changeplease.org — AI Site Grade

Analysis

The live site is entirely inaccessible to every AI crawler — and every human browser — blocked behind a Cloudflare JS challenge wall that returns 403 to all user-agents without exception.

Crawler Access

Every AI crawler tested — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, Bytespider, Applebot-Extended, anthropic-ai, ChatGPT-User, Perplexity-User — receives a 403 with a Cloudflare JS challenge page from changeplease.org. The browser baseline also gets 403. The robots.txt (archived from Shopify) contains no AI-bot-specific directives — it uses the default Shopify template with User-agent: * disallowing admin, cart, checkout, search, and policy paths, but allows the rest. The llms.txt returns the same Cloudflare 403 wall. The sitemap is also 403-blocked. The site runs on Shopify (DNS points to 23.227.38.65, Shopify's IP) behind Cloudflare with managed challenge mode enabled, meaning no crawler or browser can access any page content without solving a JavaScript challenge.

Cold-Knowledge Gap

An LLM queried cold about changeplease.org describes it as a micro-donation platform that rounds up purchases for charity — a description that matches no known entity and bears zero resemblance to the actual site. The same LLM queried about "Change Please Coffee" correctly describes a London-based social enterprise coffee company founded in 2015 by Cemal Ezel that trains and employs people experiencing homelessness as baristas, with a B Corp certification and a charitable foundation. The domain changeplease.org hosts this coffee business, but the LLM's cold knowledge associates the domain with a phantom micro-donation app — a complete category error. The actual brand (specialty coffee, barista training, homeless support) is well-documented in the model's training data, but the domain name triggers a wrong association.

Content & Schema Posture

The archived homepage (Wayback Machine, April 2026) reveals a Shopify-powered ecommerce site selling specialty coffee blends (Tall, Dark & Handsome, Happy Go Lucky, Welcome Guest, Adventurous Blend) with subscriptions and merchandise. The site contains zero JSON-LD schema of any type — no Organization, Product, LocalBusiness, or FAQPage markup. The Foundation page (/pages/changepleasefoundation-2024) reports concrete impact numbers: 214 people trained, 11,370 hours of living-wage employment, 3,204 guests on the Driving for Change buses. Despite having FAQ content, impact reports, testimonial quotes, and product listings, the site deploys no structured data to help AI engines understand its social-enterprise model, product catalog, or charitable mission.

External Signals

Web search returns zero indexed results for changeplease.org, the brand name "Change Please," or any combination of its mission terms — a complete absence from DuckDuckGo's index. The Wayback Machine shows 1,995 captures since 2015, indicating a long-standing site with rich history, but the live Cloudflare wall has effectively de-indexed the domain. The brand has a Facebook page, Instagram, LinkedIn, and TikTok presence (visible in archived footer links), but none of these external signals are discoverable through standard web search. The site's Shopify robots.txt allows most AI crawlers to access product and content pages — but the Cloudflare challenge layer above Shopify blocks them before they ever reach the server.

Findings

1. Cloudflare JS challenge blocks all AI crawlers and browsers High

   Every AI crawler tested (GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, Bytespider, Applebot-Extended, anthropic-ai, ChatGPT-User, Perplexity-User) receives a 403 with a Cloudflare JS challenge page. The browser baseline also gets 403. No crawler can access any page content.

   What to change: Disable the Cloudflare JS challenge for known AI crawler user-agents, or serve a static HTML version of the site to bots.

2. LLM cold knowledge misidentifies domain as micro-donation app High

   An LLM queried about changeplease.org describes it as a micro-donation platform that rounds up purchases for charity, which is a complete category error. The actual site is Change Please Coffee, a London-based social enterprise coffee company. The domain name triggers a wrong association despite the brand being well-documented in training data.

   What to change: Add clear structured data (Organization, LocalBusiness) and publish an llms.txt file to correct the AI's understanding of the domain.

3. Zero indexed search results for domain and brand High

   Web search returns zero indexed results for changeplease.org, the brand name 'Change Please', or any combination of its mission terms. The site is completely absent from DuckDuckGo's index.

   What to change: Remove the Cloudflare JS challenge to allow search engine crawlers to index the site, and submit the sitemap to search engines.

4. No JSON-LD schema on any page High

   The archived homepage and foundation page contain zero JSON-LD schema of any type — no Organization, Product, LocalBusiness, or FAQPage markup. Despite having FAQ content, impact numbers, and product listings, the site deploys no structured data to help AI engines understand its social-enterprise model.

   What to change: Add JSON-LD schema for Organization, LocalBusiness, Product, and FAQPage to all relevant pages.

5. llms.txt returns 403 and is inaccessible Medium

   The llms.txt file at https://changeplease.org/llms.txt returns a 403 Cloudflare challenge, making it impossible for AI crawlers to discover the site's content via this standard.

   What to change: Serve llms.txt without Cloudflare challenge, or host it on a separate static domain.

6. Sitemap returns 403 and is inaccessible Medium

   The sitemap at https://changeplease.org/sitemap.xml returns a 403 Cloudflare challenge, preventing search engines and AI crawlers from discovering the site's URL structure.

   What to change: Serve sitemap.xml without Cloudflare challenge, or host it on a separate static domain.

7. Robots.txt lacks AI-specific directives Low

   The archived robots.txt uses the default Shopify template with User-agent: * disallowing admin, cart, checkout, search, and policy paths, but contains no AI-bot-specific directives. However, the Cloudflare challenge makes robots.txt irrelevant.

   What to change: Add explicit allow/disallow rules for AI crawlers in robots.txt, but the primary fix is removing the Cloudflare challenge.

8. No external signals discoverable via web search Medium

   Despite having Facebook, Instagram, LinkedIn, and TikTok presence (visible in archived footer), none of these external signals are discoverable through standard web search. The brand has zero search presence.

   What to change: Ensure social media profiles are linked from the site and indexed, and build backlinks from reputable sources.

What's working

• Archived site contains rich content and impact data — The Wayback Machine snapshot shows a well-maintained Shopify site with detailed product descriptions, impact numbers (214 people trained, 11,370 hours of employment), and a foundation page with concrete metrics. This content is valuable for AI understanding once access is restored.
• Change Please Coffee brand is well-documented in LLM training data — LLMs correctly describe Change Please Coffee as a London-based social enterprise founded in 2015 by Cemal Ezel, with B Corp certification and a charitable foundation. The brand's mission and impact are accurately represented in training data.
• Shopify robots.txt allows most AI crawlers on content pages — The default Shopify robots.txt does not disallow AI crawlers from accessing product and content pages, meaning once the Cloudflare challenge is removed, crawlers will be able to index the site's main content.
• Site has long history with 1,995 Wayback captures since 2015 — The Wayback Machine shows 1,995 captures since 2015, indicating a long-standing site with rich historical content that can be used for recovery and indexing.