Generative Engine Optimization: Our Learnings So Far
Generative Engine Optimization (GEO) is a new technique, where best practices are still being discovered.
In this article we share some researched tips as well as our own learnings from applying them to our relatively fresh and small website in a very competitive niche (web analytics).
Spoiler alert: It worked.
But before we get to tips and examples, let’s break down the basics.
What Is Generative Engine Optimization (GEO)?
Generative Engine Optimization (GEO) is the process of making your content more visible in AI-generated search results.
AI search tools like Perplexity, ChatGPT search, Bing AI, Grok Deep Search, and Google Gemini don’t just serve up blue links. They aggregate, synthesize, and generate responses in a completely different way.
Think about it. If you search for "best web analytics tools" on a traditional search engine, you get a list of links: link one, link two, link three. You choose which to click, read the content, and decide for yourself.
With AI search, that process is different. The AI does the reading for you. It scans multiple sources, compares insights, and presents an answer directly. Instead of just showing a list of tools, it might say:
"The best simple web analytics tools are Brand A, Brand B, and Brand C, based on comparisons across multiple sources."
So how does it decide which brands to list? What signals does it need to see to pick your content? How do you ensure that your brand is part of the conversation when AI search engines generate answers?
That's what Generative Engine Optimization (GEO) is all about.
Traditional SEO vs. Generative Engine Optimization (GEO)
| Factor | Traditional SEO | Generative Engine Optimization (GEO) |
|---|---|---|
| Primary Goal | Rank high in search results ( | Be included in AI-generated responses ( |
| Core Algorithm | Keywords, backlinks, E-E-A-T | Content quality, structured data, citations, direct API connections |
| User Interaction | Users click links to find answers | AI delivers summarized answers directly |
| Content Source | Indexed web pages and site structure | Multiple sources (websites, APIs, books, research papers, etc.) |
| Optimization Focus | Keywords, backlinks, domain authority | High-quality, well-structured, factual, and authoritative content |
Looking at the table above, you might think that GEO is quite similar to SEO. However, this isn't entirely true. AI-driven search engines draw from a broader range of sources (not just web pages) and prioritize clarity, factual accuracy, and structured data which makes it a slightly different beast.
Tips for GEO and how we applied them
1. Optimize for "what", "how", "why", and keep it conversational
Identify the types of questions your audience might ask and structure content accordingly. Informational, instructional, and comparative questions are the most common types.
How to do it:
- Use FAQs and question-based content to match conversational search queries.
- Don't use complicated sentences and jargon — imagine explaining it to a 10-year-old.
- Structure content in Q&A format, which AI tools favor.
- Leverage long-tail, intent-driven keywords instead of just short, competitive keywords.
For example, our website has a Terms & Definitions section, pages from which clearly ask and answer the question "What is [Term]?" ChatGPT picks them up. But more on that later.
2. Embed authority signals
AI models weigh content credibility and authority, even though they usually don't cite sources.
How to do it:
- Embed data and statistics, communicate with numbers.
- Publish research-backed content – AI favors factual, well-researched articles.
- Name entities, e.g., "...as mentioned by Elliott Hill, the CEO of Nike."
3. Use structured data
Noticed how this article is structured? Aside from the obvious, like headings and subheadings, we use tables and bullet points to answer the "how" type of question.
Tables lay things out in a structured way, which can speed up spotting patterns or key points compared to wading through paragraphs of plain text. So, yeah, tables can have an edge for a deep search.
— Grok 3.
How to do it:
- Use clear headings, tables, bullet points, and concise sentences.
- Write in a way that AI can easily extract and summarize (avoid dense paragraphs).
- Follow the Inverted Pyramid Model by starting with the key takeaway, then expanding with details.
We have this long-tail keyword article titled "Best Web Analytics for 2025". Perplexity picked it up, because it has a clear concise bullet point list and a niching down on "2025" part. Given the freshness of our domain, there was no way for the article to compete without this approach.
4. Add an llms.txt file to your website
While robots.txt and sitemap.xml are designed for search engines, llms.txt is optimized for reasoning engines. It provides information about a website to LLMs in a format they can easily understand.
— From a Towards Data Science article on llms.txt by Derick Ruiz.
But don't panic if it's the first time you hear about it. Our research shows that the file is adopted mostly by websites with extensive documentation (Stripe, Mintlify, Eleven Labs, etc.) to solve the limited context window issue. But not so much for other reasons.
As an example, Stripe has a llms.txt at the docs.stripe.com level, but there's no llms.txt at their root domain.
It's still too early to tell what impact a root domain file has on GEO, but we decided to add it to our website. Just in case, to see what happens. You can find it in the root of our website if you're interested.
Our logic was to highlight our most important content, so we went ahead and listed our competitor comparison pages and most popular blog posts there.
It's too early to tell about the impact of adding this file, but we'll make sure to post updates with our learnings.
5. Best SEO practices work here too
Let's highlight some of the most important ones:
- Domain authority is still one of the most important things, just like backlinks.
- AI search engines rely on high-authority sources like Wikipedia, research papers, and expert publications. Being referenced in these sources increases your brand’s inclusion in AI responses.
- Google's E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness) standards are even more important in GEO. Quotes, references, lived experience stories, etc. draw a factual basis for the answer.
- Main rule still applies: write content humans would want to read.
How to monitor GEO
GEO is a new field, where best practices are still being discovered.
The main problem we have is lost attribution. Most of the time the traffic won't be referred from the AI search engines, but from Google, Bing, and other traditional search engines, where people might go and search for your brand after it got mentioned in an AI response.
We have tools created by big tech, such as Hubspot AI Search Grader, and also smaller ones, like LLM Console, that are trying to solve this, but the overall experience is not yet there.
You can also replicate this yourself by manually searching for important keywords in Perplexity, ChatGPT, Gemini, and other AI search engines, and see if your or competitors' websites show up.
And while waiting for this niche to mature, we think it's still valid to monitor GEO by looking at your website analytics.
Checking your attributed traffic is a good starting point to get a sense of which pages are being picked up by LLMs.
Here's how it could be done with a simple filter on your Seline dashboard:
This way you can clearly see which pages are being picked up by LLMs.
Get started with web analytics for free.
Sign up for Seline now - no credit card required. Cookieless, privacy-friendly, independent.
Get started
More of our examples
When someone asks "What is Campaign Source in Google Analytics?" in Perplexity, here's what they get:
Why is Seline the main source here? Our article is nicely optimized for what people are searching for. And it covers all the main questions like "what is it", "where to find it", "why it's important", etc. in a way that's easy to understand.
There are also plenty of links to our article. But do all the searchers click on them? Probably not, since they’re already getting the answers they need without clicking.
What does this mean? Not sure, but we should probably count it towards "brand awareness."
Now, look at these sweet results for "best simple analytics" prompt.
Seline's article is the first source, because it's optimized for "simple analytics" prompt, and gives a concise list of tools with short descriptions — exactly what searchers are looking for.
Hopefully these examples help you get a more practical sense of how GEO works, and how you can apply it to your website.
Generative Engine Optimization FAQ
1. What is the difference between GEO, LLMO, LSO, AIO, GAIO?
It's all the same, there is not yet a single accepted term for this new field, but GEO is the most popular one currently.
2. What is LLM search?
LLM (Large Language Model) search refers to search engines powered by AI models like GPT-4, ChatGPT, or Gemini, which don’t just provide links to content but instead generate responses directly.
3. How is GEO different from SEO?
SEO is about ranking well in traditional search engines like Google or Bing. GEO, on the other hand, focuses on getting your content into AI-generated search responses, where the AI, for example, ChatGPT, summarizes information instead of just providing links.
4. Why should I care about GEO?
AI search engines are becoming more popular, and getting your brand featured in their responses increases visibility. Even if users don’t click on your link, your brand gets seen, which can lead to more recognition and trust.
5. How do I track my GEO results?
Tracking GEO is different from regular SEO. You can use tools like Seline to monitor traffic, check AI search engines manually for your brand mentions, or use analytics tools to see which pages are being picked up by AI models.
6. How can I improve my GEO?
To improve GEO, focus on creating high-quality content that AI models can easily reference. Use clear and organized information and relevant keywords, and build your brand’s authority across different platforms to increase your chances of being featured.
