Now, I don’t talk about AI here very often. As many of you will know, I am staunchly anti-AI in all areas of my work. I am that insufferable person who adds -ai to my Google searches to stop the search engine needlessly consuming a shit ton of water just to provide partially hallucinated summaries of sort of relevant content. The only time I engage with them is when I’m doing SERPs and performance analysis for my SEO clients.
However, off the back of my recent visit to (and time on stage at) BrightonSEO, we need to talk about it.
I am also aware that the done thing in SEO now is to rebrand it into GEO or AIEO or whatever else grifters are calling it these days. The problem is, and say it with me now, the vast majority of “GEO” is just long standing SEO best practices packaged up as something new so people can charge more for it.
Don’t get me wrong, I do believe there are some specific elements of technical SEO that can support LLM citations, but certainly when it comes to content nothing has really changed.
Debunking some basic AI-SEO/GEO content “cheats” that are actually just SEO best practices
“Content chunking” → breaking content up into smaller, more readable chunks with clear, intent-driven subheadings. This has been well known UX and SEO best practice for years. I have a whole bunch of blogs about content design and UX if you’d like to learn more.
“Intent-first content” → so this one is a little murkier, because SEO has historically been laser focused on keywords for a long time. Which is why lazy or grifty SEOs will stuff their content chock full of keywords that make no grammatical sense. But the real ones have been focusing on intent-led content for years now. It’s not new.
“Topic clusters” → recently I’ve seen this one bounded about as if it’s anything new. Topic clusters have been a key part of many SEO content strategies for well over a decade to capture rankings and traffic for multiple different types of keywords around the same topic. Now, GEO “experts” tout this as way to focus on AI’s intent-based search model. As if we haven’t been doing this all along
“Updating content with new publish dates” → apparently AI loves recently published content. Cool, I won’t argue with that. But regularly updating your SEO content has also been best practice for well over a decade. In fact, I wrote this blog about updating and re-optimising old content back in 2023 and have been updating it at least once a year since.
So next time someone tells you they are a GEO specialist, seriously consider how much of what they tell you is actually new. And if someone tells you ChatGPT is the future of search, send them this blog from resident “fuck this shit” SEO expert, Nikki Pilkington.
5 reasons why AI isn’t going to save your business, your marketing, or your SEO
- AI-generated content is generic, unhelpful, and inaccurate
- Replacing juniors with AI is a long term, seriously alarming problem
- Generative AI cannot empathise with or fix human problems
- Trust and authenticity are becoming more and more important
- Building authority is next to impossible when none of the work comes from you
And if all of that wasn’t enough, there is also the fact that generative AI models were trained on stolen intellectual property and every time we use it we are causing irreparable damage to the environment.
1. AI-generated content is generic, unhelpful, and inaccurate
When I was on the judging panel for the European Search Awards this year, it was really disheartening to see how many companies put their projects forward as “creative” and “innovative” because they used generative AI on a large scale to deal with the admin of writing content. On scales that were not fact checked or edited. Just generated and spaffed out onto the world wide web.
I’m sure you’ve heard the term AI slop by now. And it gets the name slop for a reason. It’s a regurgitated, bland, personalityless, mishmash of everything that came before. Yes, you can spend hours or days or even months training an AI to sound like you. Or you could just use the time, money, and resources to create the content yourself. Even better, hire a dedicated content person to do it.
We also can’t ignore the high hallucination rates. Factual inaccuracies and coding bugs that you may not even know are there. Things that proliferate misinformation or break websites or apps. All because someone didn’t want to learn a new skill or hire a human being to create something for them.
2. Replacing juniors with AI has long term consequences
If you replace your juniors (whether that’s your marketing team, developers, or anyone else) with AI now, who will you have in 5, 10, or 15 years time to debug, decode, and correct these models in the future.
Or perhaps if this whole AI thing really is just a bubble inflated by billionaires (one can only hope), who will you have to lead teams, manage departments, and come up with strategies.
By replacing junior staff with AI, we are creating a hiring and skills gap crisis in our workforce 10-15 years down the line. If you want your business to last that long, it’s your duty to hire actual real humans to learn on the job.
3. Generative AI can’t empathise or fix human problems
Artificial intelligence is a deceiving name. Because it isn’t intelligent. It is just a relatively sophisticated prediction engine, generating the next most likely word in the sequence. It doesn’t have a point of view, it doesn’t have human experience, it’s usually willing to lie to you to tell you what you want to hear, and it doesn’t empathise.
Empathy is one of the key cornerstones of marketing. How you speak to a person, how you make them feel, and how you understand their issues goes a long way to establishing messaging and building trust with your audience.
Sure, your business may have been born from a strategic gap in the market. But you know what problems those gaps solve. You’ve spoken to your customers and helped them through various issues with your product or service. Use that experience. Just be a fucking person.
4. Trust and authenticity are becoming more and more important
The last ten years have taken us from “fake news” through to full on dystopian levels of people not knowing what is real anymore. So what do they want? Clarity, honesty, authenticity, and the opportunity to interact with an actual, real person.
Is shoving an AI chatbot on your side good for lining your pockets? In the short term, maybe. But you don’t have to look very far to see that the general sentiment is that folks want to talk to a real human person. Because they can trust that person to listen to or read what you’ve said accurately and actually try to help.
As Erin Simmons talked about at BrightonSEO last week, trust and community are becoming key pillars of marketing. Folks are getting more isolated, have less contact with humans beings, and are relying more and more heavily on Reddit threads, communities, and personal recommendations.
You will not get those recommendations if you’re hard to work with, hard to reach, or your website is full of AI slop.
5. Building authority is near impossible when none of the work comes from you
This one is two fold. One: stop denying yourself the opportunity to learn and grow by outsourcing your brain to a large language model. And two: so much of authority building comes from your own experience and expertise. The E-E-A-T of it all is undermined when you’re not putting the work in yourself.
How can you become a beacon of knowledge and authority on a topic if your content is simply an amalgamation of everything that already exists? Again, these “generative AI” models are not actually intelligent. They are advanced prediction engines that populate text with the next most likely word.
And sure, you can train an AI to sound like you and pretend to think like you but at that point, are you really saving any time or money?
Don’t fall for the sunk cost fallacy of AI investment
So you’ve spent the last couple of years, a lot of time, and more money than you were planning to creating AI-generated content that you know still doesn’t hit the mark.
Marketing stopped being a valuable step towards sales and simply became a box ticking exercise. Believe it or not, I completely understand why this happens. But your business deserves better and so do your customers.
It’s not too late to turn away from generative AI and start working with real people again. People who can empathise with your audience, understand their struggles, and know how to nudge people along the now-much-more-complicated buying journey. Take the time back you spend configuring your own personalised LLM and build your own personal creative skillset.
So where do you go from here? (specifically regarding SEO)
Search is changing. I’m not going to sit here and pretend it’s not. But that requires more human input, not loss. It’s time to focus on a broader strategy that pays equal attention to your messaging, brand, website copy, UX, and conversion as well as your SEO content.
SEO content can no longer be treated like a solo discipline. Your website builds trust and your content builds authority. You need both to succeed in the messy and constantly changing SERPs and LLM citations.
Zero-click search is increasing, so now is the time to focus on getting more from the traffic you do get. It’s understandable to be scared when you see traffic going down. But if you can keep conversions stable or even increase them, you’re onto a winner.
Resources to help you with conversion and UX
- Your homepage H1 can make or break your website
- What supermarkets can teach us about conversion copywriting
- 5 ways formatting and content design are harming your conversion rate
- More pages do not equal better SEO or better website performance
- What the enshittification of the internet means for SEO
- Context is key: capturing the right attention with conversion-driven copy