asana.com — AI Site Grade

Analysis

The OS for human-agent teams — but AI crawlers can't read the manual

Asana's homepage aggressively positions the brand as "the OS for human-agent teams" and the blog contains deep engineering content about custom LLM evaluation frameworks and AI Teammate memory systems, yet every single blog article URL returns a 404 to plain HTTP requests — the blog is a Next.js JS shell with no server-rendered article content.

Crawler Access

All 11 AI crawlers tested (GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, ChatGPT-User, anthropic-ai, Bytespider, Applebot-Extended, Perplexity-User) receive 200 with full content on the homepage — identical byte size (729KB) to browser baseline. No UA-based blocking exists. The site runs on Netlify with Next.js server-side rendering, so core marketing pages render full HTML. However, the robots.txt contains zero AI-bot-specific rules — a single User-agent: * block governs everything, disallowing /asanapedia/, /ebook/, and /resources/thank-you but offering no guidance to AI crawlers about preferred entry points. The llms.txt at /llms.txt is present and well-structured, listing 30+ core resources with descriptions and URLs — a strong positive signal.

Cold-Knowledge Gap

The cold LLM knows Asana as a "cloud-based project management platform" founded by Facebook co-founders, with features like Tasks, Timelines, Portfolios, and Goals. It mentions "Asana Intelligence (AI) features like Smart Suggestions and AI-generated status updates" from 2023-2024. This is significantly behind the site's current positioning. The homepage now leads with "The OS for human-agent teams" and "AI Teammates" — autonomous AI agents that "drive marketing campaigns, manage IT tickets, support operations." The cold model knows nothing about AI Teammates, AI Studio, the StackAI acquisition, or the "human-agent collaboration" narrative that dominates every page. The gap between what the model knows (task management tool with some AI features) and what the site sells (an agentic work OS) is wide.

Schema Posture

The homepage carries Organization, WebSite, and BreadcrumbList schema — solid but generic. The /product/ai page adds SoftwareApplication (with applicationCategory: "BusinessApplication" and applicationSubCategory: "Project Management Software") and a comprehensive FAQPage with 9 questions covering AI pricing, data security, partner models, and data retention. The /pricing page includes Product schema with Offer entries for Personal, Starter, Advanced, and Enterprise tiers. Missing: no HowTo or TechArticle schema on use-case pages, no VideoObject for the prominent demo CTA, and no WebPage schema with about linking to the AI-specific entities the site now centers on.

JS-Rendered Content Gap

The most critical finding: the Inside Asana blog listing page renders article titles in the HTML, but every individual article URL tested (/inside-asana/how-ai-teammates-build-memory..., /inside-asana/how-asana-built-a-custom-llm-evaluation-framework..., /inside-asana/ai-agents-built-for-teams...) returns 404 with a "This page doesn't exist" error page. These articles — which include detailed engineering content about LLM evaluation frameworks, AI agent memory systems, and enterprise AI transparency — are only accessible via client-side JavaScript routing. AI crawlers that do not execute JS cannot read them. The sitemap does not include blog articles. The press release page about the StackAI acquisition is similarly a JS shell with no article body in the initial HTML. Asana is publishing rich AI-agent technical content that no AI crawler can consume.

Findings

1. Blog articles about AI Teammates return 404 to AI crawlers High

   Every individual blog article URL tested returns a 404 error page with no article body in the initial HTML. The blog is a Next.js JS shell that requires client-side JavaScript to render content, making the detailed engineering content about LLM evaluation frameworks and AI agent memory systems invisible to AI crawlers that do not execute JavaScript.

   What to change: Implement server-side rendering (SSR) or static generation for blog article pages so that the full article content is included in the initial HTML response. Ensure the sitemap includes blog article URLs.

2. Cold LLM knowledge lags behind current AI Teammates positioning High

   The cold LLM knows Asana as a project management tool with basic AI features like Smart Suggestions, but knows nothing about AI Teammates, AI Studio, the StackAI acquisition, or the human-agent collaboration narrative that dominates the current site. This gap means AI agents cannot accurately represent Asana's current product capabilities.

   What to change: Publish structured data (e.g., FAQPage, SoftwareApplication with updated description) and authoritative content about AI Teammates and AI Studio on pages that are crawlable and indexable. Consider submitting updated product information to knowledge panels.

3. Press release about StackAI acquisition is a JS shell High

   The press release page about the StackAI acquisition returns a 200 status but the article body is not present in the initial HTML; it requires JavaScript to render. AI crawlers that do not execute JavaScript cannot read the content of this important announcement.

   What to change: Implement server-side rendering for press release pages so that the full article content is included in the initial HTML response.

4. Missing HowTo and TechArticle schema on use-case and blog pages Medium

   The site lacks HowTo schema on use-case pages and TechArticle schema on blog articles. The homepage and product pages have Organization, WebSite, and SoftwareApplication schema, but the AI-specific content is not semantically marked up to help AI crawlers understand the context.

   What to change: Add HowTo schema to use-case pages and TechArticle schema to blog articles. Consider adding WebPage schema with about property linking to AI-specific entities.

5. Robots.txt provides no AI crawler-specific guidance Low

   The robots.txt file contains a single User-agent: * block with no AI-bot-specific rules. While no bots are blocked from key pages, the lack of guidance means AI crawlers may waste resources on disallowed paths like /asanapedia/ and /ebook/.

   What to change: Add specific rules for AI crawlers (e.g., GPTBot, ClaudeBot) to allow access to key content and disallow low-value paths.

6. Sitemap does not include blog article URLs Medium

   The sitemap fetched from /sitemap-index.xml contains 80 URLs but none are blog article URLs. The blog subdomain sitemap at /sitemap.xml returns 404. This prevents AI crawlers from discovering blog content through sitemaps.

   What to change: Include blog article URLs in the sitemap. Ensure the blog subdomain has a valid sitemap.xml.

7. Missing VideoObject schema for demo video on homepage Low

   The homepage prominently features a demo video but lacks VideoObject schema markup. This reduces the chance of the video appearing in rich results or being understood by AI crawlers.

   What to change: Add VideoObject schema to the demo video on the homepage.

8. No WebPage schema with about property linking to AI entities Medium

   The site does not use WebPage schema with an about property to explicitly indicate that pages are about AI Teammates, AI Studio, or other AI-specific entities. This makes it harder for AI crawlers to associate content with the correct entities.

   What to change: Add WebPage schema with about property referencing AI-specific entities (e.g., AI Teammates, AI Studio) on relevant pages.

What's working

• llms.txt file is present and well-structured — The site hosts a /llms.txt file with 30+ core resources, descriptions, and URLs, providing a clear entry point for AI crawlers to discover key content.
• All 11 tested AI crawlers receive 200 with full content on homepage — The homepage returns full HTML content to all tested AI crawlers with no UA-based blocking, ensuring core marketing pages are accessible.
• Comprehensive FAQPage schema on /product/ai page — The /product/ai page includes a detailed FAQPage schema with 9 questions covering AI pricing, data security, partner models, and data retention, helping AI crawlers extract structured Q&A content.
• SoftwareApplication schema on /product/ai page — The /product/ai page includes SoftwareApplication schema with applicationCategory and applicationSubCategory, providing structured product information to AI crawlers.
• Product schema with Offer entries on pricing page — The /pricing page includes Product schema with Offer entries for each plan tier, providing structured pricing data to AI crawlers.
• Core marketing pages are server-side rendered with full HTML — The homepage, /product/ai, /pricing, and other marketing pages are server-side rendered, providing full HTML content to AI crawlers without requiring JavaScript execution.
• Trust page provides detailed security and compliance information — The /trust page contains 1078 words of security and compliance content, which is valuable for AI crawlers seeking trust signals.