For years, the answer to online visibility was relatively straightforward: if you wanted more people to find your website, you invested in SEO.
Those early days were pretty simple, because there was a list of tactics that worked and the result was largely decided by who was able to best execute them. SEO itself wasn’t simple, but we understood how to do it well. Once Google’s algorithm began to change from something designed to return the single best result to something designed to return the single best result for you in particular, it complicated into required technical foundations, useful content, keyword research, website performance, link authority, strong product or service pages, and a clear understanding of your intent as a user. But the search route we were optimising for was still familiar: people searched on Google, Google displayed a list of results, and your job was to appear as prominently as possible in those results.
As more people are using AI tools to research products, compare providers, summarise options, and ask questions that they might once have typed into a search engine, that’s becoming just one route among many. Google itself is also becoming more AI-led, with AI-generated answers appearing above, alongside, or instead of traditional search results for certain queries.
That raises understandable questions for businesses: do you optimise for SEO, or should you now be optimising for AI? Does AI optimisation look different, and if it does, what are the differences?
The answer is that the distinction is both more and less than you think. Each AI platform has unique preferences, and so the days of one route to optimise are dead – recent research has found that some of Google’s preferences directly contrast what ChatGPT looks for when deciding to cite a page. Your first job now starts with weighing up where you want to be seen, and deciding which bits of optimisation to carry out.
What do I recommend as someone who works in both SEO and AIO?
I think that it makes sense to carry out SEO first, but with an eye on the places where it crosses over. AI optimisation isn’t really a replacement for SEO, because so much of the information used to judge your business is sourced from your website by AI. If your website is technically weak, unclear, thin on useful content, or difficult for search engines to understand, it isn’t going to perform well in either traditional search or AI-led discovery.
In other words, before you worry about being recommended by AI, you need to make sure your website deserves to be found in the first place.
I’m busy, I’ll read about this later
Bad idea. If you’re a regular Google user, you’ve probably noticed that recently, you’re often served an AI answer before you get to the search results. It invites you to ask a question, and if you do, it turns into less of a browsing session and more of a conversation. If you’ve been shopping for goods or services online, that AI Overview might have included product recommendation.
Users are learning not to always expect a list of links. Increasingly, they’re spoon fed direct answers, summaries, comparisons, recommendations, and next steps. All the information needed to make a great buying decision used to be available if you were motivated to look, but now it’s distilled into an easy to read couple of paragraphs by a single press of the “enter” key. Someone looking for a local service provider might ask an AI assistant to shortlist suitable companies. A buyer researching a technical product might ask for the pros and cons of different options, and a customer planning a larger purchase might use AI to narrow down choices before ever visiting a website. I know I’ve already done at least two of those three.
None of that means traditional search is disappearing. Google remains central to how people navigate the web, and organic visibility is both commercially valuable for most businesses and (in some cases) connected to your brand’s likelihood to surface in an AI answer. The change is that search experience is becoming more of a blended medium. Users don’t just stay on Google search but will also move across AI Overviews, ChatGPT, Perplexity, social platforms, YouTube, review sites, maps, marketplaces, and brand websites before making a decision.
That means optimisation isn’t as simple as making some page adjustments and ranking for keywords that automatically bring visitors to your site. Visibility now depends on the holistic elements that make up your brand, and its effect on your presence online.
This is where some businesses risk jumping too far ahead. They’ve heard about AI search and assumed they need a completely new strategy, when in reality, many of the fundamentals haven’t changed.
AI tools still need information to work with, which means crawlable websites, trusted sources, clear explanations, consistent entities, structured content, third-party validation, and pages that answer real questions well. Those aren’t separate from SEO (well, not completely – like I said, it’s getting complicated).
SEO first, because AI still depends on the web
AI tools may feel very different from search engines, but that’s mostly the interface at work. Behind the scenes they are still heavily influenced by the information available online, and that’s what makes ignoring SEO an unfeasible growth tactic. A Large Language Model like ChatGPT is trained on an existing dataset. That dataset is vast, but it is dated to the moment it was retrieved and fed to ChatGPT. If that were the only data it had access to and the cutoff was 26 July 2025, asking a question about something that happened after that date would either generate no answer at all, or worse, a hallucination that gives you the wrong answer, but with absolute confidence.
That’s why it also has access to web search – it can retrieve the latest information that way, and fact check itself. The chance of a hallucinatory answer is reduced because there’s fresh info on tap, and the end user looking for reliable information on something goes away happy. That also increases the spectrum of information available on your business to everything online that can be crawled by a search robot. The more there is available, the more complete the picture of your brand, but there will always be some bits you wish were more available than others.
If your business has a weak digital footprint, AI systems have less to draw on. If your website is technically poor and/or light on relevant information, it becomes harder for both search engines and AI tools to understand what you offer and why you should be recommended. If your brand is barely mentioned outside your own website, there is less evidence of authority, relevance, or trust.
That’s all in scope for SEO, and why I think it makes the best starting point.
A strong SEO foundation helps you build the assets that both search engines and AI systems need. These include:
- Clear service, product, category, and landing pages
- Helpful content that answers real customer questions
- A technically sound website that can be crawled and indexed
- Structured information that makes your business easier to understand
- Strong internal linking between related pages
- Evidence of expertise and credibility
- Consistent brand information across the web
- Reviews, mentions, links, and citations from relevant third-party sources
Many of these aren’t on 2012’s list of SEO essentials, because they’re representative of the fuller picture that your digital presence is now built on.
Businesses that skip these steps and try to “optimise for AI” in isolation are likely to struggle, because they’ll most likely end up only optimising for one piece of the puzzle. Also, a lot of AI optimisation advice is still pretty sketchy. We know it can be misled because it’s already being used by criminals to defraud people, but executing the same move without resorting to the kind of cheap tactics that might get you banned later is not yet on the table.
The honest way to do it is to accept that before you can become the answer, you need to have answers worth finding.
What is AI optimisation, really?
AI optimisation is sometimes also called GEO, or generative engine optimisation. The term is useful, but it makes the discipline sound more separate from SEO than it really is.
At its simplest, AI optimisation means making your business more likely to be understood, cited, summarised, or recommended by AI-led discovery tools. It’s important to say “business” and not “website” here
That might include optimising for Google’s AI features, ChatGPT search, Perplexity, Gemini, Copilot, and other platforms that generate answers rather than simply displaying links.
The important point is that done right, AI optimisation is not about tricking AI systems. It is about making your information clear, trustworthy, well-evidenced, and easy to interpret.
If I’m working on a traditional SEO approach I might ask:
“What keywords should this page rank for?”
An AI-aware approach means I also need to ask:
“What questions should this page answer clearly?”
“What evidence proves this business is credible?”
“Would an AI assistant understand who this company helps?”
“Is the information specific enough to be included in a useful recommendation?”
“Are the claims supported by details, examples, reviews, case studies, or third-party references?”
That shift is because AI-led discovery often rewards clarity and usefulness rather than keyword repetition. A page that targets a phrase won’t be enough, as the content needs to explain, compare, evidence, and guide.
The danger of treating AI as a shortcut
Because AI search is new and fast-moving, there’s a temptation to look for quick wins. The bad advice already ranges from creating hundreds of AI-generated pages, to flooding the web with content, to using a checklist of “AI ranking factors” to appear in answers.
That’s straight out of the early days of SEO, and it’s as risky now as it was back then. It’s also a little dumber, because we have the case of SEO to show why it was a bad tactic.
Search engines have spent years trying to reduce low-quality content created purely to manipulate rankings. AI systems face the same challenge, especially as the web becomes increasingly filled with generic AI-generated material. If everyone is publishing similar AI-written content, originality, expertise, and trust become more important, because they’re the bit that AI can’t write. Not honestly, anyway.
This is particularly relevant for businesses in competitive or trust-sensitive sectors. A generic article that says broadly the same thing as every other website is unlikely to make a brand stand out. A page that gives specific advice, shows real expertise, answers nuanced questions, and reflects genuine commercial knowledge is much more valuable.
AI can be useful in the content process, but it can’t replace expertise. It can help structure ideas, identify gaps, summarise research, or produce first drafts, but your final content needs to be shaped by a combination of human knowledge, brand positioning, and customer understanding.
The winning approach isn’t having the most content, it’s creating content that is clearer, deeper, more specific, and more useful.
What should businesses optimise first?
If you are deciding where to focus, the best approach is to build in layers.
- Start with technical SEO
Your website needs to be accessible, crawlable, fast, secure, and easy to navigate. If search engines can’t properly access or understand your pages, AI tools probably can’t either.
This includes making sure important pages are indexable, page speed is good, any redirects are managed properly, duplicate content is controlled, and the site structure makes sense. If you’re not sure, PageSpeed Insights and Search Console are two free Google tools that can give you a read on your site’s technical performance.
Technical SEO isn’t glamorous, but it underpins everything else so get it done first and you might save money later.
- Strengthen your core commercial pages
Before publishing large volumes of blog content, make sure your main service, product, category, and location pages are doing their job.
These pages should clearly explain what you offer, who it is for, why it matters, and why a customer should choose you over everyone else who is selling the same thing. Answer the questions a buyer genuinely needs resolved before making contact or purchasing and you can handle those objections before they even speak to you.
For eCommerce websites, I’d recommend include improving category copy, product descriptions, filters, FAQs, buying guides, and internal links. For service businesses, it may involve clearer service pages, sector pages, case studies, testimonials, and location-specific content. The latter can be tough if you’re in professional services and need to balance internal stakeholder approval with a practical approach to content, but it will be worth it if you can persuade them to get on board.
- Build helpful supporting content
Once your commercial pages are strong, supporting content can help capture earlier-stage searches and AI-led research queries.
This might include guides, comparisons, explainers, FAQs, glossaries, opinion pieces, and advice-led articles. You don’t want to publish content for the sake of it, but to build topical authority around the subjects your customers care about.
For example, a business should not only have a page selling a service. It should also have content that explains when that service is needed, how to choose a provider, what mistakes to avoid, what costs to consider, and what outcomes to expect.
This kind of content is useful for traditional SEO and well-suited to AI discovery, because it gives both search engines and AI systems richer context. One note is that a lot of businesses have been using listicles (articles that are built around a list) to declare themselves the number one business in their field. This has led to some reasonable success with AI repeating the claim to users, but it’s now well-known enough that it’s highly likely to be one of the first things that gets punished as a “cheat” tactic.
- Make your expertise visible
AI tools need signals of trust, and so do your customers.
Your website should make it clear why your business is credible. That includes a combination of author information, accreditations, awards, reviews, case studies, client examples, media mentions, detailed product knowledge, and transparent processes.
For many businesses, this is an underused opportunity. They have genuine expertise, but their website does not show it clearly enough.
It’s something that many people struggle with, but you can’t assume people will know why you are good. Explain it, provide the evidence to prove it, and make it easy for both users and algorithms to understand.
- Improve your wider brand footprint
AI discovery isn’t limited to your website. Mentions across the web can influence how your brand is understood.
That means considering building your presence using digital PR, relevant directory listings, reviews, partner websites, industry publications, social profiles, your Google Business Profile, and other third-party sources.
This isn’t link building in the old SEO sense, it’s about building a consistent, trustworthy digital presence.
If your business is mentioned positively and consistently across reputable sources, it gives both users and AI systems more confidence in who you are and what you do. My top tip for this overall area is to look at Reddit, which has come up again and again as a place that brands need to be seen if they want to build a positive AI presence.
So when should you focus specifically on AI optimisation?
Once your SEO foundations are in place, it makes sense to look more deliberately at AI visibility.
That would include reviewing how your brand appears in AI-generated answers, checking whether your key topics are being summarised accurately, identifying which competitors are being mentioned, and looking for gaps in your own content.
Doing that isn’t always easy as AI will generate a new response with each ask, but there are tools which can help you to build a picture.
You can also adapt your content structure to make it easier for AI tools to interpret. Clear headings, concise summaries, direct answers, comparison tables, FAQs, schema markup, and well-organised topic clusters can all help.
The most important thing to consider is that these changes should support human users first. The best SEO and AI optimisation naturally improves the website for real customers too. If a page is easier for AI to summarise because it is clearer, better structured, and more informative, it’s probably easier for a potential customer to use as well.
SEO and AI are not opposing strategies
The phrase “SEO or AI” suggests a choice, but for most businesses, the two are connected.
SEO helps your website become visible, useful, trusted, and technically sound. AI optimisation builds on that by making your information easier for answer engines and AI assistants to understand, cite, and recommend.
The businesses most likely to benefit from AI search will not be the ones that evolve their SEO instead of abandoning it. .
That means moving beyond narrow keyword targeting and thinking more broadly about discoverability, building a brand presence that is consistent across the wider web.
In short, optimise for people first, search engines second, and AI as an increasingly important extension of both.
The practical answer: optimise for SEO first, but do it with AI in mind
If your website is already performing well organically, boosting your AI optimisation should be part of your next stage of development. It will help you understand how discovery is changing and where your brand may need stronger signals.
But if your SEO foundations are weak, that needs to be the first stop on your roadmap. A technically sound website, strong commercial pages, useful content, and clear evidence of expertise will support both traditional search performance and future AI visibility. Without those foundations, AI optimisation risks becoming a distraction.
The question isn’t really about whether SEO or AI matters more. What you really need to ask is “how do we build a website and digital presence that can be found, trusted, and recommended wherever customers are searching”?
For most businesses, that answer starts with SEO – and grows from there.
Need help with your AI or SEO strategy? Get in touch
If you’re in need of help with optimising your website and honing your online presence, just get in touch.
