SEO is, most emphatically, not dead. 

This apparently needs to be restated, because not a day passes without some marketer or other announcing its demise. 

But search engine optimization has undoubtedly evolved over the past few years. The standard SEO playbook (do keyword research, optimize the website, create keyword-stuffed content, build backlinks, sit back and count dollars) no longer works — if it ever did. 

Today’s marketers know they need to move with the times. Reddit dominates search results, users are exploring alternatives to Google, and AI overviews are the new “featured snippet.” 

Enter generative engine optimization. Think SEO, but make it AI. 

The marketers leading the field today aren’t just concerned about ranking on the SERPs. They’re thinking about how to write for, influence, and show up in the LLMs that are increasingly dominating search. They’re planning on how to incorporate generative AI optimization into their marketing campaigns. 

In this article, we’ll tell you what you need to know to nail your generative AI optimization strategy. 

What’s Generative Engine Optimization (GEO)? 

Generative Engine Optimization, or GEO, is the process of adapting marketing content to increase its chances of appearing in AI-generated responses. 

Traditional SEO focuses on modifying content to rank higher in search engines like Google or Bing. GEO (also known as generative AI optimization) is about structuring and distributing content so that it is easily found and shown to users of popular generative AI tools like ChatGPT or Google Bard (lately renamed Gemini). 

There is a lot of overlap between modern SEO techniques and recommended GEO best practices. Content that passes Google’s infamous E-E-A-T guidelines will probably also perform well in AI-based search tools. 

For example, as Kevin Indig points out in an article for Search Engine Journal, “ChatGPT uses Bing search results to ground and weigh answers, which means sites with strong visibility on Bing also have a high chance of being very visible in ChatGPT Search.” 

Plus, there’s a little controversy about the research paper that most GEO guidelines are based on. 

However, there are some real opportunities for content optimization that are genAI-specific, which we’re going to get into now. The good news — many of these will also help you ensure your content keeps your human readers happy too. 

7 steps to optimizing your content for generative AI

Here are some simple steps you can take to make sure your content gets featured in AI-generated search responses: 

1. Make sure your site allows search AI crawling

If you want your content to show up in genAI search results, the first step is to make sure it’s visible: 

  • Configure your robots.txt file. The robots.txt file on your website’s server tells search engine crawlers which pages or sections they are allowed to index. Make sure that key pages on your site (like your FAQs) aren’t blocked so they show up in search results. 
  • Update your meta tags. Meta tags like [meta name="robots" content="index, follow"] signal to crawlers that they should index your page and follow the links within it.

For more detailed guidelines, check out Search Engine Journal’s guide, ChatGPT Search Indexing: Essential Steps For Websites

2. Update your FAQ pages

If you haven’t looked at your FAQs for a while, now’s the time to fix that. FAQ pages are very LLM-friendly because they address common questions directly, so AI tools recognize them as the right content to use for answers. 

To make your FAQs more likely to get picked up by AI search engines: 

  • Phrase the questions you answer in full sentences, using natural language. 
  • Avoid information overload. Each FAQ should address one topic or question only. 
  • Be concise and to the point. Straightforward answers that clearly address question intent will be preferred. 
  • Be specific. For example, instead of a generic question like “What are your services?” try formats like, “What services do you offer for small business owners?” or “How can your tool help my marketing team?”
  • Use FAQ schema markups. The FAQ schema helps AI models identify and extract relevant questions and answers from your page.  

3. Create query-focused content

As well as updating your FAQs, focus your content around questions that searchers are likely to ask. Content that performs particularly well in terms of generative AI search include: 

  • “How-to” guides — in-depth, step-by-step guides that answer your target audience’s questions
  • “What is” articles — “What is” and other definitional content is easy for search AI to understand and therefore rank 
  • Comparison pages —Side-by-side comparisons between your products or services and other competing solutions tend to work well for genAI search because they line up with the way AI models process and deliver answers to user queries.

4. Update your information on knowledge repositories and databases

Generative AI models pull information from trusted sources to answer user questions. Making sure that your information is up to date, that you’ve used relevant keywords when describing your product, and that all product descriptions are current will help you show up in AI-generated search results. 

It’s worth noting that different genAI models will pull data from different repositories and databases. By keeping your details up to date across multiple sources, you’re more likely to be featured in AI search results, regardless of which AI search provider your prospects use. 

For example: 

  • For business software, make sure your products appear on G2, Capterra, and TrustRadius.
  • If your product has technical integrations, make sure you’ve got accurate documentation on Stack Overflow and GitHub.
  • Wikidata, GitHub repositories, and open-source knowledge bases also contribute to what LLMs know about your brand.  

5. Make content easily skimmable by genAI search models 

LLMs tend to prioritize clear content because that’s how they “know” that the content answers users’ questions. To make your article more AI-friendly, try: 

  • Adding sections and summaries. Short summaries at the beginning of your blogs (or at the start of each section for longer articles) can make it easier for genAI search models to digest. 
  • Check that your content structure is clear. Breaking down information into well–organized, skimmable sections isn’t just helpful for readers — it also helps genAI tools. Well-named H2s and H3s, lists, and short paragraphs are easier for LLMsI to scan and interpret. 
  • Check your internal linking for new and existing content. Make sure that you’ve linked all your relevant pages, products, and other entities within your content. This contextual linking helps AI understand relationships between products, features, and industry terminology associated with your brand.

6. Update your keyword strategy 

Standard SEO best practices will stand you in good stead for AI search. For instance, there’s a 99.5% overlap between Google’s AI Overviews and top organic results. 

That said, you may want to make sure that your keyword approach is still in line with modern SEO best practices, especially if you haven’t looked at it in a while. Recent guidance from Microsoft recommends the following AI-specific keyword strategies: 

  • Use long-tail keywords. For example, instead of “sales enablement”, try “best sales enablement platform for manufacturing.” These longer keywords reflect more specific user intents, which AI models are good at understanding. Plus, users tend to use longer, more complex questions to search via AI than they do on the SERPs, according to Indig. 
  • Use conversational phrases. AI tools will often prioritize content written in natural language patterns — and users often interact with AI in more conversational phrases. So, instead of a keyword like “Top ERP system features," try "What makes an ERP system right for your business?"
  • Remember semantic keywords. Including related keywords is helpful for AI models because they look for this information when selecting content to use. So, for instance, alongside "CRM platform," you could use related terms like "sales pipeline management," "customer data insights," and "lead tracking software."
  • Consider user behavior. Look at your current deck of keywords, and consider whether or not a user would use those phrases when interacting with an AI search engine. 

7. Keep up with the latest developments

GEO is a very new development in digital marketing, and best practices are changing as quickly as the AI tools themselves evolve. To make sure you stay ahead of the rest: 

  • Freshen up your content. AI models look for freshness and accuracy to decide what content is most relevant for users. So it’s even more important to keep your web copy up to date. 
  • Set up a system to notify search engines of changes. Microsoft recommends you use an open-source tool like IndexNow to notify search engines and AI models about changes to your content. 
  • Track changes to the algorithm and features. Keep an eye on blogs and update pages from providers. Join industry communities focused on genAI, or sign up for relevant newsletters. 
  • Monitor your performance. For example, some SEO platforms like SEMrush now include features to track your website's visibility within AI Overviews.

Keep doing what you’re doing (but keep an eye on the future) 

Many current SEO best practices will also help your content appear in generative AI-based searches. Techniques like schema markups, contextual linking, optimizing FAQ pages for clarity, using query-focused content, and promoting your brand in industry-specific directories and knowledge bases will all help you rank on search engines, while also appearing in AI overviews and AI searches. 

However, as you’ll already be aware, AI models change quickly, and the savvy marketer will be keeping an eye on the future for new developments. 

If you’re wondering about other ways that AI could be helping you tackle the marketing challenges of the future, you might want to watch our webinar: Work Less, Achieve More: How sales enablers and marketers can take back their time with AI