The useful bits

  • For a proposed test, record the market, audience, conversion definition and attribution window. Compare your measured outcomes with a relevant baseline rather than using a platform-wide adoption estimate as a return forecast.
  • Tracked brands are not necessarily active customers counted by OpenAI. A working paper and vendor panels do not establish a universal conversion rate for ChatGPT advertisers.

What the source covers

Krystal Scanlon brings together vendor estimates of ChatGPT ad reach, advertiser counts and spending mix, alongside a conversion working paper under review. These sources measure different populations and periods.

Use it in your next review

For a proposed test, record the market, audience, conversion definition and attribution window. Compare your measured outcomes with a relevant baseline rather than using a platform-wide adoption estimate as a return forecast.

Read the evidence in context

Tracked brands are not necessarily active customers counted by OpenAI. A working paper and vendor panels do not establish a universal conversion rate for ChatGPT advertisers.

Keep learning

What would you add?

Bring your experience to the conversation. The next useful perspective could be yours.

Talk with the collective