Back to the blog
    AcquisitionAcquisition and marketing

    AI Is Already Driving Your Sales, but Your Reports Credit Another Channel

    The IAB has released a framework for measuring AI's role in purchases: sales influenced by AI assistants often show up as search or direct traffic, pushing budget toward the wrong channel. What it says and what to do about your attribution now.

    Sebastián RondónBySebastián RondónCEO y fundador de KMS Agency Group
    6 min
    Blue-toned illustration on a light background: a marketer examines an analytics dashboard with a magnifying glass while an AI assistant traces dotted paths toward an online store, some hidden behind clouds, with search, funnel, chart and coin icons.

    "Sales are up, but we're not quite sure why." It's a line we hear in more and more marketing meetings. Branded search grows, direct traffic grows, and no new campaign explains the jump. Increasingly, part of the answer sits somewhere your reports can't see: a conversation the customer had with an AI assistant before ever reaching your site.

    On November 12, the Interactive Advertising Bureau (IAB), the leading trade organization for digital advertising, released the AI-Era Attribution Blueprint, a framework for classifying AI's role in a conversion. A week later, on November 19, Microsoft announced that its ads now appear next to AI-powered answers on Bing. At KMS Agency, we read these two pieces of news together: AI is already part of the buying journey, yet most companies still measure it with tools built for a world of clicks.

    The problem: AI drives the sale, another channel gets the credit

    Caroline Giegerich, the IAB's VP of AI, put it plainly: as AI takes on more of the purchase journey, the clicks and referrals that attribution depends on are thinning out or disappearing. AI-influenced sales are sometimes reported as search or direct traffic, which can send the next marketing investment to the wrong channel.

    The figures released alongside the blueprint show how big the gap is:

    • 45% of ad buyers name comparing AI-driven and traditional customer journeys as their top measurement challenge, according to IAB's 2026 Outlook Study: September Update.
    • 44% rely on branded search and direct traffic as proxies for AI influence they can't track.

    Digiday's coverage adds Similarweb data cited in the IAB report: consumers who receive a brand recommendation from an AI assistant are 2.5 times more likely to visit that brand's website within seven days, but only 8.8% of those visits come directly from AI referrals. 55.9% are attributed to search and 19.9% to direct traffic.

    In plain terms: a buyer discovers your brand in ChatGPT, Gemini or Copilot, then searches for you on Google a few days later. Your dashboard hands all the credit to search.

    What the IAB framework proposes

    The blueprint doesn't replace existing attribution systems; it works alongside them. It starts from three questions: who or what created the intent, who made the decision, and who completed the purchase. Based on the answers, every conversion falls into one of three pathways:

    • Human-Owned: the person handles the journey on their own.
    • AI-Assisted: AI plays a part, recommending or comparing, but the person decides.
    • AI-Executed: AI completes the purchase.

    The pathway determines which parties are eligible to make a claim and what evidence that claim requires. The IAB is careful about one nuance: a claim is not credit; it is the right to be evaluated for credit.

    The framework is also candid about its limits. According to Digiday, the working group hit a wall: you can't standardize what you can't see. Recommendations, decisions and even transactions happen inside AI platforms, and that data isn't yet available to advertisers or measurement providers. That's why, in Giegerich's words, the document became more of a call to action for the industry.

    Two of those calls apply directly to any business running paid media:

    • Brands can estimate the share of their conversions that falls into each pathway.
    • Agencies can apply a single claim-and-evidence standard across platforms, separating platform-reported claims from independently verified ones.

    Meanwhile, ads are moving into AI answers

    The second announcement shows why this can't wait. Microsoft Advertising says ads in AI Overviews on Bing are now available in all markets except China. There's no separate campaign setup: they run through the campaigns advertisers already use, including Performance Max and AI Max. Citing internal data, Microsoft reports a 21% increase in purchases after a Copilot interaction.

    At the same time, Digiday reports that OpenAI is testing a dashboard that would show how often a business appears in ChatGPT's answers and how many clicks those appearances drive. Platforms are starting to open up some data, but they still don't connect it to sales. And Profound, a measurement company quoted in the same piece, found in a browser-session analysis that one in three consumers had a relevant AI conversation about their purchase within a month that went unattributed.

    What to do about AI marketing attribution right now

    Industry standards will take time. At KMS Agency, our advice is not to wait, and to start building your own evidence:

    1. Ask customers how they found you

    It's the simplest step and the most underrated: a form field, or a question in the first WhatsApp message or on the sales call ("How did you hear about us?"), with an option that reads "An AI assistant recommended you." Within a few weeks, you'll have your own estimate of how many conversions are AI-assisted.

    2. Break out traffic from AI assistants

    Set up a dedicated channel group in your analytics for referrals from AI platforms instead of lumping them under "referral" or "other." It's only part of the real effect, but it's the part you can actually see and compare month over month.

    3. Read branded search and direct traffic differently

    If branded search or direct traffic grows without a campaign to explain it, don't give all the credit to Google or SEO. Cross-check those spikes against your visibility in AI assistants and what customers tell you in step 1. We covered the limits of traditional attribution in our article on attribution after iOS and cookies; AI adds a new layer to the same problem.

    4. Audit how AI presents your brand

    Ask ChatGPT, Gemini and Copilot for the best options in your category and market, and check whether you show up, how you're described and which sources they cite. If you're missing, the issue isn't measurement but visibility, and that's where SEO for LLMs (GEO) comes in.

    5. Separate what platforms report from what you can verify

    With ads inside AI answers, every platform will report its own conversions. That's useful, but it isn't neutral. Track results in your CRM, match real sales with the sources customers report, and base budget decisions on that, not only on each platform's dashboard.

    Conclusion

    AI already shapes how customers discover and choose brands, even if most reports don't show it yet. The IAB framework doesn't solve the problem, but it names it clearly: when an AI-influenced sale shows up as search or direct traffic, the next budget may go to the wrong channel.

    Companies that start measuring more honestly today, by asking, separating channels and verifying with their own data, will invest better once platforms open up their data. Those that keep reading only the last click will keep rewarding the channel that closes, not the one that convinces.

    At KMS Agency, we build acquisition and paid media strategies with clear measurement, combining campaigns, content and visibility in search engines and AI assistants. If you'd like to review how you're attributing your sales and where your next budget should go, let's talk.

    Sources and further reading

    Let's design your acquisition, sales and operations system.