Winning the AI shelf: How to compete for visibility in the LLM era

AI is changing the way people access and interact with information. Some studies find that 70% of people already use the technology for search, with up to 14% using it daily.

As Google ramps up testing of AI Mode and AI Overviews, we’re starting to see what the future of search looks like. Brands know they need to optimise for LLM visibility, but a serious lack of data is putting marketing budgets at risk - and the likes of Google aren’t helping.

LLM visibility doesn’t solve everything

Everyone in the industry is talking about LLM visibility, but showing up in AI tools isn’t the end goal. You have to convert visibility into revenue for it to have any business value, so make sure you’re asking the right questions when optimising for LLMs:

• Does AI visibility drive traffic?

• Does LLM traffic convert?

• Can you attribute AI visibility to traffic, conversions, and revenue?

• Can you optimise for consistent visibility (when every AI response is different)?

• Does AI visibility consistently drive enough revenue?

Brands need to see an ROI from AI visibility if they’re expected to optimise it. This is where Google, OpenAI, and the other leaders in AI search need to step up. If they want businesses to invest in organic visibility, they need to give us access to the necessary performance data.

We need the ability to track LLM visibility, click-through rates, and engagement metrics (most importantly, conversions), separately from other traffic sources. Until this happens, brands risk wasting huge amounts of money on LLM optimisations that don’t deliver an ROI. Business leaders will only tolerate this for so long before they start questioning where their money is better spent.

A word of caution on AIO, GEO, and the great SEO rebrand

Many search marketers are pushing hard to rebrand SEO as AIO, GEO, and every acronym possible. This is part of an effort to drive SEO investment, under a new name, by making it sound like everything has changed in this new age of AI search.

It’s more profitable to sell GEO services that sound radically different, than adapting existing SEO strategies. You can see this in the trend of renaming established SEO techniques as if they’re new strategies.

New buzzwords	Translation
Audience AI platform behaviour analysis	Audience research
AI crawlability and indexability	Technical SEO
Topical breadth and depth	Content strategy
Chunk-level retrieval optimisation	Clear, structured writing
Answer synthesis optimisation	Clear, accurate writing
Citation-worthiness	Accurate information
Content authoritativeness signals	Accurate information, brand citations, link building, thought-leadership (ie: SEO and digital PR)
Multi-modal optimisation	Multimedia content
Personalisation resilience	Content relevance, and intent optimisation

Yes, we have to adapt these techniques for AI search, but we’re not rewriting any rules here, and we certainly don’t need to rename the SEO fundamentals.

Be wary of any advice pushing you to invest serious budget into big SEO changes. The names that dominated traditional search still dominate AI search. Not because they’ve overhauled their SEO strategies. Because they’ve got the brand reputation, authority, natural links and citations, etc.

Adapting your marketing strategy for the LLM era

You don’t need to overhaul a functional SEO strategy to optimise for LLM visibility.

Although we’re working with limited insights during these early days, every reliable data source says the SEO fundamentals are as important as ever in the LLM era:

• 76.10% of pages cited by AI Overviews rank in the top 10 positions.

• Brands ranking on page one of Google show a strong correlation (~0.65) with ChatGPT and other LLM mentions.

• Strong correlation (0.664) between branded web mentions and AI Overview citations.

• Significant correlation (0.527) between branded anchor text (backlinks) and AI Overview citations.

Ultimately, the most important factors for LLM visibility align with traditional SEO fundamentals:

1. Strong organic search rankings for relevant topics

2. Brand mentions across authoritative sources

3. Branded backlinks on third-party content for relevant topics

Yes, we have to adapt these strategies to the intricacies of AI search, but a working SEO strategy gets you 95% of the way there. Don’t try to break and rebuild what already works. Adapt the most impactful strategies that influence traditional search and LLM visibility to maximise gains across the board.

Link building

Every reliable data source shows a strong correlation between quality backlinks and mentions across AI Mode, AI Overviews, ChatGPT, and other AI search tools.

Effective link building doesn’t only generate backlinks. It builds brand awareness, increases brand citations, encourages publishers to write about you, and increases online chatter about your brand - all signals LLMs trust. Don’t think of link building as a standalone strategy. Consider the cumulative impact of links, citations, brand awareness, etc. on building your authority as a trusted source of information..

Citations

Brand citations are one of the most influential factors in LLM visibility. AI systems source information from multiple sources, so repetition in brand mentions and contextual information counts for a lot. Let’s say you want LLMs to recommend you as the best brand in a particular category. Well, then you need plenty of third-party sources making the same recommendation - including the top citation sources listed above.

Digital PR

LLM systems are affected by digital PR strategies far more than traditional search engines. Everything published online has the potential to influence AI-generated responses, including web content, social posts, customer reviews, press coverage, guest posts, content partnerships, etc.

Adapt your digital PR strategy (including link building and citations) to optimise for LLM visibility, and influence what they’ll say about you by shaping the online narrative around your brand.

Platform-specific optimisation

AI search is producing genuine competition for Google, far beyond anything traditional search engines managed. Each LLM system has its own algorithm favouring slightly different signals, including their preferred information sources. The latest insights on where AI tools source their citations from reveal a strong bias to major, trusted resources and publications:

AI Overview’s top citation sources

1. Reddit 21%

2. YouTube 18.8%

3. Quora 14.3%

4. LinkedIn 13.0%

5. Gartner 7.1%

ChatGPT’s top citation sources

1. Wikipedia 47.9%

2. Reddit 11.3%

3. Forbes 11.3%

4. G2 6.7%

5. TechRadar 5.5%

Perplexity’s top citation sources

1. Reddit 46.7%

2. YouTube 13.9%

3. Gartner 7.0%

4. Yelp 5.8%

5. LinkedIn 5.3%

Optimising for multiple platforms isn’t limited to LLM visibility. Search itself is facing competition from other channels as online experiences evolve at a faster rate than ever.

Some people are already ditching Google in favour of ChatGPT and other LLM systems. Others simply use their favourite social media apps for most of their online activity - particularly Gen Z, which increasingly searches on TikTok and Instagram.

Another group has actively switched to Reddit amid growing frustration over the quality of Google Search results. Meanwhile, most product searches happen on Amazon, people search for takeaways on Deliveroo, and holiday searches can start and end on platforms like Airbnb and Skyscanner. Aside from adapting to AI search, brands need to adopt a more channel-agnostic approach to marketing - one that builds and converts visibility everywhere.

Building brand power beyond search and LLM visibility

We’ve developed a culture in this industry of accepting everything Google does, and jumping through hoops to accommodate it. Brands are entirely dependent on a handful of tech providers because of this attitude. And the same dependence allows tech giants to do whatever they want - like take every piece of content you’ve ever produced without any compensation.

The only long-term strategy that makes sense at this point is to build enough brand power that tech giants need you, not the other way around. Look at the top sources of information for each LLM system. They’re trusted because of their reputation and authority.

Reddit is the top source of information for AI Overviews, and Reddit results are all over Google’s organic SERPs now. It’s no coincidence that many people prefer using Reddit over Google as a search tool. That’s brand power.

Likewise, pay attention to where LLMs source their statistics, quotes, and factual information. Garner is a top source for Google and Perplexity, while industry-specific sources like HubSpot are favourites for relevant prompts.

Instead of optimising for reliance on LLM visibility, why not become the source of information that LLMs rely on? That’s the kind of brand power that drives performance across all marketing channels, without dependence on any particular platform.

If you’re worried about the impact of AI on your marketing strategy, our Performance team is constantly testing strategies for LLM visibility. Leave your details below and a member of our team will be in touch to schedule a call.