esource.com — AI Site Grade

Analysis

The sitemap points to a subdomain (esd9.esource.com) that does not resolve in DNS, meaning every URL in the sitemap is effectively dead to crawlers — the site's entire content inventory is invisible.

Crawler Access

All major AI crawlers — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, Bytespider, Applebot-Extended — receive a 200 with full HTML content identical to browser delivery. No UA-based blocking exists. The site runs on nginx/1.18.0 (Ubuntu) on AWS (EC2, no CDN/WAF layer), built on Drupal 11. The robots.txt contains a generic User-agent: * rule with no AI-bot-specific directives — a neutral posture that neither blocks nor prioritizes AI crawlers. The llms.txt endpoint returns 403 Forbidden, a surprising block given the site has an anthropic-domain-verification TXT record (ClaudeBot verification) suggesting active ClaudeBot engagement.

Content & Schema Posture

Every page fetched — homepage, about, customer solutions, grid solutions, insights — contains zero JSON-LD schema of any type. No Organization, WebSite, FAQPage, Article, or BreadcrumbList schema is present. The canonical tags use http://www.esource.com/ (non-HTTPS), a minor but persistent inconsistency. The homepage delivers ~820 words of visible text with strong H2/H3 heading structure and clear answer-format signals (case studies, event listings, insight teasers), but no FAQ, comparison tables, or definition patterns. The content is well-written and specific to utility industry challenges (arrearages, CIS transformations, heat pump markets), making it naturally AI-answerable — but without schema, AI engines have no structured hooks.

Cold-Knowledge Gap

The LLM prior knows E Source as a utility research and consulting firm founded in 1986, focused on customer experience, energy efficiency, and grid modernization, serving North American utilities. It mentions a "Customer Insights" platform and annual forums. The actual site positions itself more aggressively: "Total clarity. Accelerated impact." — a modern, execution-oriented brand voice emphasizing three solution pillars (Customer, Business/Technology/Operations, Grid). The site mentions Grid Solutions (vegetation intelligence, storm/wildfire intelligence) and arrearage benchmarking as flagship offerings — none of which appear in the cold LLM knowledge. The LLM also knows nothing about the DEFG acquisition (visible in sitemap press releases) or the E Source Forum 2026/2027 events prominently featured on the homepage.

External Signals & Structural Issues

The sitemap.xml (114KB, ~50 URLs) points exclusively to esd9.esource.com — a subdomain that fails DNS resolution (no A/AAAA/CNAME records). This means every press release, case study, newsletter, and report in the sitemap is unreachable to any crawler. The main www.esource.com domain has no sitemap entries at all. The Wayback Machine has no snapshots of the homepage, suggesting the site may be recently rebuilt or has blocking directives. External search results for E Source are sparse — no reviews, Reddit threads, or press coverage surfaced in search, indicating a low external citation footprint that limits the brand's AI-answer authority. The only external link on the site is to LinkedIn.

Findings

1. Sitemap points to non-resolving subdomain esd9.esource.com High

   The sitemap.xml lists URLs under esd9.esource.com, which has no DNS records, making all sitemap entries unreachable by crawlers.

   What to change: Update the sitemap to use the main www.esource.com domain and ensure all listed URLs resolve correctly.

2. No JSON-LD schema on any page High

   Every fetched page lacks any structured data markup (Organization, WebSite, Article, etc.), reducing AI engines' ability to extract and cite content.

   What to change: Add JSON-LD schema for Organization, WebSite, Article, and BreadcrumbList to all relevant pages.

3. llms.txt returns 403 Forbidden Medium

   The llms.txt endpoint is blocked despite the site having an anthropic-domain-verification TXT record, indicating a configuration inconsistency.

   What to change: Allow access to llms.txt and populate it with a summary of the site's content for AI crawlers.

4. Canonical tags use non-HTTPS URLs Low

   Canonical tags reference http://www.esource.com/ instead of the HTTPS version, a minor inconsistency that may confuse crawlers.

   What to change: Update canonical tags to use the HTTPS version of the URL.

5. Low external citation footprint Medium

   Web searches for E Source returned no reviews, Reddit threads, or press coverage, limiting the brand's authority in AI answer generation.

   What to change: Encourage customer testimonials, case studies, and industry press to build external backlinks and citations.

6. LLM prior lacks key offerings and events Medium

   The LLM prior does not mention Grid Solutions, arrearage benchmarking, DEFG acquisition, or E Source Forum events, which are prominent on the site.

   What to change: Publish more content on these topics and ensure they are indexed and linked from authoritative sources.

7. Main domain has no sitemap entries High

   The sitemap.xml only lists URLs under esd9.esource.com, leaving the main www.esource.com domain without any sitemap coverage.

   What to change: Generate a sitemap that includes all pages on www.esource.com and submit it to search engines.

8. No Wayback Machine snapshots of homepage Low

   The Wayback Machine has no snapshots of the homepage, suggesting the site may be recently rebuilt or has blocking directives.

   What to change: Ensure the site is not blocking archiving bots and consider submitting to the Wayback Machine.

What's working

• All major AI crawlers receive full HTML content — GPTBot, ClaudeBot, PerplexityBot, and others get a 200 with full HTML, identical to browser delivery, ensuring content is accessible.
• Robots.txt has no AI-bot blocking directives — The robots.txt contains only a generic User-agent: * rule, neither blocking nor prioritizing AI crawlers, maintaining open access.
• Content is well-written with clear headings and answer-format signals — Pages have strong H2/H3 structure and specific utility industry content (arrearages, CIS transformations) that is naturally AI-answerable.
• Anthropic domain verification TXT record present — The DNS TXT records include an anthropic-domain-verification entry, indicating active ClaudeBot engagement setup.
• Site runs on Drupal 11 with nginx on AWS — The technology stack (Drupal 11, nginx, AWS) is modern and capable of supporting SEO and AI visibility improvements.