From Users to Algorithms: How AI Search Personalisation is Reshaping Brand Visibility

From Users to Algorithms: How AI Search Personalisation is Reshaping Brand Visibility

Let’s start with the uncomfortable truth: brand visibility isn’t what it used to be. Remember the days when SEO was all about ranking #1 on Google? Get your keywords right, optimise your meta tags, maybe sprinkle in a few backlinks, and job done. But today? That playbook is borderline ancient history.

Why? Because search isn’t about everyone anymore. It’s about *you*. Or rather, the specific "you" sitting behind the screen: your preferences, your browsing history, your intent in that exact moment. This shift is what we call AI search personalisation, and it’s changing how brands are discovered, or not discovered, online.

Here’s the kicker: even if you’re visible in traditional search engines, there’s no guarantee you’ll appear in personalised AI results. I’m talking about ChatGPT, Google’s AI Overview, Claude, Perplexity, the tools people are increasingly turning to for answers, recommendations, and decision-making. If AI can’t find your brand, does it even matter how well-optimised your website is?

In this piece, I’ll break down how AI search personalisation works, what it means for enterprise visibility, and what brands can do to adapt. I’ll also share data from the 3897 AI prompts we’ve run at Contxt to highlight just how wide the visibility gap can be. Buckle up, this one’s worth your time.

How AI Search Personalisation Works

Let’s demystify the concept. AI search personalisation tailors results based on the user’s context, preferences, and behaviour. It’s not just about what you search for, but how, when, and why. These systems are trained to interpret intent behind queries, pulling from massive datasets, including your search history, location, and even the device you’re using.

Take ChatGPT, for instance. When you ask it for a restaurant recommendation, it doesn’t just spit out a random list of places. It pulls in variables like the time of day (are you searching at lunchtime or 10 PM?), prior preferences (do you lean toward Italian or sushi?), and broader data about what similar users prefer. Related reading: The Real Cost of Ignoring LLM Visibility: A Business Risk.... For a deeper look, have a read of Lessons from Award-Winning LLM Visibility Platforms: What.... Suddenly, the “best” restaurant isn’t universal, it’s algorithmically determined for you.

That sounds brilliant for users, right? Sure, but for brands, it’s a minefield. AI search personalisation introduces bias. Not necessarily malicious bias, but bias nonetheless. The algorithms are trained on finite datasets that inherently favour some brands over others. If your brand isn’t in the mix, or worse, if the algorithm hasn’t even been trained on your data, you’re invisible. Related reading: this piece about why 39% of ai systems disagree on brand recommendations (.... Full stop.

Case in point: across our network of 3897 prompts, we’ve seen a 38% divergence between ChatGPT and Google AI Overview on brand recommendations. That means over a third of the time, these two systems, which ostensibly answer the same queries, surface completely different answers. If your brand’s recommended on one, there’s no guarantee it exists on the other.

The Implications for Brands

So, what does this mean for brands? A lot, frankly. Visibility is fragmented in a way we haven’t seen before.

Let me give you an example from our tracking data. Of the 43 brands we monitor, seven are stuck in what I’d call the “low visibility zone”, appearing in less than 30% of AI-driven searches. One of the worst cases is Castle Pumps, a manufacturer in the marine sector, with just 6.3% visibility across our dataset. That’s despite having a solid presence in traditional SEO. On the flip side, brands like Oak&More and Specialist Blinds are smashing it with 100% visibility. You can visibility tracking tools to see for yourself. You can the Contxt scanning process to see for yourself. They’re everywhere.

But here’s what I keep seeing: most brands don’t even realise this is happening. They’re focused on their old metrics, Google rankings, website traffic, maybe some social performance. Meanwhile, they’re asleep at the wheel when it comes to AI algorithms. And that’s a problem, because AI personalisation isn’t just a niche phenomenon. It’s how users will find what they need moving forward.

Think of it this way: the battle for visibility is no longer just about users. It’s about winning over algorithms first. If AI systems don’t know who you are (or worse, if they don’t trust you), you’ve lost before you even start.

The Bias Problem in Personalised AI Recommendations

Here’s a question: do you trust AI to make unbiased recommendations? If your answer’s yes, I’ve got some bad news, it’s a lot messier than that. AI search personalisation is inherently shaped by the data it’s trained on, and that data comes with gaps.

At Contxt, we ran 58 experiments across 14 businesses to test how brand visibility varies by query type, and the results were eye-opening. A brand might be recommended for one type of query (say, “best sports tech for football fans”) but completely absent from another (“indoor football experience near me”). Visibility is as much about how the query is framed as it is about the brand’s actual authority in the space.

And then there’s the issue of bias favouring certain providers. Across the same set of prompts, ChatGPT had a mention rate of 74.4% for tracked brands, while Google AI Overview clocked in at just 45.1%. That’s a massive gulf. If your business is banking on being discoverable across all platforms, you’re in for a nasty surprise. No two systems rank brands the same way, and it’s not just about algorithms, there are commercial factors at play too.

Strategies to Navigate the AI Personalisation Shift

So, how do you win when the rules are so fluid? It’s not rocket science, but it does take a strategic rethink. Here’s where I think brands need to focus their energy:

1. Understand Your Current Visibility

You can’t fix what you don’t measure. The first step is benchmarking your brand’s presence across major AI systems. This means running prompts, tracking visibility rates, and identifying where you’re absent. At Contxt, we’ve built tools to do exactly that, testing prompts for everything from buying intent to knowledge recall to competitor displacement. For a deeper look, have a read of this piece about invisible metrics: how llm tools miss the mark on measuri.... I've written about this separately in this piece about open-source vs commercial llm visibility tools: lessons l.... If you’re not auditing your AI visibility, you’re already falling behind.

2. Optimise Content for AI

Traditional SEO isn’t dead, but it needs an upgrade. AI models thrive on structured, clear, and authoritative information. Think FAQs, well-labelled product data, and conversational content that mirrors how users actually ask questions. One brand we work with jumped from 13% to 38% visibility just by tweaking their content to align better with how AI processes queries.

3. Diversify Your AI Strategy

Remember that 38% divergence I mentioned earlier between ChatGPT and Google AI? That’s your wake-up call to stop relying on one platform. If you’re visible on ChatGPT but invisible on Google’s AI, you’re leaving money on the table. Test across multiple systems, and don’t assume what works on one will automatically work on the others.

4. Feed the Algorithms

This one’s often overlooked: make sure your brand data is actually accessible. AI systems pull from structured datasets like product feeds, Wikipedia entries, and supplier databases. If your business isn’t contributing to these data pools, you’re voluntarily removing yourself from the conversation. Visibility in AI search personalisation starts with being part of the training data.

5. Monitor Changes Over Time

AI evolves fast. What works today might not work tomorrow. In our testing, we’ve seen algorithm updates dramatically shift which brands are prioritised. One month you’re riding high; the next you’re nowhere to be found. Regular monitoring isn’t optional, it’s essential if you want to stay ahead.

The Opportunity in AI Personalisation

Look, I know this might sound like doom and gloom. But here’s the thing: for brands willing to adapt, the opportunity in AI search personalisation is enormous. Personalised AI recommendations can give the right brand unparalleled access to hyper-targeted audiences. You’re not competing for generic eyeballs; you’re reaching users who genuinely want what you offer.

Take Specialist Blinds, for instance. Their 100% visibility rate isn’t by chance. They’ve nailed the basics, authoritative content, consistent data signals, and proactive testing across multiple platforms. That’s not luck; that’s smart strategy. And it’s exactly what every brand should be aiming for.

I’d argue this shift is forcing us to be better marketers. It’s no longer enough to shout the loudest or spend the most. You’ve got to be relevant, trustworthy, and easy for algorithms to recognise. Innovation is about asking the why, why is a competitor outperforming us? Why are we absent for specific queries? Why do certain algorithms favour other brands? Digging into those questions is where the magic happens.

The Road Ahead

AI search personalisation isn’t going anywhere. If anything, it’s only going to get more sophisticated. As the tech evolves, the gap between visible and invisible brands will widen. The question is, where will your brand sit?

My theory? Brands that take this seriously now will dominate the next decade. They’ll be the ones showing up in personalised AI recommendations, capturing intent-driven traffic, and building authority with both users and algorithms. Everyone else will scramble to catch up, or worse, get left behind entirely.

So, what’s your next move? Start by understanding where you stand. Test your visibility. Experiment with how your content resonates with different AI systems. Stay curious, stay proactive, and, above all, stay visible. This is the future, and it’s here now.

If you’re ready to stop guessing and start owning your AI visibility, let’s talk. The brands that act today are the ones that win tomorrow.