Real Estate Agent Websites
A real estate agent website earns its cost through three things: live IDX search that keeps buyers on your domain, lead capture that reaches you before the lead cools, and content that search engines and AI assistants can actually read. The third is where most agent websites quietly fail.
What should an agent website actually do?
Three jobs, in order of how often they are neglected:
- Keep buyers on your domain. If your search experience is worse than the portals, buyers leave and the lead belongs to someone else.
- Capture and route leads fast. An inquiry that sits unanswered overnight is substantially less valuable than one answered in minutes.
- Be readable by machines. Both the search index and, increasingly, the AI assistants that now answer a large share of queries directly.
Why is the third job suddenly the hard one?
Because the crawler landscape changed and most website platforms did not.
Independent crawler analyses have found that GPTBot, ClaudeBot, PerplexityBot, OAI-SearchBot and similar agents fetch raw HTML and never execute JavaScript. Google's Gemini rides Googlebot's rendering infrastructure and AppleBot uses a browser-based crawler, but they are the exceptions.
The consequence is blunt. If your website assembles its content in the browser — which most template-driven agent sites do — then to those crawlers your pages are empty. Not low quality. Empty. No amount of good writing fixes a page that is never seen.
How do you tell whether your current site has this problem?
Open your site, view source (not the inspector — actual view-source), and look for your article text. If the body copy is not in that raw HTML, neither search engines on first fetch nor AI crawlers at all can read it.
What does the fix look like?
Prerendering: generating real static HTML for every page at build time, so the content is in the initial server response. Your Agent Site does this for every article and landing page, along with:
- Per-page titles, meta descriptions and canonical URLs
- JSON-LD structured data identifying the article, the breadcrumb trail and the agent entity
- A sitemap and robots.txt regenerated on every build
- Direct-answer passages near the top of each page, which is the specific content shape generative engines quote
What about local search?
Town-level landing pages, structured around genuine local intent, plus a topical content cluster that links them together. A single "Listings" page cannot rank for eight different towns; eight pages that each answer one town's questions can.
Note the one thing no website can do for you: the Google local map pack requires a verified Google Business Profile with consistent name, address and phone. That is a business-identity task, not a website feature, and any platform claiming otherwise is overselling.
Frequently Asked Questions
Do AI crawlers read JavaScript websites?
Mostly no. Crawler studies show GPTBot, ClaudeBot, PerplexityBot and OAI-SearchBot fetch raw HTML and never execute client-side code. Google’s Gemini rides Googlebot’s rendering pipeline and AppleBot uses a browser-based crawler, but they are exceptions.
How do I check whether my site is readable by AI crawlers?
Use view-source on a content page rather than the browser inspector. The inspector shows the page after JavaScript runs; view-source shows what a crawler receives. If your body copy is missing there, non-rendering crawlers cannot see it.
Will a new website get me into the Google map pack?
No website alone will. The local pack requires a verified Google Business Profile with consistent name, address and phone details. A website supports local ranking but cannot substitute for that verified business identity.