Key Takeaways
- Waiting for the IAB to define AI advertising measurement standards means ceding control of your attribution model to committees that don’t know your brand.
- AI agents are already making purchase decisions on behalf of consumers, which means traditional last-click and multi-touch models are already undercounting real influence.
- CPG and DTC brands that build first-party data infrastructure now will have a meaningful advantage when agentic attribution standards are finalized.
- Most brand measurement stacks were not built to track a touchpoint that exists between a consumer’s intent and their purchase, which is exactly where AI agents operate.
- The brands that pressure-test their measurement architecture before a standard is imposed will have more negotiating leverage with platforms when the rules change.
AI advertising measurement is broken for CPG and DTC brands right now, not at some future date when the IAB finishes its framework. AI agents are already filtering, summarizing, and acting on product information before a consumer ever reaches your product page, retailer listing, or cart. If your attribution model doesn’t have a way to account for that intercept, you’re already flying blind on part of your funnel.
The Funnel Has a New Middle Layer and Most Brands Are Ignoring It
The traditional purchase funnel had a predictable structure. A consumer saw an ad, visited a page, made a decision. Attribution models, even messy ones, could be mapped to that journey.
AI agents break that structure entirely. A consumer asks an AI assistant what protein bar to buy. The agent synthesizes product information, reviews, and potentially ad-influenced content, then returns a recommendation. The consumer buys it. Where in your attribution model does that interaction live?
Right now, for most brands, it doesn’t live anywhere.
This is not a future problem. Tools like ChatGPT, Perplexity, and Google’s AI Overviews are already shaping purchase decisions at scale. We have written about how ChatGPT is functioning as a retail media channel and what that means for brands willing to move early. The point stands: the channel is live, the measurement infrastructure is not.
What “AI Agent Attribution” Actually Means
AI agent attribution refers to the process of assigning credit to advertising or marketing touchpoints that influenced a purchase decision made by or through an AI agent, rather than a human clicking directly through a traditional path.
The challenge is that AI agents operate as intermediaries. They consume inputs, including potentially ad-influenced content and branded product data, and output decisions or recommendations. Measuring the advertising influence within that process requires a fundamentally different approach than pixel-based or cookie-based attribution.
It is a hard problem. But it is a solvable one if brands start building the right infrastructure before the industry’s answer is handed to them.
Why Industry Standards Alone Won’t Save Your Attribution Model
The IAB building a framework is useful and we aren’t dismissing it. But industry standards are built by consensus, and consensus takes time and tends to protect the largest players at the table.
By the time a formal AI advertising measurement standard is ratified, the major platforms will have already shaped it in ways that favor their own reporting. This isn’t cynicism. It’s the history of every prior measurement debate, from viewability to cross-channel attribution to cookie deprecation.
CPG brands in particular cannot afford to wait. The category is intensely competitive, margins are tight, and the distance between a consumer’s intent and their checkout is compressing fast because of agentic commerce. We covered the broader implications of this shift in our piece on agentic commerce and what CPG brands must do now.
Waiting for the standard is a passive strategy in an active environment.
What You Should Actually Be Doing Right Now
Here is how performance-focused CPG and DTC brands should approach AI attribution before any standard is set:
- Audit your current attribution model for agentic blind spots. Map every step where an AI tool could intercept a consumer decision and ask whether your current model captures any signal from that step. Most models will come up empty.
- Invest in first-party data capture at every owned touchpoint. When platform-level attribution gets murky, owned data becomes your most reliable measurement asset. Email capture, loyalty programs, direct purchase data: these give you signal that does not depend on a platform’s reporting API.
- Run incrementality tests now, before the environment gets noisier. Geo-based holdout tests and matched market experiments give you a baseline that survives attribution chaos. They’re harder to run later when the measurement landscape is shifting under your feet.
- Start tagging and tracking traffic from AI referral sources. Tools like Perplexity and ChatGPT are beginning to surface referral data. Set up your UTM structure and analytics to capture that traffic distinctly so you have historical data when volume grows.
- Pressure-test your media mix model against agentic scenarios. Ask your agency or analytics team to model what happens to your reported ROAS if 10 or 20 percent of influenced purchases move through an AI agent that your current model cannot track.
The Measurement Stack Most Brands Have Is Already Out of Date
Most brand measurement stacks were built for a world where a human consumer moved through a predictable digital path. Click, visit, convert. Even sophisticated multi-touch models assume a human is touching each point.
AI agents do not behave like human consumers. They do not click ads in the traditional sense. They ingest information, weight it according to their own logic, and produce recommendations. Your pixel doesn’t fire on that process. Your view-through window doesn’t capture it. Your last-click model definitely doesn’t credit it.
The brands that recognize this early and rebuild their measurement thinking around it will have cleaner data and better decisions when the rest of the industry catches up. That is a real competitive advantage, and it’s available right now.
—
FAQ
What is AI advertising measurement and why does it matter for CPG brands?
AI advertising measurement refers to tracking and assigning credit to ads that influence purchases made through or by AI agents, not just human clicks. It matters for CPG brands because AI tools are already shaping consumer purchase decisions at scale, and most current attribution models have no way to capture that influence.
How are AI agents changing the purchase funnel for DTC brands?
AI agents create a new layer between consumer intent and purchase action. Instead of a consumer clicking an ad and visiting a product page, they may ask an AI assistant for a recommendation and act on that. DTC brands lose attribution visibility at that intercept point because traditional tracking tools are not built to follow that path.
Should CPG brands wait for the IAB framework before updating their measurement approach?
No. The IAB framework will take time to finalize and will reflect compromises made by large platform stakeholders. Brands that build first-party data infrastructure, run incrementality tests, and audit their attribution models now will be better positioned when the standard is set, not scrambling to comply with it.
What is the most important first step in preparing for agentic attribution?
Audit your current attribution model for blind spots at the AI intercept layer. If you cannot identify where an AI agent recommendation would show up in your current reporting, you already have an untracked portion of your funnel. That is the starting point.
—
Ready to build a measurement architecture that does not depend on industry standards catching up to your reality? Talk to Junction 37 about performance media built for the way consumers actually buy today. Or see what that looks like in practice at our services page..
—
Chris Pyne, Founder, Junction 37 – 30+ Years in Performance Media