digitaltwentyfour.com — AI Site Grade

Analysis

Digital 24 has an OpenAI domain-verification TXT record (openai-domain-verification=dv-...) but no llms.txt — the site registered with OpenAI to prove domain ownership for GPT crawlers, yet offers no structured AI-friendly content map, creating a contradiction between verification intent and execution.

Crawler Access

All major AI crawlers (GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, ChatGPT-User, anthropic-ai) receive HTTP 200 with full HTML content identical to browser baseline. Only Bytespider (ByteDance/TikTok) returns 403 blocked. The robots.txt contains no AI-bot-specific directives — a single catch-all User-agent: * rule with standard WooCommerce exclusions and a Yoast-generated empty allow block. The site runs on nginx behind Cloudflare nameservers, with no WAF layer that discriminates by UA. AI crawlers see the full Elementor-rendered page (~426KB), including case studies, service pages, and the FAQ schema.

Cold-Knowledge Gap

The LLM's prior knowledge describes Digital 24 as a Belfast agency offering "SEO, PPC, social media marketing, and web design/development" serving "SMEs across Northern Ireland and the UK." The actual site does not mention web design or development as a service — the service menu covers PPC, social media advertising, social media management, content creation, email marketing, influencer marketing, digital strategy, and training only. The cold knowledge also omits the agency's explicit anti-contract stance (no minimum contract length, a core differentiator on the FAQ page) and its CIM Best Agency NI award displayed prominently site-wide.

Schema Posture

The site uses Yoast SEO schema with Organization, WebSite, WebPage, Article/BlogPosting, BreadcrumbList, and Person types. The FAQ page correctly implements FAQPage with Question/Answer entities — a strong answer-format signal. However, no LocalBusiness schema is present despite the agency having a physical Belfast address with phone and email. The Organization schema lists only Facebook, X, and Instagram as sameAs — LinkedIn is missing despite the site linking to a LinkedIn profile in the footer. Case study pages lack ItemProp or Review schema to mark up the specific results (follower growth, ROAS, CPA figures).

Content Freshness & Blog Quality

The blog is actively maintained with posts dated 2026 (e.g., "Why you shouldn't use Chat GPT for your social media strategy in 2026" published May 1, 2026). The Q Radio case study is dated March 2026. This forward-dated content indicates the site publishes ahead of its time or uses future-dated timestamps — a potential trust signal issue for AI engines that parse datePublished. The blog contains substantive, well-structured articles (1,600+ words) with named authors, not thin content.

External Signals

The site has near-zero indexed external mentions from web search. No review sites, Reddit threads, or press coverage surfaced in queries. The only off-domain signals are the Google Partner directory listing and social media profiles (Facebook, Instagram, LinkedIn). The DNS TXT records include an OpenAI domain verification token (openai-domain-verification=dv-eJAiWe2f7vMyAAc4UkNMB4N0) — the site has proactively verified ownership with OpenAI for GPT crawler access, yet has not created an llms.txt to guide what those crawlers should prioritize.

Findings

1. No llms.txt despite OpenAI domain verification Medium

   The site has an OpenAI domain-verification TXT record but no llms.txt file, meaning GPT crawlers have verified access yet receive no structured guidance on which content to prioritize.

   What to change: Create an llms.txt file at the root that lists key pages (services, case studies, FAQ) and provides a brief summary of the agency.

2. Cold knowledge omits actual service offerings High

   LLM prior knowledge describes Digital 24 as offering web design/development, but the site does not list that service. The actual services (PPC, social media, content, email, influencer, digital strategy, training) are not reflected in the LLM's knowledge.

   What to change: Ensure the site's service pages are well-structured with schema markup and that the LLM knowledge base is updated via structured data and external citations.

3. Cold knowledge misses key differentiators Medium

   The LLM's prior knowledge does not include the agency's anti-contract stance (no minimum contract length) or its CIM Best Agency NI award, both prominently featured on the site.

   What to change: Add structured data (e.g., FAQPage for the anti-contract stance, Award schema for the CIM award) to help AI crawlers extract these differentiators.

4. No LocalBusiness schema despite physical address Medium

   The site has a physical Belfast address, phone, and email but does not implement LocalBusiness schema, reducing local AI visibility.

   What to change: Add LocalBusiness schema with address, phone, email, and opening hours to the contact or about page.

5. LinkedIn missing from Organization sameAs Low

   The Organization schema lists only Facebook, X, and Instagram as sameAs, but the site footer links to a LinkedIn profile.

   What to change: Add the LinkedIn URL to the Organization schema's sameAs array.

6. Case study pages lack result markup High

   Case study pages contain specific results (follower growth, ROAS, CPA) but no ItemProp or Review schema to mark up those figures, making them invisible to AI crawlers as structured data.

   What to change: Add Review or ItemProp schema to case study pages to mark up key metrics and results.

7. Blog posts use future dates (2026) Medium

   Blog posts and case studies are dated in 2026, which may cause AI engines to distrust the datePublished field and reduce credibility.

   What to change: Use accurate publication dates or ensure the datePublished field reflects the actual publish date.

8. Near-zero indexed external mentions High

   Web searches for the agency name and domain returned no results from review sites, Reddit, or press, indicating very low external citation signals.

   What to change: Build external citations through guest posting, directory listings, and PR to increase off-domain signals.

9. Case studies archive page is content-thin Medium

   The case studies archive page contains only 11 words, providing no summary or context for AI crawlers to understand the available case studies.

   What to change: Add descriptive text and summaries to the case studies archive page.

10. News archive page is content-thin Low

    The news archive page contains only 40 words, providing little context for AI crawlers.

    What to change: Add introductory text and category descriptions to the news archive page.

What's working

• All major AI crawlers receive full HTML content — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, and others get HTTP 200 with full HTML, identical to browser baseline.
• FAQ page implements correct FAQPage schema — The FAQ page uses FAQPage with Question/Answer entities, providing strong answer-format signals for AI crawlers.
• OpenAI domain verification TXT record present — The site has an openai-domain-verification TXT record, indicating proactive registration with OpenAI for GPT crawler access.
• Blog contains substantive, well-structured articles — Blog posts are 1,600+ words with named authors, not thin content, providing valuable text for AI crawlers.
• Yoast SEO schema provides solid foundation — The site uses Yoast SEO schema with Organization, WebSite, WebPage, Article, BreadcrumbList, and Person types.
• No AI-bot-specific blocking in robots.txt — The robots.txt has no AI-bot-specific directives, ensuring all AI crawlers can access the site.