I've been writing digital due diligence reports for venture and private equity buyers for some years. The shape of those reports was stable for most of that time. Buyers wanted an honest read on the SEO moat, the paid efficiency, the technical debt, the team risk, the brand. We delivered. The numbers held.
In 2025 I started adding a section called "AI search exposure". I added it because some of our agency clients were losing organic traffic to AI Overviews and chat-based search interfaces. By the second half of 2025 it wasn't a curiosity — in some cases it was materially affecting revenue. By early 2026 it had moved from the back of the report to one of the first things I look at.
This post is for buyers — and for the bankers and lawyers who advise them — still treating AI search exposure as a footnote. In a meaningful share of the digital businesses changing hands right now, it's the most important number on the page.
The mechanism, briefly
When a user asks a question in ChatGPT, Perplexity, Copilot or Google's AI Overviews, the model generates an answer. Sometimes it cites and links a source. Sometimes it doesn't. Either way, the user's need is often met without a click.
The further upstream the question is in the buyer's journey, the more likely the model's answer is sufficient. Informational queries — "what is X", "how do I do Y", "compare A and B" — are most exposed. They're also, historically, the queries that drive the bulk of organic traffic for content-led businesses, comparison sites, and many e-commerce category pages.
What this means in practice: businesses whose organic traffic profile leans heavily into informational queries have been losing real volume to AI search over the last twelve to eighteen months. The rate of loss is accelerating. Some categories have seen 20–40% drops in organic clicks at the same impression volume. The traffic isn't going to a competitor. It's not going anywhere. The question got answered before the click.
Why this isn't yet showing up in standard diligence
Three reasons.
Data infrastructure is weak. Search Console tells you about Google clicks. It doesn't tell you that fewer people are clicking because the AI Overview answered the question. You see the symptom — declining click-through rate at the same impression volume — without the diagnosis.
Exposure isn't uniform. A site can have category pages that are fine and content pages getting decimated. The blended traffic numbers wash the signal out. Without segmenting, the picture looks like normal organic volatility.
The most exposed assets look healthiest on trailing data. Content-led, informational-query-heavy assets have been the darling of digital M&A for years. They're also, in 2026, the most structurally vulnerable to the shift in how people search. Standard diligence sees the trailing organic numbers, ticks the box, moves on. The structural risk is invisible to the framework.
How we measure it
We run AI search visibility audits as a parallel exercise to traditional SEO diligence. The mechanics aren't complex but they have to be done deliberately:
Define a representative set of high-intent prompts the target's customers might be using — typically thirty to sixty prompts spanning the buyer journey. Run them through ChatGPT, Perplexity, Google AI Overviews and Copilot. Capture the responses. Classify each: was the target brand mentioned, was a competitor mentioned, was a source cited, was the user's need adequately met without a click.
From that, build a share-of-voice profile and an exposure profile. Share of voice tells the buyer how present the target is in the AI-generated answers their customers are seeing. Exposure tells the buyer what proportion of the target's existing organic traffic is for queries AI is now answering directly.
The exposure number is the one that should keep buyers up at night. We've seen targets whose ostensibly healthy six-figure monthly organic traffic was sitting on a queries portfolio where 60-plus percent of impressions are now subject to direct AI answers. That asset is worth materially less than its trailing numbers suggest.
What buyers should do
In rough order of priority:
Insist AI search exposure is in scope. Not a section your DD provider includes if they happen to have the capability — standard. If your provider can't do it, find one who can, or get a specialist alongside.
Look at the queries portfolio, not just the traffic number. Headline organic traffic tells you what the asset earned in the past. The queries portfolio tells you whether that traffic is structurally durable. A site driving 200,000 monthly clicks from "how to" queries is not the same asset as a site driving 200,000 clicks from "[brand] reviews" or "best [category]".
Quantify the downside. Take the queries you've identified as AI-exposed, model the traffic loss under conservative, central, and aggressive scenarios. Apply the asset's revenue-per-click. The range of outcomes is usually wider than buyers expect — and it should change the price you're willing to pay.
Build the mitigation into the 100-day plan. This isn't a problem you fix with traditional SEO levers. The mitigation requires investment in being a citation source the AI engines will consistently surface — structured content, authoritative claims, brand mention frequency in earned media. The work takes six to twelve months to start paying back. If you're acquiring an exposed asset, build that runway into your value-creation plan from day one.
What sellers should do
Briefer note for the other side of the table. Sellers and their advisors taking a digital asset to market in 2026 should be doing the AI search visibility analysis themselves before the buyer does.
Two scenarios play out very differently:
The seller has done the analysis, knows the exposure, has begun to invest in mitigation, and presents the buyer with a realistic but managed picture of the risk. Price holds.
The seller hasn't done the analysis, the buyer's DD team finds the issue, and the seller is responding to a sophisticated buyer's concerns with surprised explanations during the negotiation window. Price drops, sometimes dramatically.
The asymmetry of preparation is doing real work in deals right now.
The broader point
I keep writing about this not because AI search is the only thing that matters in digital diligence in 2026 — it isn't. The standard areas — paid efficiency, content concentration, technical debt, key-person risk, brand health — all still matter, sometimes more.
But the standard diligence framework was built for a world where Google was the primary mechanism by which buyers reached digital businesses' content. That world is changing materially, and the framework hasn't caught up. Buyers who price the new risk into their decisions will pay more accurate prices. Sellers who acknowledge it will close deals at sensible valuations. The middle ground — pretending the framework still works as it used to — is the most expensive place to be.
If you've got a deal coming up and want a view on AI search exposure as part of the broader DD package, drop me a note. It's the section of every report I write that gets the most attention from investment committees. There's a reason for that.