makeheroes.co — AI Site Grade

Analysis

I have everything I need. Here's the audit.

Cold-Knowledge Catastrophe: LLM thinks this is a Shopify agency

The LLM's cold knowledge describes Heroes Technology Ltd as a Shopify Plus Partner building custom apps (Heroes Checkout, Heroes Subscriptions, Heroes Theme). The site itself describes a multi-brand nursery aggregator that acquires and scales brands like Tutti Bambini, Snuggle Me Organic, Boba, Trunki, Beco, and iL Tutto. These are two entirely different companies. The model has no awareness of the nursery rollup business, the brand portfolio, or the acquisition model.

Crawler Access

The robots.txt is a bare WordPress default — Disallow: /wp-admin/ with no AI-bot rules whatsoever. No llms.txt exists (404). GPTBot, ClaudeBot, Google-Extended, Applebot-Extended, Bytespider, and anthropic-ai all receive zero-byte responses with server disconnects — the nginx server actively terminates their connections. ChatGPT-User, OAI-SearchBot, PerplexityBot, and Perplexity-User get full 200 responses with identical content to browser. This is a partial, inconsistent AI-blocking pattern: the crawlers that power training (GPTBot, ClaudeBot, Google-Extended) are cut off, while the retrieval/search bots (ChatGPT-User, Perplexity-User) are allowed.

Content & Schema

The homepage has zero JSON-LD schema of any type — no Organization, no WebSite, no Product for the portfolio brands. The blog section is in critical disrepair: all six blog posts (/on-the-radar/, /on-the-radar-2/, /on-the-radar-3/, /growth-strategy/, /growth-strategy-2/, /exit-strategy/) return ~17 words of visible text — the body content is JS-rendered behind Elementor popups and invisible to plain HTTP crawlers. Two posts contain placeholder "What is Lorem Ipsum?" / "Why do we use it?" headings, indicating unfinished content. A sample-page with WordPress default text ("I'm a bike messenger by day...") is live and indexed in the sitemap. The /landing-page/ contains the richest brand portfolio content (Snuggle Me Organic, Tutti Bambini, iL Tutto acquisitions) but is not linked from main navigation.

External Signals

The press page lists four media mentions from 2021-2022 (The Times, Business Insider, Financial Times) covering the Amazon rollup space. No press coverage newer than January 2022 is cited. Web searches for "Heroes Technology" return zero indexed results — the domain has minimal external footprint. The brand portfolio brands (boba.com, bumco.com, niimo.co) sit behind Cloudflare challenge pages that block crawlers entirely.

Stale Infrastructure

The sitemap is a WordPress default (wp-sitemap.xml) with 19 URLs including a sample-page, a landing-page with no nav presence, and greenhouse-job-descriptions containing only a heading and footer. Copyright notices show both 2023 and 2024. The careers page lives on a separate subdomain (talent.makeheroes.co) with proper FAQ schema and active job listings — this is the most AI-friendly page on the entire domain.

Findings

1. LLM cold knowledge describes wrong business model High

   LLMs describe Heroes Technology Ltd as a Shopify Plus Partner building custom apps, while the site is a multi-brand nursery aggregator. This mismatch undermines AI-driven discovery.

   What to change: Publish structured data (Organization, WebSite) and a clear value proposition on the homepage to correct LLM knowledge.

2. Homepage has zero JSON-LD schema High

   The homepage lacks any JSON-LD markup (Organization, WebSite, Product), making it invisible to AI crawlers that rely on structured data.

   What to change: Add Organization and WebSite JSON-LD schema to the homepage, and Product schema for portfolio brands.

3. Blog posts render as near-empty JS shells High

   All six blog posts return only ~17 words of visible text; body content is JS-rendered behind Elementor popups, invisible to plain HTTP crawlers.

   What to change: Server-side render blog content or use progressive enhancement so crawlers can read the full text.

4. Inconsistent AI bot blocking pattern High

   GPTBot, ClaudeBot, Google-Extended, Applebot-Extended, Bytespider, and anthropic-ai receive zero-byte responses with server disconnects, while ChatGPT-User, OAI-SearchBot, PerplexityBot, and Perplexity-User get full access. This partial blocking confuses AI visibility.

   What to change: Allow training bots (GPTBot, ClaudeBot, Google-Extended) or block all AI bots consistently; update robots.txt accordingly.

5. No llms.txt file published Medium

   The site returns a 404 for /llms.txt, missing an opportunity to guide AI crawlers to key pages.

   What to change: Create an llms.txt file listing important pages like /landing-page/ and /careers/.

6. Domain has zero indexed results in web search High

   Web searches for 'makeheroes.co' and 'Heroes Technology Ltd' return zero results, indicating minimal external footprint and poor SEO.

   What to change: Improve SEO fundamentals: fix blog content, add schema, build backlinks, and ensure pages are crawlable.

7. WordPress sample page is live and indexed Medium

   A default WordPress sample page with placeholder text is live and included in the sitemap, diluting content quality.

   What to change: Remove the sample page and exclude it from the sitemap.

8. Blog posts contain placeholder Lorem Ipsum text Medium

   Two blog posts feature 'What is Lorem Ipsum?' and 'Why do we use it?' headings, indicating unfinished content that harms credibility.

   What to change: Complete or remove placeholder blog posts.

9. Key landing page not linked from main navigation Medium

   The /landing-page/ contains the richest brand portfolio content but is not accessible from the main navigation, reducing its visibility to users and crawlers.

   What to change: Add a link to the landing page in the main navigation or footer.

10. Portfolio brand sites blocked by Cloudflare Medium

    Portfolio brand domains (boba.com, bumco.com, niimo.co) return Cloudflare challenge pages that block crawlers, preventing AI from connecting them to the parent company.

    What to change: Allow AI crawlers through Cloudflare or add links from the parent site to these domains.

11. Press coverage is over two years old Low

    The press page lists media mentions from 2021-2022 only, with no recent coverage, suggesting a decline in PR activity.

    What to change: Pursue recent press mentions and update the press page.

12. Copyright notices show both 2023 and 2024 Low

    The site displays inconsistent copyright years, indicating maintenance neglect.

    What to change: Update all copyright notices to the current year.

What's working

• Careers subdomain has FAQ schema and active listings — The talent.makeheroes.co subdomain includes proper FAQ schema and active job listings, making it the most AI-friendly page on the domain.
• Robots.txt allows most crawlers to access the site — The robots.txt only disallows /wp-admin/, leaving the rest of the site open to crawlers that are not blocked by the server.
• Sitemap is published and accessible — A WordPress sitemap with 19 URLs is available, helping crawlers discover pages.
• Landing page contains rich brand portfolio content — The /landing-page/ details acquisitions of Snuggle Me Organic, Tutti Bambini, and iL Tutto, providing valuable context for AI crawlers.
• Press page lists media mentions from reputable outlets — The press page includes mentions from The Times, Business Insider, and Financial Times, providing external credibility signals.
• Supply terms and conditions page is accessible — A terms of supply page is published, adding legal transparency.
• Greenhouse job descriptions page exists — A page listing job descriptions is available, supporting recruitment efforts.
• DNS and HTTP headers are properly configured — The site has valid DNS records and returns 200 status with nginx, indicating basic infrastructure health.