Words: Alex Coe – Growth Director
Category(s): Content
Date: 1st August 2026
Many B2B marketers are trying to work out how their LLM optimisation efforts should dovetail with their other organic growth tactics. The reassuring news is the emergence of LLMs as a useful and popular research tool shouldn’t change the digital marketing calculus much at all.
AI search is a new discovery channel
Our job as marketers is (and has always been) to try and be present in all the places our target audience hangs out and pays attention, whether they’re actively searching for what we offer or not. By doing this we maximise our chances of being top of mind when our audience decides to buy something we are selling.
As part of this effort we need to optimise for search engines, for social media channels like YouTube, Reddit and LinkedIn, and now for LLMs. Of course, optimising for Google in and of itself requires us to optimise for its AI overviews, which appear at the top of almost all search queries nowadays. And then there’s Google’s AI Mode tab which offers a similar interactive chat experience to other LLMs.
In fact, look at the top of the Google search page and you’ll also see Images, Video, Maps, Shopping, and more. The point being, optimising for various channels and platforms is par for the course for digital marketers.
This brings us to the question of whether GEO (generative engine optimisation) deserves its own acronym. We do use the term quite regularly at Beneath, not least because clients and prospects often do. We also used it in a recent blogpost called What makes a good SEO or GEO strategy? (because it’s an important keyword we are optimising for). But, really, given the term SEO applies to so many different channels and platforms already, it might be more useful to use the term ‘optimising for AI search’ over GEO.
How AI search differs from traditional search
LLMs don’t rank pages in the same way Google does. Rather, they compile answers based on retrieving information from brands, people and sources that appear consistently across the web – citing them in the process.
At the risk of oversimplifying the matter, optimising for AI search is an entity authority play: helping LLMs know who you are, what you do, and what you know – then cite you. Traditional search on the other hand still relies heavily on keyword relevance and query matching: helping you rank for those keywords, and making it easier for people to find you (although, of course, topical authority, brand signals, trustworthiness, expertise and user satisfaction are still important).
The goal for B2B brands should be to appear consistently in conversations related to their expertise across the web – for example, on social media platforms, in the news, on review platforms, in industry publications, on podcasts, at events, on partner websites, and in research reports.
Consistent mentions across trusted sources help AI systems build confidence in who your organisation is, what it specialises in, and when it should be cited as an authority. And the more original your thoughts, the better. Original research, proprietary data and distinctive points of view are far more likely to be cited.
Technical considerations for marketers
Weaving in priority keywords, crafting headlines that mirror search queries, adding links to webpages, soliciting backlinks from external websites with a high domain authority . . . all these ‘traditional’ search optimisation tactics remain relevant and important.
But specifically optimising for AI search – including Google’s AI Overviews – requires a few additional considerations. The key is to make it easy for AI to understand information. For example:
- Answer questions directly
- Use clear headings and structure
- Include FAQs and lists where appropriate
- Support claims with evidence
- Ensure important information is easy to extract and cite
As with all good digital marketing in the B2B space, you should plan your content marketing strategy around user intent and the role the content plays in the customer journey.
Bottom-of-the-funnel content designed to capture high-intent traffic – such as service pages, product pages, pricing pages and comparison articles – should be optimised for SEO in the traditional way, with specific, relevant keywords and backlinks from external websites with a high domain authority.
Content that deserves extra AI optimisation effort would include research reports, industry explainers, and expert commentary – because LLMs are likely to cite the sources of this kind of information when generating responses.
In reality, there’s a lot of overlap between optimising for traditional search engines and LLMs: both reward high-quality, trustworthy and well-structured content. So the best approach will often be to optimise for both.
But remember: just because you can optimise for AI doesn’t mean you should. If you lean too hard into these tactics, you can end up creating sections that might feel odd to the reader, or structuring a piece with questions that nobody would ask.
The end result might be that, yes, you get cited in an LLM. But because you’ve created a sub-optimal content experience for your reader, the content is less likely to engage, and its performance in traditional search engines might be hindered.
Let’s not forget that content that audience members spend time reading, interact with, like, share and reference tends to rank higher. Conversely, it sends a bad signal to the search engine if audience members spend a few seconds reading the content before looking for another article to read on the same subject.
So don’t sacrifice a potentially commercially powerful piece of content for the sake of being cited in an LLM for an audience member that might not be too far along the buying journey at all.
One final important point. Using AI to produce large volumes of low-value content is risky. In its spam policies, Google states that ‘using generative AI tools or other similar tools to generate many pages without adding value for users’ is likely to lead to the content being penalised. The takeaway? Put your faith in good writers to produce compelling, original, trustworthy content.
Allocating budget to AI search
Calculating how much time and effort to invest in a particular channel is one of the trickiest challenges digital marketers face because, to be frank, it requires a fair amount of guesswork. At Beneath, we use a roulette analogy. We allocate our chips (budgets) across the tables (channels) we expect to yield the greatest expected return. Then we gradually move our chips to the opportunities that pay the most out.
There’s no doubt that optimising for LLMs is becoming increasingly important, with visits steadily rising in the US and Europe, according to Datos’s quarterly State of Search survey. However, the data confirms that optimising for Google should still be the priority as it handles hundreds of times more search-like queries than LLMs. So back to our roulette analogy: think carefully about the timing and logic of moving your chips off the Google table and onto the AI one.
Don’t lose sight of marketing fundamentals
Search optimisation is a crucial marketing tactic. But it’s even more important to focus on being noticed and remembered by your target audience. If your brand has a distinctive point of view, publishes content your audience wants to read, and generally appears in important conversations that are relevant to your offer, then you’re on the right track. Optimisation then becomes a useful amplifier.
AI search may be changing how people discover information, but it doesn’t change the fundamentals of good marketing.
For a chat on any of the above, get in touch with our search experts.