicking?
- Plan for agreement: Language models tend to agree with their users. A study by Anthropic calls this “sycophancy” and finds it in several widely used assistants. So ask for problems, not for approval, and check whether the criticism also comes when you do not ask for it.
- Do not adopt numbers you cannot back up: assistants like to suggest specific percentages. If the number is not in your own material, it does not belong in the ad.
Three questions for every piece of ad copy
- Does someone without prior knowledge understand what is on offer?
- Does the copy match what the reader just searched for or saw?
- Does it say how the result comes about, not just that it does?
For headlines in particular: headline analyzer.
Test ad copy with 99takes in Claude
99takes writes challenger versions from your material, without numbers that are not backed up. Three versions of a headline cost about 6 credits (challenger versions). If you already have several drafts, the shortlist sorts them for 3 to 5 credits.
The best version then competes with your current one, read by personas from your audience. With 12 personas that costs about 73 credits, with each persona also reading under the others’ jobs; one-to-one about 19.
Whether competitors already make your core claim is answered by the claim check for 3 to 7 credits.
All of this is a simulation before launch, not a market measurement. It helps you go into the live test with the stronger versions.
Frequently asked questions
Can ChatGPT test ad copy?
ChatGPT can rate copy, find unclear passages and write variants. It does not measure behaviour, and without details about your audience it judges from general knowledge. To find out which version your audience prefers, you need readers from that audience, real or simulated.
Test ad copy with AI or run an A/B test?
Both, one after the other. With AI you shortlist while many versions are still in the race and no budget should flow yet. The A/B test with real ads decides in the end what brings more clicks and enquiries.
What is a simulated audience?
Personas played by a language model that are meant to read copy the way people from your audience would. How well that works depends on what they are based on.
What does it cost to test ad copy with 99takes?
Writing three versions of a headline costs about 6 credits, a shortlist 3 to 5, a comparison with 12 personas about 73. You start with 350 credits, no payment details.
Related
Shortlist ad copy before the budget runs
Create an account, connect 99takes to Claude and have two versions read against each other. You see the price before it starts.
Start for free