dncu.com — AI Site Grade

Analysis

DNCU's homepage dateModified is set to 2026-05-19 — a date that has not yet occurred — which means every AI crawler ingesting schema metadata is being fed a publication timestamp from the future, undermining temporal credibility signals.

Crawler Access

All major AI crawlers — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, ChatGPT-User, Applebot-Extended, anthropic-ai — receive a full 200 response with identical content to a browser baseline (116,581 bytes). The sole exception is Bytespider (ByteDance's crawler), which gets a Cloudflare 403 block. The robots.txt at https://www.dncu.com/robots.txt is a bare Yoast-generated file with a single User-agent: * Disallow: rule and no AI-bot-specific directives whatsoever. No llms.txt exists (404). The site runs on Cloudflare (DNS via Cloudflare nameservers, IP 104.17.127.5) with no apparent WAF-based UA filtering beyond the Bytespider block.

Cold-Knowledge Gap

The LLM prior knows DNCU as a not-for-profit cooperative based in Santa Fe, New Mexico, originally chartered in 1952 as Los Alamos Schools Credit Union, serving northern New Mexico counties. The site itself tells a slightly different origin story: chartered in 1954 as Los Alamos Scientific Laboratory Credit Union, serving 60,666 members with $1.0 billion in assets across Los Alamos, Espanola, Santa Fe, and Rio Rancho. The cold model knows the "Green Checking" product — but the site does not mention "Green Checking" anywhere. The site's actual checking lineup is Express, Star, and Free 4 Me accounts. This is a meaningful gap: the model would describe a product the site no longer offers or has rebranded.

Schema Posture

Every page inspected uses Yoast-generated JSON-LD with WebPage, WebSite, BreadcrumbList, and (on blog posts) Article schema types. No FinancialProduct, BankOrCreditUnion, LoanOrCredit, Product, or FAQPage schema is present anywhere — despite the site having extensive rate tables, loan comparison grids, and an FAQ page with structured Q&A content. The homepage schema declares dateModified: 2026-05-19, a future date. The FAQ page's dateModified is 2022-02-11 — over four years stale. The checking page has a rich comparison table (Express vs Star vs Free 4 Me) with rates, fees, and feature columns, but this is rendered as plain HTML with zero structured data markup.

Content & Answer Signals

The homepage is a 660-word JS-rendered page (WordPress) with multiple H1s, a member testimonial, and a prominent interactive "How can we improve your life?" selector that lets visitors self-identify (homeowner, student, parent, etc.) and get product recommendations. This interactive tool is invisible to crawlers as static text. The checking page contains a genuine comparison table with three account types side-by-side. The FAQ page has 12+ real Q&A pairs but uses no FAQPage schema. The blog has 377+ indexed posts dating back years, covering community events, financial literacy, and promotions — a strong content corpus for AI training, but none of it surfaced via an llms.txt.

External Signals

External search results for DNCU are remarkably sparse. No Reddit threads, no review-site snippets, no news articles surfaced in DuckDuckGo searches. The brand has near-zero off-domain citation footprint in the search results sampled. This means AI models have very little third-party signal to triangulate against — they rely almost entirely on the site's own content and whatever training data included the brand. The DNS TXT records show two Google site verification tokens and a Kinsta mail SPF record, confirming the hosting stack.

Surprising Details

The dateModified of 2026-05-19 on the homepage is the most anomalous single finding — it appears to be a placeholder or a Yoast bug, but it means every schema-aware crawler ingests a future timestamp. The Bytespider block is the only UA-based restriction, which is unusual since Bytespider is typically the least aggressive AI crawler. The site has a WordPress XML-RPC pingback endpoint exposed (xmlrpc.php), a known security surface. The sitemap index contains 8 sub-sitemaps (post, page, FAQ, news, category, tag, faqs_category, author) with 377+ URLs — a healthy crawl surface — yet the robots.txt sitemap directive points to sitemap_index.xml which does resolve correctly (200).

Findings

1. Homepage schema dateModified set to future date 2026-05-19 High

   The homepage's JSON-LD schema declares a dateModified of 2026-05-19, a date that has not yet occurred. This feeds AI crawlers a publication timestamp from the future, undermining temporal credibility signals.

   What to change: Update the dateModified field in the homepage schema to the actual last-modified date or remove it if not accurately maintained.

2. No FinancialProduct or BankOrCreditUnion schema on product pages High

   Despite extensive rate tables, loan comparison grids, and checking account details, no FinancialProduct, BankOrCreditUnion, LoanOrCredit, or Product schema is present. This limits AI models' ability to extract structured financial data.

   What to change: Add appropriate FinancialProduct, BankOrCreditUnion, and LoanOrCredit schema markup to all product and rate pages.

3. Cold LLM knowledge describes a checking product not offered on the site High

   The LLM prior knows DNCU for a 'Green Checking' product, but the site's actual checking lineup is Express, Star, and Free 4 Me. This mismatch means AI models may describe a product that no longer exists or has been rebranded.

   What to change: Ensure the site prominently features current product names and descriptions, and consider updating external listings or schema to align with the actual product lineup.

4. FAQ page lacks FAQPage schema markup Medium

   The FAQ page contains 12+ real Q&A pairs but uses no FAQPage schema. This prevents AI crawlers from directly extracting structured Q&A content for featured snippets or knowledge panels.

   What to change: Add FAQPage schema with Question and Answer markup to the FAQ page.

5. FAQ page dateModified is over four years stale Medium

   The FAQ page's schema declares dateModified as 2022-02-11, which is over four years old. This may signal to AI crawlers that the content is outdated.

   What to change: Update the dateModified field to reflect the actual last update or remove it if not maintained.

6. Interactive product selector is invisible to crawlers Medium

   The homepage features an interactive 'How can we improve your life?' tool that lets visitors self-identify and get product recommendations. This content is rendered via JavaScript and is not visible to static crawlers, missing an opportunity to surface personalized product signals.

   What to change: Ensure the tool's content is also available in static HTML or provide a text-based fallback for crawlers.

7. Near-zero external citation footprint in search results Medium

   Searches for DNCU on DuckDuckGo returned no Reddit threads, review snippets, or news articles. This lack of third-party signals means AI models have little external data to triangulate against, relying almost entirely on the site's own content.

   What to change: Encourage member reviews on third-party platforms and engage in community forums to build off-domain citations.

8. No llms.txt file available for AI crawlers Medium

   The site returns a 404 for llms.txt, meaning there is no curated file to guide AI crawlers to key content or provide a summary of the site's offerings. This is a missed opportunity to improve AI visibility.

   What to change: Create an llms.txt file that lists important pages and provides a brief site summary for AI crawlers.

9. Bytespider crawler is blocked by Cloudflare Low

   Bytespider (ByteDance's crawler) receives a 403 block from Cloudflare, while all other major AI crawlers are allowed. This may limit visibility on ByteDance's AI platforms.

   What to change: Review the Cloudflare WAF rules to ensure Bytespider is not inadvertently blocked, or explicitly allow it if desired.

10. WordPress XML-RPC pingback endpoint exposed Low

    The site exposes xmlrpc.php, a known security surface that can be used for pingback attacks or brute-force attempts. While not directly an AI visibility issue, it is a security concern.

    What to change: Disable XML-RPC if not needed, or restrict access to trusted IPs.

What's working

• All major AI crawlers allowed with full content access — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, and others receive a full 200 response with identical content to a browser baseline. No AI-bot-specific disallow rules exist in robots.txt.
• Sitemap index with 8 sub-sitemaps and 377+ URLs — The sitemap index contains 8 sub-sitemaps covering posts, pages, FAQs, news, categories, tags, and authors, providing a comprehensive crawl surface for search engines and AI crawlers.
• 377+ blog posts covering community and financial topics — The blog has a large corpus of content dating back years, covering community events, financial literacy, and promotions, which provides rich material for AI training and visibility.
• Yoast-generated JSON-LD schema on all pages — Every page inspected includes Yoast-generated JSON-LD with WebPage, WebSite, and BreadcrumbList schema, providing a baseline of structured data.
• Rich comparison table for checking accounts — The checking page features a genuine comparison table with three account types side-by-side, including rates, fees, and features, which is valuable content for AI crawlers.
• Cloudflare provides security and performance — The site uses Cloudflare for DNS and CDN, which offers DDoS protection, caching, and performance benefits, contributing to site reliability.