You Don't Build Authority When You Suddenly Need It

SEO still sends far more website traffic than AI does for most businesses, and I don’t think that’s worth disputing.

But traffic volume may be the wrong way to judge what’s actually changing.

Someone arriving from traditional search could be at almost any stage of their thinking. Someone arriving after an extended conversation with an AI assistant may already have researched the market, compared alternatives, and narrowed the field before your website ever enters the picture.

The visitor count may be smaller. The significance of the visit may not be.

And there’s another complication: AI can influence the decision to visit a business without ever receiving attribution for that visit at all.

The Part the Traffic Numbers Miss

• SEO still drives far more traffic than AI for most businesses, and pretending otherwise doesn’t help anyone.
• Traffic volume alone misses something important: AI-influenced visitors can arrive much later in their decision process than search visitors do.
• It also misses attribution: someone can be influenced by an AI conversation and still show up in analytics as an ordinary Google or direct visit, so AI’s actual influence may be larger than any traffic report shows.
• Waiting until AI-assisted discovery is large enough to notice in your analytics means starting the real work, building authority, evidence, and clarity, only once it’s already needed.
• The economics favor some businesses more than others: high-ticket, low-frequency purchases need fewer AI-influenced opportunities before the investment becomes economically meaningful.
• None of this is an argument against SEO. It’s an argument for paying attention to a channel whose real influence may be larger than its traffic numbers suggest.

The Traffic Numbers Aren’t Wrong

I want to be clear about something before making the rest of this argument: organic search is still the dominant channel for most businesses I talk to, by a wide margin. If someone tells you AI has already overtaken search as a traffic source, they’re probably talking about a narrow slice of a specific industry, not the general picture. I’m not interested in dressing up a smaller trend as a bigger one.

But I’ve started to think traffic share is the wrong lens for deciding whether this matters yet.

What the Traffic Numbers Don’t Show

Picture two visitors landing on the same page. One typed “wealth manager near me” into Google fifteen seconds ago. They might be starting a serious search, or they might be years away from actually switching advisors and just curious what’s out there. There’s no way to tell from the click alone.

The other visitor spent twenty minutes in a conversation with an AI assistant, asking about fee structures, comparing a handful of local firms, reading about investment philosophy and credentials, and narrowing the list down to two names. Your firm is one of them. By the time they land on your site, they’ve already done a version of the work a search visitor might spread across several separate sessions, and they’ve already formed an opinion about where you stand.

Those two visitors show up identically in a traffic report. They are not remotely the same visitor.

There’s a further wrinkle worth naming. That wealth-manager visitor might never show up as an AI referral at all. They could close the AI conversation, open a new tab, search your firm’s name directly, and arrive looking like ordinary organic or direct traffic. Analytics can tell you how someone arrived at your website. It can’t necessarily tell you what convinced them your website was worth arriving at. That’s the difference between attribution and influence, and it means AI’s actual role in a decision can be larger than any traffic report will ever show.

This is also where the math starts to favor certain businesses more than others. Building the kind of authority AI systems check for takes real, sustained work, and the traffic it sends is comparatively small. For a business built on high volume and thin margins, that trade can be hard to justify against the numbers alone. For a business built on a handful of high-value relationships, a wealth manager, a luxury real estate agent, an M&A advisor, the math changes: the lower the frequency and the higher the value of the decision, the fewer AI-influenced opportunities it may take for the investment to become economically meaningful. A luxury agent doesn’t need hundreds of AI-influenced leads to make the economics work. One valuable listing opportunity can materially change the calculation.

The Trap Hiding in “Not Yet”

Here’s the part I think gets missed. The natural response to “AI traffic is still small” is to wait: revisit this once the numbers move, once it shows up clearly on the dashboard, once it’s obviously worth the effort.

The problem is what has to exist before an AI system has good reason to recognize or recommend a business at all: a clear, consistent identity, credible corroboration, demonstrated expertise, and a track record that holds up when examined. None of that can simply be switched on when demand arrives.

Authority is closer to reputation than a landing page. It accumulates over time.

That’s the trap. If you wait until AI-assisted discovery is large enough to justify the attention, you may be choosing to begin building something inherently slow at precisely the moment you need it quickly.

Why This Isn’t a Case Against SEO

I want to head off two opposite misreadings here.

The first is “AI has replaced search.” It hasn’t. Search still sends the traffic, still converts, and is still where most of the marketing budget probably belongs today.

The second is “AI traffic is too small to matter yet.” That conclusion doesn’t actually follow from the traffic numbers alone, for the attribution reasons already covered above: influence and traffic share aren’t the same measurement.

My actual position sits between those two. SEO matters today. AI-assisted discovery is still developing. Its influence may be under-attributed, its visitors are sometimes unusually far along already, and the authority required to compete for them can’t be manufactured overnight. Waiting for the traffic graph to prove the case may be exactly the wrong time to start.

What Building Authority Before You Need It Actually Looks Like

In practice, this isn’t dramatic. It looks like making sure your business is described the same way everywhere it appears online. It looks like having real, checkable evidence behind the claims on your site instead of just confident language. It looks like the kind of groundwork that feels almost too boring to prioritize, right up until the day it’s the only thing standing between you and being left off a shortlist you never even knew existed.

None of it happens overnight, which is exactly the point.