Large language model optimisation (LLMO)

Table of contents:

• What are LLMs?
• What is LLMO?
• LLMO, SEO, AEO and GEO what are the differences?
• Why you should start LLMO now
• How we help you with LLMO
• LLMO best practices
• Example from practice
• How do you get started with LLMO?
• AI SEO checklist
• Blogs about AI SEO
• Frequently asked questions about LLMs

What are Large Language Models (LLMs)?

Large language models (LLMs) are smart AI systems trained to deal with language. They can write texts, answer questions, create summaries and much more. They can do this because they have been trained on huge amounts of text. So they learn to recognise patterns and connections. Based on that, they create new texts that fit what you ask.

A well-known example is ChatGPT from OpenAI, With ChatGPT you can do all sorts of things: write texts, make translations, ask questions, have code generated or even come up with new ideas.

But as a ondernemer, what good is an LLM?

LLMs (such as ChatGPT) do not provide a list of links, but instant answers. For many people, that feels faster and clearer than clicking through to websites themselves. For businesses, this changes the rule of the game: it is not enough to rank high in Google. You want to be in the answer itself. Because that's what people really read. So LLMs are a like an extra channel to get your goods or service onder noticed! Are you not being featured enough in these AI Models? Then you are missing out on potential customers!

What is LLMO?

LLMO stands for Large Language Model Optimisation. (in Dutch: Groot taalmodel optimalisatie). This is a new way of optimising online, targeting AI systems like ChatGPT and Perplexity. It is similar to SEO, but works differently.

Instead of aiming for search results in Google, here you influence the answers given by AI systems. You do this, for example, by writing your content in such a way that LLMs can include it in their answers. You can also influence what information they use.

Because AI works differently from a search engine, LLMO requires a different approach. So companies should not only be findable, but also discussed in AI's answers and quoted as sources to the company's website.

The beauty of LLMO optimisation is that we can measure how well you perform in LLMs like ChatGPT and also compare this with your competitors. On our page about LLM tracking we tell you all about it! The image hereonder shows how we track prompts from potential clients and how often our content is cited as a source in them.

LLMO, SEO, AEO and GEO what are the differences?

LLMO is the strategy that ensures your brand is reflected in LLMs' answers. You can think of it as the overarching approach on top of SEO, AEO and GEO :

• SEO: the technical basics - zonder good content, technology and authority will get you nowhere.
• AEO (Answer Engine Optimization): focused on systems that provide instant answers, such as Google AI Overviews or voice assistants.
• GEO (Generative Engine Optimization): focuses on visibility in generated AI content, such as summaries in ChatGPT or Perplexity.

LLMO connects these three layers and is the key to sustainable online visibility in an increasingly AI-driven world.

Why you should start LLMO now

How Onder helps you with LLMO

Is your website already mentioned in the answers of ChatGPT, Gemini or Perplexity? At Onder, we make sure AI systems recognise and cite your organisation as a reliable source. You focus on your business, we make you visible. We translate LLMO - think AI SEO, AEO and GEO - into a clear approach. From content and technology to strategy: we make sure your brand stands out in AI answers, where one result makes all the difference.

Wondering what we can do for you? Then contact us without obligation!

What are the best practices for LLMO?

LLMO is still new, and good onder search is still lacking. Still, you can do a lot right now. LLMs (like ChatGPT) build their answers from what they have learned from existing content. By cleverly capitalising on those sources, you increase the chances of your brand appearing in the answer.

Spread your brand online

LLMs often extract information from database websites, knowledge aggregators or major publishers such as the Financial Times or Forbes. Examples of database websites are business directories or assessment platforms. Knowledge aggregators are, for example, Wikipedia or YouTube. Within LLMO, it's all about visibility: your brand should show up on external websites and in influential databases. Use PR campaigns or paid content to get mentioned on platforms that LLMs care about.

Five key guidelines

1. Align your content structure with the structure of the LLM.

Ask an LLM directly how it would describe or structure an onder object, product or service. Use this as the basis for your own writing and onder back it up with structured data. This may seem like imitation, but it is an effective way to be taken on board by an LLM.

2. Use clear and informative language

LLMs process information best when it is clear, simple and rich in information. Texts with clear comparisons, advantages and disadvantages are ideal.

3. Avoid long ononderbroken texts

Long blocks of text zonder headings are difficult for LLMs. Provide short sections, start longer pieces with a summary, and use sub-headings so that the content is more digestible.

4. Add citations and dates

Quotes and references to well-known experts or figures can increase findability and relevance in LLM results. Statistics and quantitative data also strengthen the LLMO strategy.

5. Keep optimising continuously

As with SEO, LLMO is an ongoing process. Companies must constantly evolve their strategy and respond to technological innovations and competitors. Stagnation means loss of visibility.

Example from practice

Suppose you have a page about the importance of sunscreen.

With traditional SEO, focus mainly on keywords such as "buy sunburn" or "best sunscreen". But with LLMO, it goes beyond keywords. An LLM like ChatGPT or Gemini needs to understand your content, be able to quote it and use it in a response.

• Write a clear explanation zonder jargon
  For example, "Sunscreen protects your skin from harmful UV radiation and reduces the risk of skin ageing." That way, everyone understands the gist immediately.
• Answer the questions your target audience really asks
  Think about: "How often should I apply sunscreen?", "What is the difference between factor 30 and 50?" or "Which sunscreen is suitable for children?".
• Provide a logical structure with clear headings and bullet points
  Create separate sections with H2s such as "How does sunscreen work?" or "Frequently asked questions about sunburns". Add bullet points so answers are scannable.
• Reinforce reliability with E-E-A-T
  Onder Support your content with sources, such as a reference to the KWF Cancer Society or dermatological onder searches. Also add signals of expertise, for example by quoting a dermatologist.

That way, you increase the chances of an LLM selecting your text and using it in an AI response. Not only will you be found via Google, but also served in the new generation of search and response systems.

Make sure your text is ready to be quoted in AI-generated answers. Stop thinking only about "ranking" and start thinking about "being served" in an AI response.

How do you get started with LLMO?

Would you prefer personal advice? Then feel free to contact the specialists at Onder. We will be happy to help you with all your questions about LLMO, AEO, GEO and SEO - and together build a content strategy that is ready for the future.

Would you rather get started with LLLMO yourself, but don't know where to start? Then start with our AI SEO checklist!
In this checklist, you will find the most important quick-wins you can already implement on your website to perform better AI models.

AI SEO checklist

Blogs about LLM

• How are you mentioned in ChatGPT and other AI chatbots?
• Preparing for the arrival of AI mode
• AI developments and the impact on SEO

Frequently asked questions about LLM

What does LLMO mean?

LLMO stands for Large Language Model Optimization. Its strategy is to optimise content and online presence for major language models. Where SEO focuses on search engines, LLMO focuses on findability in AI systems such as ChatGPT, Gemini or Perplexity.

What does LLM mean?

LLM stands for Large Language Model. It is a type of AI that is trained on huge amounts of text and can therefore understand and generate human language. LLMs can write texts, answer questions, translate and summarise.

Is ChatGPT an LLM?

Yes, ChatGPT is an LLM. It is built on OpenAI's GPT models and can understand, generate and structure human language. ChatGPT is one of the best-known applications of LLM technology.

Is Gemini an LLM?

Yes, Google's Gemini is an LLM. It is the successor to Bard and belongs to the family of multimodal large language models that combine text, images and sometimes audio.

What is the difference between LLM and AI?

AI is the collective term for artificial intelligence. LLMs are a specific type of AI specialising in language. All LLMs are AI, but not all AI is an LLM. AI can also do image recognition, robotics or prediction, while LLMs focus on text and language.

What are the benefits of LLMO?

The advantages of LLMO are:

• Visibility in AI answers such as ChatGPT and Gemini
• Higher chance of being quoted as a reliable source
• Being prepared for the shift in search behaviour towards LLMs
• A sustainable strategy combining SEO, AEO and GEO