Is AI Killing Websites? What Actually Changed and What Didn’t

There is a question small business owners are carrying right now, and it usually shows up late at night, somewhere between checking analytics and wondering if the website is still worth the maintenance fee.

If AI answers questions before people click, what is my website for now?

It is a fair question. The headlines have been loud. AI Overviews appear above search results. ChatGPT and other tools hand back full answers without sending anyone anywhere. Traffic charts that used to climb are flat or sliding. Reddit threads pile up with variations of the same fear: Is SEO dead? Is Google quietly killing small businesses? Are AI Overviews the death of comparison sites? Are niche sites finished?

The fear is not imaginary. Something real is happening. But the answer to “did AI kill websites” is not the one the headlines are selling you.

AI did not kill websites. It changed what websites are for.

That distinction matters, because the response to “websites are dead” is panic and a rebuild. The response to “websites have a new job” is clarity and a refactor. Those lead to very different decisions about your time, your money, and what you do next.

This article walks through what the research actually shows, what changed, what did not change, and what it means for the website you already have — especially if it runs on WordPress.

What is actually happening with AI and website traffic

Before getting to the research, one quick definition. AI Overviews are the AI-generated summaries that now appear at the top of many Google search results, giving the user a direct answer before they see the list of websites underneath. They are part of what changed how discovery works, and most of the data in this section is measuring their impact.

The research tells a more nuanced story than the headlines.

Pew Research Center found that users clicked a traditional search result in 8% of visits when an AI summary appeared, compared with 15% when no AI summary appeared. That is a meaningful drop. It is not zero. People still click. They just click less when AI hands them an answer up front.

Ahrefs studied click-through rates more directly. Their 2025 analysis found that AI Overviews reduced clicks to top-ranking informational pages by about 34.5%. Their 2026 update reported that figure had grown to 58%. The trend is intensifying as AI Overviews expand and user behavior adjusts. That deserves attention. It does not deserve a full-scale rebuild of your website.

Ahrefs also noted that 99.2% of the keywords triggering AI Overviews in their dataset were informational in intent. That is a significant detail. The pages most affected are the ones answering simple, general questions — the “what is,” “how to,” and “best of” content. Service pages, contact pages, booking pages, and proof pages are not the primary target.

Then there is the SparkToro and Datos zero-click study, which found that 58.5% of U.S. Google searches and 59.7% of EU searches ended without a click in 2024. That was before AI Overviews became the headline. Most searches were already ending without a click. AI accelerated a shift that was already well underway. It did not invent it.

Put those numbers together and the picture sharpens. People are clicking less when AI answers their question for them. That trend is real and growing. But the websites being affected most are the ones doing work that AI can now do more efficiently — summarizing, comparing, defining, listing.

That is not the same as websites being dead. It is websites being asked to do a different job.

What AI actually changed

Five things shifted, and naming them clearly makes the rest of the conversation easier.

Discovery moved upstream. The old path looked like this: someone searches, scans results, clicks a website, reads, and decides. The new path looks more like this: someone asks an AI tool or sees an AI summary, evaluates that answer, and then visits a website to verify, compare, or take action. Your website may no longer be the first stop. It is increasingly the confirmation stop.

Traffic became a less reliable success metric. When AI summaries absorb the simple informational clicks — definitions, quick comparisons, basic how-tos, “best of” lists, simple cost and timing questions — your traffic can drop even when your website is still useful. The visitors you do get may be further along in their decision than before. Traffic alone never told the whole story. AI just made that obvious.

Websites became source material. AI systems need reliable public information to summarize, cite, compare, and validate. A clear, well-structured website helps both humans and machines understand what your business does. When AI mentions a business or recommends a service, it is pulling from somewhere. If your website is not clear about who you are, what you do, who you help, and where you do it, you are harder to surface and harder to verify.

Generic content got weaker. Thin posts that answer simple questions are more vulnerable now. The content that holds up has specifics: real examples, real experience, real process, real proof, local context, original perspective, and actual decision support. The kind of writing that sounds like it could have come from anywhere is the kind of writing AI replaces most easily.

Validation became more important. People may get a first answer from AI, but they still need to confirm: Is this business real? Can I trust them? Do they serve my area? What do they actually offer? What does it cost? How do I contact them? What happens next? Those questions get answered on a website, not in an AI summary.

What did not change about websites

This is the part that is getting lost in the noise.

People still need to trust you before they hire you. AI can summarize options. It cannot replace the confidence someone needs to hand you money, book a service, or pick you over a competitor. Trust is still built through your About page, your testimonials, your reviews, your case studies, your process pages, and the way your site feels when someone arrives.

Businesses still need a destination. A website is still where contact happens, where booking happens, where service information lives, where proof gets shown, where pricing context appears, where lead capture happens, and where conversion paths exist. None of that moved to AI. AI may point people toward you, but the work of becoming a customer still happens on your site.

Search still depends on websites existing. This is the contrarian point that does not get said enough. AI systems and search engines need source material. If small business owners abandon or neglect their websites, they reduce the pool of reliable information AI has to work with. Your website is not just a destination for humans. It is part of the infrastructure AI uses to learn about and recommend businesses like yours.

Technical fundamentals still matter. Crawlability. Mobile usability. Page speed. Clear headings. Internal links. Accessible content. Structured data. Clean navigation. Clear page purpose. Google’s AI optimization guidance states directly that SEO remains relevant for generative AI search because AI features in Google Search are rooted in its core ranking and quality systems. Google is not telling you to abandon SEO and chase a new playbook. It is telling you that the fundamentals still apply.

Local and service businesses still get validated on the website. AI may help someone discover options, but when a homeowner is choosing a contractor, a parent is choosing a tutor, or a small business is choosing a consultant, they still visit websites. They check credentials. They look at reviews. They read the About page. They look for signs that this is a real business run by real people. None of that goes through an AI summary.

The reframe: your website’s new job

Here is the shift, in plain terms.

The old website model was: search, click, read, decide. The website was a traffic destination, and traffic was the success metric.

The new model looks more like this: AI, search, or referral hands someone an answer or a recommendation. They validate it. They visit your website. They look for trust. They make a decision. They take an action.

Your website is no longer only a traffic destination. It is becoming a source of truth, a trust hub, a validation layer, a conversion path, and a structured signal for the search and AI systems that may surface your business to people who have not heard of you yet.

That is a meaningful change. But it is not a death. It is a job change.

The clearest way to say it: AI may change how people discover you. It does not remove their need to trust you.

That trust still gets built on your website.

What this means for WordPress owners

Most WordPress owners reading this do not need to rebuild their website because of AI. They likely need to refactor and strengthen the website they already have. The bones are usually fine. The job those bones are doing is what needs to change.

After 18 years of helping WordPress owners through WP Sites Success, the pattern I see is consistent — the bones of most sites are fine; the job those bones are doing is what needs to change.

Two terms worth defining before going further. A refactor means strengthening the website you already have — clarifying its pages, improving its structure, updating its content, and tuning the parts that are not working. A rebuild means replacing it — new design, new build, new investment. Refactor before rebuild. That is the right framing for almost every WordPress site I see.

What does refactor actually mean in practice?

It starts with the homepage and top service pages. They should make it obvious what you do, who you help, where you work, why someone should trust you, and what they should do next. Many WordPress sites bury those answers under hero images, vague headlines, and marketing language that could describe anyone. Clarity wins. Specificity wins. Plain language wins.

It includes FAQ sections built from real customer questions — not the generic questions a template suggests. Real buyer questions. Pricing context. Timeline questions. Process questions. The objections people actually raise before they hire you. Answering these clearly helps your buyers and gives AI systems clean material to summarize when someone asks about businesses like yours.

It means strengthening the trust pages. About. Contact. Reviews. Testimonials. Case studies. Process. Credentials. These are the pages people land on when they are deciding whether to hire you. They deserve more care than most sites give them. Google’s E-E-A-T guidance — experience, expertise, authoritativeness, and trust — is not a checklist you optimize for. It is a description of what real trust looks like, and it shows up on these pages or it does not show up at all.

It includes improving how your content is structured. Clear headings. Internal links that connect related pages. Schema and structured data where appropriate. Service information that is easy to parse. Yoast’s guidance on structured data points to the same idea Google’s documentation does: clear structure helps machines understand your content, and that understanding is part of how you stay visible. Many WordPress sites can improve this through plugin configuration and content cleanup, not a full rebuild.

It involves reviewing technical health. Mobile usability. Site speed. Crawlability. Broken links. Indexing. Plugin bloat that slows the site down without adding value. These have always mattered. They matter more now because the cost of a slow, hard-to-parse website is higher when AI systems are competing for attention with your site.

It requires shifting how you measure success. Traffic was never the whole story, and it is even less useful as a single metric now. Track leads. Track inquiries. Track booked calls. Track email signups. Track branded search — people typing your business name into Google. Track conversions. Those numbers tell you whether the website is doing its job. Traffic alone does not.

It means improving existing content before creating more. Most WordPress sites have years of old posts that are thin, dated, or generic. Updating them with examples, perspective, proof, and connection to your actual services is almost always more valuable than publishing new content on top of weak content. Google’s helpful content guidance continues to emphasize originality, specificity, experience, and trust.

And it includes building direct audience paths. An email list. A newsletter. Repeat-visitor paths. Owned communication channels that do not depend on Google, AI, or any other platform deciding to surface you. If AI changes how discovery works, the businesses with direct relationships to their audience are the ones with the most stable ground to stand on.

None of that is a rebuild. All of it is work you can do on the website you already have.

Key takeaways

• AI reduced clicks to some pages, especially basic informational content — but it did not eliminate website visits.
• Zero-click search was already common before AI Overviews became the headline; AI accelerated a trend, it did not create one.
• Websites still matter for trust, validation, conversion, and as source material AI systems read.
• Most WordPress owners need a refactor, not a rebuild.
• Traffic is no longer a reliable single success metric. Leads, inquiries, bookings, and branded search are.

Where this leaves you

If you have been wondering whether your website is still worth the maintenance, the answer is yes — but the question to ask is different now.

The right question is no longer “is my website getting enough traffic.” The right question is “is my website doing the job it needs to do now.”

Does it make clear what you do, who you help, and where you work? Does it answer the questions buyers actually ask? Does it show enough proof and process that someone can trust you? Does it give both people and search systems a clean path to understand your business? Does it convert the visitors who do arrive?

If the answer is yes, you are in better shape than you might think. If the answer is no, the fix is rarely a rebuild. It is usually a clarity audit and a focused refactor.

A clarity audit is a structured review of what your website is saying, who it’s saying it to, and whether it’s doing the job your business needs it to do. That is what a WPSS initial consulting session is for. We sit down with your existing website, look at what it is doing well, identify where it is unclear or working against you, and map out what to refactor before you spend money on anything bigger. No upsell to a rebuild you may not need. Just a clear-eyed look at the website you already have and what it would take to make it do its new job well.

Before you spend money replacing a website that does not need replacing, start there.

Frequently asked questions about AI and websites