A guide for established businesses navigating AI search, content quality, and why your experience is suddenly your most valuable asset
Here’s a number worth sitting with: 58% of internet searches now end without a single click. The traffic didn’t disappear, it migrated. And if your marketing strategy was built on capturing that traffic, it’s time to rethink the whole playbook.
But here’s the thing. If you’ve spent years building a genuine reputation, deep expertise, a distinct voice, loyal clients who refer others, the answer isn’t to market faster. It’s to avoid damaging what already works. The risk AI poses to established businesses isn’t being left behind. It’s being lumped in with lower-quality competitors who are flooding the internet with machine-generated noise.
What’s actually gone wrong
Most businesses misunderstand what AI is. Teams have started treating it as a content engine rather than a support tool. Hand the whole process over to a language model and you get what the industry has started calling AI slop: grammatically correct, soulless text that says precisely nothing of value.
This isn’t a moral judgement — it’s a technical one. Large language models are trained to predict the most statistically average next word. Relying on them to write from scratch, on your behalf, is literally instructing your brand to be average. If you’ve spent years developing a strong, opinionated, specialised voice, using AI alone will slowly strip that away until you sound like everyone else.
Google’s spam update reduced “unoriginal content” by 45%. The algorithm doesn’t care how you made your content. It cares whether it adds anything new.
People misread this as Google banning AI-written content. That’s not it. The question isn’t whether you used AI, it’s whether your content adds something the internet didn’t already have. A summary of the top ten search results isn’t useful. A perspective shaped by twenty years of first-hand experience, structured with AI assistance, is.
The search landscape has changed shape
That missing 58% of clicks? It went to AI summaries. Google’s AI Overview sits at the top of results. ChatGPT, Claude, Perplexity, and Gemini have become the first stop for questions that used to generate blue-link traffic. People type a question, get an answer synthesised from across the web, and move on.
This is why you’ll hear people talking about AEO (Answer Engine Optimisation) and GEO (Generative Engine Optimisation). Think of traditional SEO as giving someone a map of a library. AEO is more like having a research assistant read the books and give you the answer. The game continues, but there’s a new level to it: being cited in that AI answer.
AI hasn’t changed the search game, it’s added a level to it.
What search engines actually want: EEAT
Google has been unusually clear about what they’re looking for. It’s summarised in four letters:
E
Experience
Did you actually do the thing? First-hand, lived perspective, the only signal AI genuinely cannot fake.
E
Expertise
Is the information technically correct? Formal credentials, specialised skills, demonstrated knowledge.
A
Authoritativeness
Are you a recognised leader? External validation, links, mentions, citations from trusted sources.
T
Trust
Are you honest and safe? Transparency, accuracy, and consistency across your digital presence.
The first E (Experience) is the most recently added and is arguably the most important, because it’s the one thing AI cannot manufacture. A language model can simulate expertise. It cannot simulate having done the work.
How to respond: part one – content
Pillars and clusters
Content marketing has become more important than ever, not less. The framework that works is pillars and clusters: choose three to five broad topics where you genuinely have depth (your pillars), then build out articles on specific subtopics that link back to each (your clusters). This does two things simultaneously, it helps humans navigate your site, and it signals to AI what you actually do.
When choosing your pillars, think about where you have lived experience, what genuinely serves your audience’s pain points, and whether a topic can sustain at least eight to fifteen sub-articles. Search data should validate the investment, but specificity beats breadth. You don’t need to own a huge topic, you need to own a corner of it.
Search everywhere optimisation
AI doesn’t only train on websites. It scrapes YouTube, social platforms, Amazon, Reddit, and that last one matters more than people realise. Subreddits are where human consensus lives: unfiltered opinions, authentic discussions, real validation. When your brand or product is mentioned positively in those spaces, AI systems treat it as ground truth. You’re not trying to rank on Reddit, you’re trying to create genuine positive sentiment that AI can cite.
Google’s Hidden Gems update is actively elevating authentic, first-hand experience from forums and personal blogs above polished corporate marketing pages. The algorithm is explicitly deprioritising content that reads like it was written by a committee.
Writing from experience
Here’s a practical shift that changes everything: convert “How to” articles into “How we” articles. It sounds small, but it’s psychologically significant. It forces you to write from your own experience rather than generalising, and it produces content that only your brand can tell. An article called “How we handle X for clients in Y sector” is, by definition, irreplaceable.
When using AI to help produce content, work in chunks rather than handing over a full brief. Fact-checking a sprawling AI-generated article is exhausting, and the longer it gets, the more likely you are to let inaccuracies slide. Generate, check, refine, repeat.
How to respond: part two – technical infrastructure
Having a strong human voice is half the battle. The other half is making sure machines can read it. If AI can’t understand your content, you’re invisible to it. And if it summarises you badly, based on incomplete signals, that misrepresentation becomes your de facto positioning.
Schema markup
Schema is a set of structured signals embedded in your website that tells search engines and AI systems what they’re looking at. Think of it as name tags for your content, it removes ambiguity and dramatically reduces the risk of your content being misrepresented or skipped.
Answer-first formatting
Lead every major section with a direct answer of 40 to 60 words before you elaborate. AI systems are looking for clear, citable statements, don’t bury the lead. Be specific about who you serve and what you do. Build individual pages for each service and each target audience. Don’t make the machine guess.
Pricing pages
This one is uncomfortable for a lot of B2B businesses. The usual objections are predictable: every project is unique, competitors will undercut us. But leaving pricing unanswered creates two concrete risks.
Risk one: AI systems will generate a best-guess price based on what they’ve seen elsewhere — which may be completely wrong for your market.
Risk two: If a competitor has published pricing and you haven’t, AI is more likely to send searchers to them. You don’t control the narrative if you’ve left the most important question blank.
Even a starting-from figure with an explanation of what affects the final cost is enough. Give the machines something accurate to work with, or they’ll fill the gap themselves.
Where AI belongs, and where it doesn’t
A wave of new AI marketing tools has arrived, all promising to automate your content. The business model of these platforms requires outputs that work for everyone in an industry, which means template-driven results. The more you rely on them, the more your content converges with every competitor using the same tool. Sameness removes personality. Generic content weakens authority. Over-automation leads, directly, to AI slop.
AI should support your expertise, not replace it. Here’s where it earns its place:
- Idea generation and brainstorming
- Organising messy notes into a first draft
- Repurposing finished articles for social
- Research acceleration
- Feedback on your writing
- Drafting refinement and editing
But always add what AI cannot: experience, opinion, and real insight. The tool is the scaffolding. The building is still yours.
If an AI model were trained exclusively on your brand’s existing digital footprint right now, what kind of personality would it output?
Would it sound like someone with hard-earned experience? Or just another robot?
AI will reshape marketing. It won’t replace trust, authority, or reputation. It will, however, punish sameness, and that’s genuinely good news for businesses who already know who they are. The algorithm is, finally, catching up with quality.
Be as specific as you can about what you do. Give the machines what they need. And if AI is going to summarise you, make sure you’ve given it something worth summarising.
