A developer's checklist for Generative Engine Optimization (GEO)

Developers now need to consider Generative Engine Optimization (GEO) as traffic increasingly originates from sources like ChatGPT, Gemini, Perplexity, and Google AI Overviews. The goal is to have your product featured in those answers, which is achieved through engineering work. Here's a developer's checklist for GEO:

1. Allow AI crawlers access. Ensure your site allows GPTBot, PerplexityBot, ClaudeBot, or Google-Extended to crawl it by removing any broad disallow rules from robots.txt. Without crawl access, your content cannot be cited.

2. Serve content in HTML. Avoid client-side rendering, as it can hinder visibility. Make sure the main content, headings, and links are present in the server response.

3. Implement structured data. Utilize JSON-LD formats for Organization, SoftwareApplication, FAQPage, and Article types. This provides engines with clear information about your site and offerings.

4. Focus on answer-first pages. Provide a direct answer to a single clear question at the top of each page, then expand upon the topic. Short definitions, step-by-step instructions, and comparisons tend to be quoted most often.

5. Create an llms.txt file. This small file lists your key pages, helping AI agents locate the relevant content more efficiently.

6. Measure your performance. Select 20 to 50 prompts your users commonly ask and track which engines mention your site in response over time. Automating this process can streamline the optimization effort. The Raechal AI project aims to automate this entire workflow, auditing sites, fixing technical issues, generating missing content, publishing it to the CMS, and tracking citations across various platforms.

A free plan is available for those who wish to try it on their own website. The full checklist is openly available on GitHub under the name geo-checklist.