Digital Marketing Insights

SEO isn’t dead. But the old way of doing SEO is starting to age fast.

Google’s AI Overviews now reach more than 1.5 billion users every month, and one research found that users who see an AI summary click traditional search results only 8% of the time, compared with 15% when no AI summary appears.

Why? Because AI search is already changing how people discover brands, compare options, and decide what to buy. Ethinos observed that this shift is pushing brands to think beyond rankings and traffic, and focus instead on whether AI can actually find, understand, trust, and repeat their content.

So the question is no longer just “How do we rank?” It is “How do we become the answer?”

1. Write for the Way People Prompt AI, Not Just the Way They Google

Traditional SEO was built around keywords, but AI search is built around questions, context, and intent. Ethinos observed that AI does not just read the query at face value. It layers in the user’s intent, persona, location, and source set before deciding what answer to give.

That means “glass skin products” is no longer enough if the real prompt is closer to “best skincare routine for glass skin with sensitive skin and uneven texture.” This is why prompt-style keyword research matters.

Instead of only chasing broad keywords, brands need content built around specific buyer questions, use cases, product comparisons, pricing searches, and problem-based prompts. Research found that58% of U.S. adults in its browsing study ran at least one Google search that produced an AI-generated summary, which means more users are already being served answers before they reach the traditional links. So the job is not just to rank for a keyword. It is to make your content specific enough that AI knows when your brand belongs in the answer.

2. Structure Your Content So AI Can Extract It

Getting picked up by AI is not just about having the right topic. It is also about making the answer easy to pull from the page. Ethinos observed that AI does not work well with long, unstructured blocks of content because it may truncate, skim, or miss the most important points. Instead, content needs clear headers, short sections, schema, FAQs, tables, and direct question-and-answer formatting.

That lines up with Semrush’s analysis of thousands of AI search citations, which found that AI citation behavior does not simply mirror traditional Google ranking patterns. In other words, a page can be “optimized” in the old SEO sense and still be hard for AI to use. Each section should give AI something clean to extract, whether that is a definition, comparison, product detail, recommendation, or direct answer. The easier your page is to parse, the easier it is for AI to quote, summarize, and repeat.

3. Give AI Proof It Can Cross-Check

AI does not recommend brands based on the product page alone. Ethinos observed that AI builds an association graph around who is talking about the brand, where they are saying it, what product they are linking it to, and whether the sentiment is positive. That is why brand mentions only matter when they connect the brand to the actual product category.

A skincare brand does not just need people saying the name. It needs people connecting the name to glass skin, sensitive skin, texture, reviews, routines, ingredients, or whatever buying context the brand wants to own. This is also why platforms like YouTube, Reddit, podcasts, review sites, publishers, and expert content matter. They give AI outside signals it can use to validate the brand instead of relying only on the brand’s own claims.

BrightLocal’s 2026 consumer review research found that 97% of consumers read online reviews, which reinforces how heavily third-party proof shapes trust before purchase. The goal is not to “get more mentions” for the sake of noise. It is to build the right proof in the right places so AI can connect your brand to the answer.

4. Separate Product Facts From Bigger Educational Claims

One mistake brands make is forcing one page to do two jobs: sell the product and explain the whole category. Ethinos observed that product pages should stay clear, factual, and transactional, while deeper educational claims should live on separate, properly supported pages. This matters because Google’s Search Quality Rater Guidelines place extra weight on trust for pages involving health, finance, transactions, product reviews, or advice.

A product page should give AI the hard facts: size, ingredients, format, price, usage, delivery details, and other concrete product information. But if the brand wants to talk about bigger concerns like sensitive skin, uneven texture, barrier support, or glass skin routines, that content needs credible support. That means expert review, references, clear author details, and markup that helps AI understand why the page can be trusted.

The goal is not to separate content for the sake of neatness. It is to stop AI from treating product details, marketing claims, and expert-backed explanations as if they all carry the same weight.

5. Track AI Visibility, Not Just Website Traffic

Traditional SEO reporting makes traffic look like the main win, but AI search changes that equation. Ethinos observed that brands may see fewer clicks because more of the discovery journey is happening inside AI answers. But that does not always mean demand is weaker. In fact, Ethinos observed that AI-referred traffic can convert at nearly 4x higher rates, while some B2B partners saw visibility increase by 4x and conversion impact double even without major traffic growth.

That means brands need to track more than sessions and rankings. Look at whether the brand appears in target AI prompts, whether impressions are increasing in Search Console, whether branded search is growing, and whether direct or branded traffic is converting better. The goal is to stop treating lower clicks as automatic failure. In an AI-led search environment, visibility, recommendation frequency, brand demand, and conversion quality may tell a clearer story than traffic alone.