Attribution in AI Search: Measuring B2B Marketing Right

Attribution in AI search: the problem behind the problem

You’re not losing clicks. You’re losing the big picture.

Picture a technical buyer. He needs a component, a service provider, a solution. A few years ago he would have Googled it. Today he asks ChatGPT.

He types in his question. The AI names three providers. One of them is you, or maybe not.

He reads the answer. He doesn’t click. He remembers the name.

Two weeks later he calls. Or he fills out your contact form. Your analytics then shows: “Direct.”

Now here’s the part that hurts. You think the problem is “AI is stealing my clicks.” It isn’t.

The real problem sits one level deeper. Your measurement doesn’t capture the AI source at all. You shift budget by numbers that stopped telling the truth long ago.

For marketing leads in industry, that’s especially awkward. Your buying cycles are long. Your deals are big. An invisible source at the start skews everything you steer by.

That’s the difference between “less traffic” and “flying blind.” Less traffic, you notice. A blind spot in your measurement, you don’t.

What visibility in AI search actually means

Visibility in AI search comes down to one simple question: does your brand show up in the answers AI searches give?

We mean the big answer engines:

• ChatGPT: OpenAI’s chat assistant, often the first stop for research.
• Perplexity: an answer engine that cites its sources directly.
• AI Overviews: the AI summaries at the top of Google’s results.
• Gemini and others: Google’s own assistant and comparable tools.

In classic SEO, your position in the results list counts. In AI search, something else counts.

Here it’s about whether the AI names you. And whether it names you in the right context: as a solution, not a footnote.

The technical term for this is GEO (Generative Engine Optimization, optimizing for generative answer engines). SEO gets you into the results list. GEO gets you into the answer.

Before you measure AI attribution: is your baseline even right?

Here’s the uncomfortable part. Most people talk about AI attribution before their normal tracking even runs cleanly.

It’s like building a second floor without ever checking the foundation.

A real example. At a B2B client, I ran a first tracking check. I found three leaks.

• GA4 had counted zero conversions since setup. The conversion event never fired. Every report was built on a zero.
• 98% of LinkedIn clicks landed as “Direct.” The channel that brought the leads was practically invisible in the report.
• HubSpot and GA4 contradicted each other. Two systems, two truths. Nobody knew which one was right.

The reporting was completely skewed. Talking about AI attribution made no sense yet.

The order is clear. First set up the measurement cleanly. Then add the AI sources.

Otherwise you just add a new blind spot to the old ones. And you steer even more off course than before.

Why classic attribution fails here

Your current tools are built on clicks. Google Analytics, UTM parameters, referrer data: they all need a link that someone clicks.

AI search often delivers no click. The user gets the answer right in the chat. They see your name, but they don’t visit your site.

Even when they do click, the source is often missing. Many AI tools don’t pass a clean referrer. The visit lands as “Direct” in your report.

The result: a growing share of your pipeline is invisible. You win leads, but you don’t know where they came from.

How to track your visibility in AI search

Once the measurement runs cleanly, the second step follows. You can’t close this gap with a single tool. You need several signals together. Four ways have proven practical.

#### 1. Watch brand mentions in AI answers

Ask the AI the questions your customers ask. Ask for providers on your topic, your industry, your region.

Note whether your brand shows up. Note the context too, and which competitors it stands next to.

Repeat this regularly. That’s how you see whether your visibility is rising or falling.

#### 2. Ask about first contact in conversation

This is the most direct way, and it costs nothing. Ask every new lead how they found you.

One question is enough: “Where did you first see me?” Answers like “ChatGPT recommended you” are worth their weight in gold.

Build this question firmly into your first call and your contact form. Over a few weeks, a clear picture emerges.

#### 3. Count the rise in direct inquiries with no click path

Look at your “Direct” inquiries over time. Are they rising even though no campaign is running?

A growing base of sourceless inquiries is a signal. Often AI search is behind it, leaving no referrer.

This number alone proves nothing. Together with the answers from your conversations, it gets meaningful.

#### 4. Use specialized monitoring tools

There are tools that query AI answers automatically. For defined questions, they check whether and how your brand is named.

Such tools save manual work and deliver historical data. They’re one building block, not a cure-all.

Important: no tool shows you every mention with total certainty. AI answers vary by user and by moment. Take the numbers as a trend, not an exact measurement.

Why this matters for your B2B pipeline

In B2B, the buying journey is long. Several people, many weeks, many touchpoints.

AI search often sits right at the start of that journey. That’s where the first shortlist of providers takes shape.

If you’re missing from that list, you drop out early. You don’t even notice, because no inquiry comes in.

Measure here and you gain two things. First, you see whether the AI knows you at all. Second, you can improve in a targeted way instead of guessing in the dark.

That’s the difference between guessing and steering. You shift budget where it works, not where the click happens to stay visible.

And that only works on a clean foundation. First the measurement, then the AI sources. In that order.

Next step

Want to know whether your numbers even add up, before you start talking about AI? Then let’s look at your tracking and your AI search visibility together. Honest, no sales pressure.

[Get your tracking and AI search visibility checked](/contact/?ref=ai-search-visibility-track-attribution-for-b2b-in)

Official primary sources and further documentation:

• Google Analytics: Referral Traffic and Source/Medium. support.google.com/analytics/answer/6205762
• Google Search Central: AI Features in Search. developers.google.com/search/docs/appearance/ai-features
• Google Search Central Blog. developers.google.com/search/blog