6 SEO essentials to rethink with search AI

Yes, SEO and AEO have a lot of overlap, but they are not the same. Sticking to your tried and true SEO method will not get you AI search visibility.

So what needs to be changed?

Rather than revisiting content structure in AI search, I’ll focus on three key elements that are important to AI search and three that are less important.

The 3 essentials of SEO are to emphasize

Establish brand authority and strong associations

AI systems need to “know” that your product exists and what it stands for before it can mention you. Business recognition is central to AI visibility in a way that traditional search has never been, although Google’s Knowledge Graph has long highlighted products using this approach. LLM training data rewards brands with consistent, cross-platform presence.

For our agency clients, we focus on making sure product information is consistent across Wikipedia, LinkedIn, Crunchbase, industry directories, and anywhere else LLM might pull business data.

PR and SEO/AEO teams also need to work closely together because earned media buzz is now a hallmark of business building.

The EEAT was already pointing to this, but architects’ associations are more important now. Branded experts with a credible web presence add authority to the content they produce.

If we can invest in business architecture, we often see immediate AI citation benefits when we produce solid content because the infrastructure is already in place.

Build topic depth with content collections

AI systems like sources that show total authority on a topic, not just individual pages that rank for individual keywords. Thin content sequences are more exposed in AI search than they were in traditional search.

Keyword and keyword content planning needs to provide content ownership for the topic. Stop asking, “What do we stand for?” and start asking, “What topics do we want AI programs to cover?”

Internal linking becomes more important because it shows the thematic relationships between the pieces of the LLM import. Content testing should also be a regular activity to identify gaps in content integration, not just pages that aren’t performing well.

If we can go deeper into the niche, we often see the content cited in many related queries. A single strong content collection can generate a wide range of AI visibility.

Managing a topic group in a problem solves the client’s product as an authoritative resource before the sales conversation begins. We’re also getting more feedback that buyers are seeing in LLMs when they’re doing research before buying.

Gain offline brand mentions and social presence

LLMs are trained on the wider web, beyond pages with backlinks. Mentions of your product on Reddit, Quora, a niche forum, or an industry community carry weight even without a link.

AI systems pattern-match what the web says about you from multiple sources, not just ranking on Google. Managed content alone cannot create that signal.

Trusted third-party communities like Reddit carry some weight because LLMs are highly trained in them and often treat their content as authentic user sentiment.

Social engagement and digital PR are now top priorities alongside SEO. Getting your name mentioned in the right places is important, even if the link is attached.

Overseeing non-affiliated brand mentions is as important as tracking backlinks. Tools like Brandwatch and Mement, as well as the manual monitoring of Reddit and Quora, can show where your brand is and not physically.

Talk to your team less about who is connecting with you and more about where your product is being discussed, and whether those conversations are good and accurate.

Brands with an active presence in relevant communities tend to show up naturally in AI responses to conversational, recommendation-style questions, such as “What does Reddit think of X?” or “What is the best Y according to users?”

For challenger companies trying to break into the category, leveraged social media can build AI’s perceived authority faster than traditional link building, which takes time to accumulate.

B2C brands, in particular, benefit from a real social presence. Consumer AI questions tend to lean more toward community testimony and peer recommendations than authoritative publications.

The 3 essentials of SEO are to stress less

Chasing high volume keywords with little content

The AI ​​overview takes clicks to get general knowledge questions. Ranking number 1 for a broad term increasingly means that you have put in a lot of effort to attract traffic that never came.

Volume alone is no longer a proxy for probability. A query with 50,000 monthly searches that triggers AI Overview may bring less traffic than a query with 2,000 searches that does not.

Create specific, authoritative content that answers a small question better than anything else available. Focus on queries where the searcher needs to take action, make a comparison, or access something that only your site offers. Those are hard for AI to fully solve.

Keyword traffic power is no longer the primary metric. First ask if someone will still need to click after the AI ​​answers the question. If the answer is no, the opportunity is not the same as before.

Pursuing similarities and creating a deceptive link

Low quality link volume does nothing for AI citation chances. LLMs measure the authority and relevance of cited sources, not raw link counts. The key publications in AI citations are those with real programming standards, which can be played out the way networks can connect.

Focus on getting coverage and links from outlets where AI systems are actually taken from, such as trade publications, reputable blogs, and near-academic sources. Create content that should be referenced instead of relying on reach designed to generate links.

A hundred low quality links will not get you cited on ChatGPT. Five links from publications that your target audience may read. The important metric is source authority, not link volume.

Optimizing CTR for regular blue links

A growing share of informational questions are solved without a click. As a result, optimizing titles and meta descriptions for CTR in queries dominated by AI Overview provides diminishing returns. The time and resources spent on improving the CTR of zero-click queries could be better spent on earning citations within an AI response.

Be sure to be the source cited within the AI ​​answer rather than the blue link below it. For questions where clicks are still happening, focus on purchase intent and navigation because those are more resistant to AI correction.

CTR optimization assumes that someone chooses between results. For an increased share of questions, that selection is made before the blue links appear. The game went up on the page.

Profit is not always full of traffic

There are many shifts to consider, but these are the first ones I will make. You may lose some volume on traditional SEO metrics like impressions and clicks, but that should have little downstream impact on the most important metrics, including conversions, pipeline, and revenue. That’s an increasingly rewarding AI search.