A year ago, the scariest thing that could happen to a small business online was slipping from the first page of Google to the second. Now it's something else entirely: Google answers the question before anyone ever clicks your website.
That answer box at the top of the results β the one labeled "AI Overview" that sums up an answer in a few sentences β is what everyone is panicking about. And the panic makes sense. If Google hands your customer the answer, why would they visit your site at all?
I do SEO (search engine optimization β the work of getting your business found on Google) for clients across the US, and I get this exact question every week now. So let me tell you what I'm actually seeing. The businesses losing traffic and the businesses growing aren't separated by budget or luck. They're separated by one decision about what kind of content they publish.
What AI Overviews actually changed (and what they didn't)
Here's the part the panic skips over. AI Overviews are good at one thing: answering simple, factual questions. "How many ounces in a cup." "What time does the post office close." "What's the capital of Ohio."
If your website traffic came from questions like those, yes β that traffic is shrinking, and it isn't coming back.
But that traffic was never worth much. Nobody who Googles "how many ounces in a cup" was about to hire you. They wanted a number, they got a number, they left.
What AI Overviews are bad at β and this is the whole point β is the messy, specific, "it depends on my situation" questions people ask right before they spend money. Google won't confidently answer "is a med spa franchise worth it in Scottsdale" or "should I replace or repair a 12-year-old commercial AC unit." Those answers depend on context, judgment, and trust. That's where real customers live.
Why keyword-chasing SEO is suddenly the losing move
For fifteen years, SEO advice was basically: find the words people type into Google, then build a page stuffed with those words. Rank for the word, get the traffic.
That's the game AI Overviews are quietly ending. When a search is a simple keyword with a simple answer, Google now answers it itself, and the page you built to "rank for that keyword" gets skipped.
So businesses still optimizing page after page around single keywords are, without realizing it, building exactly the content the answer box replaces. They're running faster on a treadmill that keeps speeding up.
The fix isn't to write more keyword pages. It's to stop competing with the answer box on its own turf.
What the businesses that are thriving do instead
The clients of mine who are growing right now all made the same shift: they moved from writing for keywords to writing for intent β the actual reason behind the search.
In plain terms: instead of a thin page targeting "best running shoes," they publish the page a real runner needs β "I have flat feet and knee pain, which running shoes actually help" β with specifics, trade-offs, and a clear recommendation. Google's AI box can't replace that, because that's not a fact. It's judgment.
Intent-focused content wins right now for three reasons:
If this is the shift you know your content needs but you don't have time to make, it's exactly the kind of work I do β let's talk about your site. I'd rather tell you honestly whether it's worth it than sell you ten pages you don't need.
Here's the difference laid out plainly:
| The old way (keyword-focused) | The new way (intent-focused) | Which one survives AI Overviews |
|---|---|---|
| A page targeting "accountant near me" | A page answering "do I need an accountant, or just better software, for my one-person business" | Intent β Google won't answer this one for you |
| 20 thin pages, one keyword each | 6 deep pages built around real customer decisions | Intent β fewer pages, far more trust |
| Written to please a search engine | Written to genuinely help a buyer | Intent β and it reads better to humans too |
A real example: an accounting firm in Charlotte
Last quarter, a small accounting firm in Charlotte, North Carolina came to me worried. Their blog traffic had dropped about 30% over six months, and their previous SEO provider's only plan was "publish more keyword posts."
We did the opposite. We retired 14 thin posts that all targeted slight variations of "Charlotte accountant," and replaced them with five honest, specific guides β things like "S-corp or LLC: which one actually saves you money in your first profitable year" and "What a bookkeeper does that QuickBooks (the accounting software) doesn't."
The whole project was about $3,800 over five weeks. Within three months their organic traffic wasn't just back β it was up 22% over where it started, and two of those five guides began showing up cited inside Google's AI Overview for their topics. Better still, the contact-form inquiries improved: fewer tire-kickers, more "I read your S-corp guide, can we talk." That's the whole game.
How to make your content AI-proof, starting this week
You don't need a big budget or a rebuild to start. You need a different question driving your content. Here's where I'd begin:
None of this requires you to understand how Google works. It requires you to know your customer better than a robot can β which, as the actual expert in your field, you already do.
The Bottom Line
AI Overviews aren't killing SEO β they're killing lazy SEO. The businesses panicking built their traffic on questions a machine can now answer in one sentence; the businesses thriving moved their content to the specific, judgment-heavy questions only a real expert answers well.
Make that shift and you stop fearing the answer box β you start getting cited in it. If you want a straight answer on whether your content is built for the old game or the new one, tell me about your business and I'll take a look.
