Key Takeaways
- ChatGPT’s ad business is now running at a $1 billion annual rate, making it a real media channel, not a future experiment.
- Brands that test this channel now will pay significantly lower CPMs than those who arrive after auction competition matures.
- ChatGPT ads sit closer to purchase intent than most social inventory, which changes how CPG and DTC brands should evaluate them.
- Self-serve access expanding to Europe signals OpenAI is moving fast toward scale, compressing the early-mover window.
- Attribution inside AI-native environments is still broken, so testing requires a measurement framework built before you spend, not after.
- Most brands are waiting for “proof,” but the brands who built proof on paid search in 2004 and Meta in 2009 were the ones who won the decade.
ChatGPT is advertising for CPG brands, and when a platform crosses a billion-dollar revenue run rate, most CPG and DTC brands need to move fast, because the window is closing faster than their planning cycle allows for.
Why $1B in Ad Revenue Changes the Conversation
A billion-dollar run rate does not mean a billion dollars has been collected. It means that if current monthly revenue holds, that is where the annual number lands. That distinction matters. It tells you this channel is scaling now, in real time, not in a roadmap deck.
It also tells you that advertisers are already there. Someone is spending. The question is whether it is your competitors or just early-adopter tech brands running broad awareness tests that will not translate to your category.
We think it is early enough that CPG and DTC brands can still get in at rational CPMs. We also think that window is months away from closing, not years.
This Is Something More Valuable Than Search
The instinct is to slot ChatGPT ads into the search budget and evaluate them on the same terms. That framing undersells what is actually happening inside these interfaces.
When someone asks ChatGPT which protein bar has the cleanest ingredient list, or which dish soap is safe for a baby, they are not browsing. They are one step from a decision. That intent signal is sharper than most keyword queries and dramatically sharper than anything you are buying on social.
We wrote about this dynamic in ChatGPT Is a Retail Media Channel. Here’s How to Test It, and the core argument holds even more weight now that the platform is generating real ad revenue at scale.
The format is also different from a standard SERP ad. No sidebar. No second or third position. The ad appears inside a conversation that a human initiated with a specific need. That context changes how creative should work and how you should think about what a click means.
What CPG and DTC Brands Should Do Before CPMs Spike
A practical framework for getting into this channel before auction dynamics make it expensive:
- Define your intent categories first. Map out the questions your buyer is likely asking ChatGPT. Not keywords. Questions. “What is the healthiest cereal for toddlers?” is different from “toddler cereal” in ways that should reshape your creative and your landing page.
- Build a measurement baseline before you activate. Attribution in AI-native environments is genuinely unsolved right now. We covered this directly in AI Agent Attribution Is Broken: What CPG and DTC Brands Should Do. Without a pre-spend measurement plan, you will not be able to evaluate performance or justify more budget.
- Treat this as a test budget, not a channel shift. You do not need to move dollars from paid search or Meta to test ChatGPT ads. Carve out a small, time-boxed budget with a clear hypothesis. What question are you trying to answer in 60 days?
- Get your product content ready for conversational surfaces. Your current ad creative was built for interruption formats. ChatGPT ads appear in a context where the user asked for help. Your messaging needs to answer something, not just announce something.
- Watch CPM trends monthly. Early-stage inventory is priced by supply and demand just like everything else. Once major CPG holding company brands flood in, CPMs will move. You want data from before that happens.
The Waiting Game Is a Budget Decision, Not a Strategy
We hear the hesitation. Measurement is unclear. Formats are still evolving. ROI is hard to prove. All of that is true. It was also true of programmatic in 2012 and TikTok in 2019.
The brands that waited for certainty did not avoid risk. They transferred their risk from “this might not work” to “we are now paying three times the CPM and fighting for scraps against entrenched competitors.”
At Junction 37, we work with CPG and DTC brands that do not have the luxury of burning budget on trend-chasing. Everything we recommend has to connect to outcomes. But this one connects. The intent quality inside ChatGPT conversations is genuinely different from what most brands are buying right now, and the price has not caught up to the value yet.
That gap does not stay open. It never does.
If you want to know what a smart, bounded test actually looks like for your brand and category, talk to our team. We will tell you honestly whether it makes sense for you right now, and if it does, we will build the framework to test it properly.
—
Frequently Asked Questions
What Does ChatGPT’s $1 Billion Ad Revenue Run Rate Actually Mean?
It means ChatGPT’s current monthly ad revenue, when multiplied by 12, equals approximately one billion dollars. It is not money already collected over a full year. It is a real-time snapshot of where the platform sits today, and it signals that the channel is scaling fast enough to take seriously as a media investment.
Are ChatGPT Ads Right for CPG and DTC Brands Specifically?
Yes, for a specific reason. CPG and DTC buyers ask conversational, intent-rich questions inside ChatGPT. Questions about ingredients, safety, sustainability, and product comparisons. Those queries sit much closer to a purchase decision than most social inventory. The format rewards brands with clear product stories and strong differentiators.
How Should Brands Handle Attribution for ChatGPT Ads?
Carefully, and with a plan built before you spend. Attribution in AI-native environments does not behave like search or social. You need to establish your baseline metrics, agree on what success looks like in a bounded test period, and not rely on last-click models to evaluate the channel. Incrementality thinking is more useful here than standard conversion tracking.
When Is the Right Time for a Brand to Start Testing ChatGPT Advertising?
Now, if your brand has a clear point of difference and a product that answers a question people are actually asking. The early-mover CPM advantage is real and temporary. Waiting for the channel to “mature” means paying higher rates for less differentiated inventory in a more crowded auction.
—
Chris Pyne, Founder and CEO of Junction 37 and its sister venture Series A. He built Cortex, J37’s proprietary AI-driven planning ecosystem, and pioneered the integration of predictive marketing science into client strategy. Previously, Chris held C-suite roles at OMD USA and MediaCom, where he led planning for $7B in billings and 700+ employees.