OpenAI’s ChatGPT Ads just got another targeting piece that advertisers know from Google and Meta. LiveRamp has expanded its partnership with OpenAI so brands can bring first-party audiences into campaigns through RampID.
The integration covers data from CRM systems, loyalty programs, websites, apps and other owned sources. Marketers can reach or suppress specific customers and prospects, then apply bid multipliers once those audiences are attached.
What changed
LiveRamp and OpenAI first connected in June, when LiveRamp became a measurement partner for ChatGPT Ads. The new move extends that relationship from measuring conversions to activating audiences.
For brands already on LiveRamp, ChatGPT becomes another destination for the same first-party data they use across other media plans. The rollout starts in 11 markets, with more to follow as ChatGPT Ads expands.
The mechanics will look familiar:
- Reach or exclude customers and prospects using first-party data.
- Apply bid multipliers between 0.1x and 10x at the ad group level.
- Activate RampID audiences rather than maintaining separate CSV or TXT uploads.
No LiveRamp? You can still upload lists
LiveRamp is not required. OpenAI already supports Custom Audiences directly in Ads Manager. Advertisers can upload CSV or TXT files with emails, phone numbers, hashed identifiers or Google Advertising IDs, then include or exclude those groups at campaign level.
The partnership mainly reduces setup work for existing LiveRamp customers. Megan Morris, Director of Integrated Media Solutions at Allegiance Group & Pursuant, said the platform is becoming an increasingly popular engagement surface for donors, which makes it worth measuring alongside other channels.
Why marketers should test suppression first
Having Google- and Meta-style controls does not mean ChatGPT ads will behave like Google or Meta. Those platforms have years of conversion history. ChatGPT Ads is newer, and conversation context influences which ads users see.
There is almost no public performance data for first-party audiences in ChatGPT yet. That makes suppression the lowest-risk starting point: exclude recent buyers from acquisition campaigns, or stop current subscribers from seeing introductory offers.
Using first-party data to find and bid up audiences carries more unknowns. OpenAI’s audience-size requirements limit narrow segmentation, and there are no public benchmarks against ChatGPT’s other targeting options.
Treat this as a small test, not a budget shift
Brands already running LiveRamp across multiple platforms can add ChatGPT without building a separate data process. That lowers the cost of a test, but easy setup alone is not a reason to move money.
Advertisers should look for a specific hypothesis: reaching known customers at a different research moment, testing whether conversational context changes response, or finding incremental conversions from an already-known audience. Until OpenAI or LiveRamp publishes comparative results, a smaller test is easier to defend than pulling budget from campaigns with established performance.
Pricing is also still an open question. OpenAI has not published separate pricing for Custom Audiences, and there are no benchmarks showing how first-party audiences perform against native targeting. Bid multipliers give control, but advertisers will need conversion rates and customer value data to judge whether the cost is worth it.
For now, first-party targeting is available in ChatGPT Ads through two paths. What marketers don’t have yet is enough performance history to know how much it should change their investment.
Source: Search Engine Journal
