How To Know If AI Search Actually Threatens Your Business (A Diagnostic)

I had a VP of Marketing ask me last week if they should "pivot away from evergreen content" because of what's happening with AI search.

I asked her three questions:

1. How much of your traffic comes from informational search queries?

2. Could someone get the same answer from five different sources?

3. Does Google distribution drive conversions, or just vanity metrics?

She paused. "I... don't actually know those numbers."

That's the problem.

Everyone's panicking about publishers losing 27% of their traffic and zero-click searches hitting 69%. But almost nobody is asking the right question:

Does this actually matter to YOUR business?

Here's how to figure that out.

The Data That Matters (And What To Do With It)

Yes, the numbers are dramatic. Zero-click searches jumped from 56% to 69%. When AI Overviews appear, click-through rates drop from 15% to 8%. Only 1% of users click source links inside AI summaries.

Traffic to the top 500 publishers dropped 27% year-over-year. Chegg lost 49% of its traffic. The Planet D shut down after losing 90%.

But here's what nobody's asking:

What kind of traffic are they losing?

Because that's the diagnostic question that determines whether you should care.

Look at who got hit:

• Chegg - Informational queries about homework problems. Zero commercial intent until conversion.

• The Planet D - Informational queries about travel destinations. Long consideration cycle.

• CBS News - 75% of their top keywords now generate zero clicks with AI Overviews. These are news queries, not purchase-intent queries.

Notice the pattern? High-volume informational queries with no immediate commercial intent.

Now here's the diagnostic framework you need to run on your own business.

The Diagnostic: Five Questions To Assess Your Risk

Run these numbers for your own business. If you can't answer them, get the data first before you make any strategic decisions.

Question 1: What percentage of your organic traffic comes from informational vs. commercial queries?

• Informational = "how to," "what is," "best practices," definitions, guides Commercial = product comparisons, "buy," "pricing," "vs," vendor searches

• If 70%+ of your traffic is informational queries that could be answered in a paragraph, you're exposed. If most of your traffic is late-stage commercial intent, you're probably fine.

Question 2: How unique is your answer compared to what's available elsewhere?

• Go to Google. Search for your top 5 keywords. Look at positions 1-10.

• If you could write your article by reading those ten results and synthesizing them, AI can too.

• If your content includes proprietary data, specific customer examples, or methodologies that don't exist elsewhere, AI can only cite you, not replace you.

Question 3: What's the conversion path from organic search traffic?

• If your model is: search → article → newsletter signup → nurture sequence → conversion... and the conversion cycle is 6+ months, you're vulnerable. Traffic disruption hits you hard.

• If your model is: search → pricing page → demo request → close in 30 days, you're less exposed. The traffic coming to you already has intent. AI Overviews don't answer "show me pricing for enterprise software."

Question 4: Is your content strategy built on distribution or differentiation?

• Be honest. Are you ranking because you have genuinely unique insights, or because you executed SEO tactics well?

• If you took away Google distribution, would people still seek out your content specifically? If no, that's a distribution dependency—and it was always a risk.

Question 5: What percentage of your pipeline actually comes from organic search?

• Not "traffic" - pipeline. Qualified opportunities. Revenue.

• If organic search drives 5% of your pipeline and you're freaking out about AI Overviews, you're optimizing for the wrong thing. If it drives 60% of your pipeline, now we're talking about real business risk.

How To Interpret Your Results

Based on your answers, here's how to think about your exposure:

High Risk Profile:

• 70%+ informational query traffic

• Content that synthesizes publicly available information

• Long conversion cycles dependent on sustained traffic

• Distribution-based content strategy

• 40%+ of pipeline from organic search

If this is you, yes, you need to act. But not by "pivoting to additive content" or whatever the trend piece told you. You need to fix the fundamental business model issue: you're dependent on a distribution channel you don't control.

Medium Risk Profile:

• 40-70% informational traffic

• Mix of unique and commoditized content

• Moderate conversion cycles

• Some differentiation, some distribution dependency

• 15-40% of pipeline from organic search

You have time, but you need to be strategic. Figure out which content assets actually drive revenue and which are vanity metrics. Double down on the former, automate or sunset the latter.

Low Risk Profile:

• High commercial intent traffic

• Proprietary insights or data

• Short conversion cycles

• People seek you out specifically

• <15% of pipeline from organic search, or organic traffic that converts at high rates

You're probably fine. Monitor the situation, but don't overreact to industry panic. Your business model isn't built on capturing informational search traffic.

Here's the part nobody wants to hear: 

If you're high risk, the problem isn't AI Overviews. The problem is you built a business model on rented land. AI just made the landlord raise the rent.

What To Do Based On Your Diagnostic

Your next moves depend entirely on what your diagnostic revealed. Stop following generic advice.

If You're High Risk:

• Your priority isn't "shifting to additive content." It's reducing dependency on a channel you don't control.

• Start building owned audiences. Email lists. Slack communities. Podcasts. Anything where you have the relationship directly.

• Audit which content actually drives pipeline vs. just traffic. Kill or automate everything that doesn't convert.

• If you have unique data or insights, extract them into standalone research reports that can't be easily replicated. This is what companies like Gartner, Forrester, and G2 have always done—they're not worried about AI Overviews because their value is in proprietary data, not information synthesis.

• Consider whether you're in the media business or the software business. If you're software pretending to be media (content marketing as your primary channel), that's the deeper problem to fix.

If You're Medium Risk:

• You need to be strategic, not reactionary.

• Run a content audit segmented by: (1) traffic volume, (2) conversion rate, (3) replicability.

• Keep the high-converting unique content. Sunset the high-volume, low-converting, easily-replicated content. Stop feeding the algorithm with content that doesn't serve your business.

• Invest in depth over breadth. One genuinely differentiated piece of content is worth more than ten "how-to" guides that say the same thing as everyone else.

• Track new metrics: LLM citations, Share of Voice in AI responses, brand mentions in ChatGPT. But don't obsess over them. Track them alongside pipeline contribution.

If You're Low Risk:

• Don't panic just because everyone else is.

• Monitor the situation. Set up alerts for traffic changes. But don't overhaul a working strategy because of industry trend pieces.

• The mistake would be abandoning what's working to chase what's trending.

Universal Truth Regardless of Risk Level:

• If your content strategy relies on someone else's algorithm staying favorable to you, that was always a vulnerability. AI didn't create this problem. It exposed it.

• The question isn't "what's happening with AI search?" The question is "do I have a defensible business model?"

The Bottom Line

When people ask me "should I worry about AI search," I ask them: "Do you know what percentage of your pipeline comes from organic search?" Most don't.

They know traffic numbers. They know rankings. They know click-through rates. But they don't know if any of it actually matters to their business.

That's the diagnostic gap.

The marketers who are going to win aren't the ones following the loudest trend pieces. They're the ones who actually understand their own business model.

They know the difference between traffic and revenue. They know whether their content is defensible or commoditized. They know if they're building on owned land or rented land.

The data about zero-click searches and publisher traffic losses is real. But whether it matters to YOUR business is a different question entirely.

Run the diagnostic. Answer the five questions honestly. Then act based on your actual exposure, not on industry panic.

Because the only thing worse than being disrupted by AI search is pivoting your entire strategy based on someone else's disruption.

Sources:

1. Similarweb data on zero-click search trends and publisher traffic (May 2024-2025)

2. Pew Research Center study on AI Overviews impact on click-through rates (March 2025)

3. Digital Content Next survey on publisher referral traffic losses (May-June 2025)

4. Chegg company statements and antitrust lawsuit filings (February 2025)

5. BrightEdge research on click-through rate changes with AI Overviews