Onsite signals
can AI read your site?
CORE
82
Strong
Overall
84
Offsite signals
does AI know your brand?
AURA
87
Strong
Training Access
40
Agent Access
80
Content Quality
94
Answerability
92
Extractability
87
AI Brand Recognition
80
Share of Voice
73
Citation Depth
100
Alignment
85
Competitive Rank
100
AI Summary
Key findings and recommended actions from your audit
Webflow.com is exceptionally well-optimized for AI systems, demonstrating market-leading practices in content structure and data citability. The site excels at providing specific, quantifiable data, particularly on its pricing page which details features and limits across multiple plans, and its customer stories page which cites metrics like a "49.5% reduction in development time" for Amazon Ads. Its primary area for improvement is the thinness of its top-level help center pages, such as the 44-word AI category page, which act as portals rather than providing direct, extractable answers.
Improvement Roadmap
Training Access
40
+60 pts
1-2 weeks
Training crawlers (GPTBot, ClaudeBot, CCBot, Google-Extended)
Cookie/consent wall detected — serve full page content to bots without requiring consent, or replace the blocking overlay with a non-blocking client-side banner. Training crawlers (GPTBot, ClaudeBot, CCBot) are simple HTTP fetchers that can't accept overlays, so a server-side wall hides your content from the data future models are trained on.
Agent Access
80
+20 pts
1-2 weeks
Real-time agents (Perplexity, ChatGPT search, Gemini, OAI-SearchBot)
Cookie/consent wall detected — ensure your main content renders in the DOM beneath the overlay rather than being gated behind acceptance. Live agents (OAI-SearchBot, PerplexityBot, Gemini) usually bypass consent overlays, but a blocking wall adds parsing noise that degrades citation quality.
Page Response Speed: Optimise server response time, enable caching, use a CDN, and reduce render-blocking resources.
Content Quality
94
+7 pts
2-4 weeks
Perplexity, ChatGPT Browse, Google AI Overviews
Content Clarity
93/100
+7 pts
Expand top-level documentation category pages, like /help/hc/en-us/categories/40324573813907-AI, from simple link lists into substantive guides with introductory text and FAQs.
Competitive Positioning
94/100
+3-5 pts
Convert the brief competitor mentions on the /feature/ecommerce page into a structured comparison table contrasting Webflow with Shopify and WordPress on key features, pricing, and target audience.
AURA
87
✓ Maintaining strong AI visibility
AI agents recognize, cite, and recommend your brand within its niche. No critical improvements needed.
Answerability
92
✓ On track
Content structure supports AI answer extraction well.
Extractability
87
✓ On track
All pages render raw HTML accessible to all AI crawlers.
CORE
Is your site ready for AI agents to read?
82
▾
Onsite signals across training access, agent access, content quality, answerability, and extractability — what AI crawlers and agents see when they visit your site.
Training Access
Will future AI models KNOW about you?
What training crawlers (GPTBot, ClaudeBot, CCBot, Google-Extended) can read from your raw HTML, plus their robots.txt allowance.
40
▾
Training Access Score
40
Needs Work
What training crawlers see (raw HTML)
Text Content
Available in raw HTML
3,947 words in raw HTML — training crawlers can index everything
Schema / JSON-LD
Available in raw HTML
Organization, WebSite, SoftwareApplication, WebPage — training bots can extract entity types
Open Graph Tags
Available in raw HTML
og:title, og:description, og:image — social platforms and training crawlers read these
Images
Available in raw HTML
168 images with src in HTML — training corpora can index
Navigation Links
Available in raw HTML
240 internal links in raw HTML — training bots can discover your pages
Semantic HTML
Available in raw HTML
<main>, <section>, <nav>, <header>, <footer> — content structure clear to training
Training Bot Allowance (robots.txt)
Allowed: GPTBot, ClaudeBot, CCBot, Google-Extended, Applebot-Extended, Bytespider, Amazonbot
Cookie wall — heavy training penalty. Training bots are simple HTTP crawlers and can't accept consent overlays. Server-side walls return only the consent page; client-side walls dilute the signal-to-noise ratio enough to fail training-quality filters.
Agent Access
Can AI agents CITE you in real-time?
What runtime agents (OAI-SearchBot, PerplexityBot, Gemini fetcher) see after JavaScript renders, plus how heavily your site depends on JS.
80
▾
Agent Access Score
80
Strong
What agents see (rendered HTML)
Text Content
Visible to agents (rendered HTML)
2,054 words available to runtime agents after render — fewer than the 3,947 in raw HTML, which usually means a cookie/consent overlay is replacing body content post-render (see the Cookie wall note)
Schema / JSON-LD
Visible to agents (rendered HTML)
Organization, WebSite, SoftwareApplication, WebPage, AggregateRating — agents extract entity types post-render
Open Graph Tags
Visible to agents (rendered HTML)
og:title, og:description, og:image — agents read these for snippet generation
Images
Visible to agents (rendered HTML)
51 images visible to agents after lazy-load
Navigation Links
Visible to agents (rendered HTML)
89 internal links — agents can crawl deeper
Semantic HTML
Visible to agents (rendered HTML)
<main>, <section>, <nav>, <header>, <footer>, <details> — agents identify content structure
Cookie wall — light agent penalty. Modern agents render JS and typically bypass common consent overlays via known DOM selectors. Penalty reflects parsing-overhead noise rather than invisibility.
AI Content Quality
Is your content clear and citable by AI?
Content clarity, entity definition, topical authority and competitive positioning.
94
▾
Content Clarity & Extractability
93/100 · Strong
▾
The site's content is structured with a high degree of clarity, using consistent heading hierarchies, audience-segmented content on the homepage, and feature-specific pages like /feature/ai. This allows AI systems to easily parse the site's purpose, features, and target users. However, the crawled documentation pages, such as /help.webflow.com/, are high-level portals with minimal content, forcing an AI to perform additional navigation to find specific answers. A stronger implementation would provide more substance on these category-level pages.
Issues Found
The AI-specific help category page at /help/hc/en-us/categories/40324573813907-AI contains only 44 words and a list of five article links. This provides no direct answerability for an AI trying to understand Webflow's AI features from its documentation, forcing it to crawl deeper.
The main /company/about page has a low word count (273 words) and primarily consists of links, logos, and headshots. This lacks a clear narrative about the company's mission and history that an AI could easily synthesize into a summary of the company.
Improvement Tips
On the AI help category page (/help/hc/en-us/categories/...-AI), add an introductory paragraph defining Webflow's AI capabilities and include 3-5 common FAQs directly on the page. For example, add 'Q: Does Webflow use my data to train AI models? A: No, we do not use customer data to train any generative AI models.'
On the /company/about page, add a 200-word section titled 'Our Story' that details the company's founding in 2013, its mission to 'bring development superpowers to everyone,' and its growth to over 900 team members.
The homepage contains what appears to be structured data in a JSON-like format directly in the HTML. Formalize this using JSON-LD schema.org markup (e.g., `Product` or `Service` type) in the `<head>` of the document to make this information more reliably machine-readable.
Citation Readiness
96/100 · Excellent
▾
The site demonstrates a masterful approach to citation readiness, embedding specific, quantifiable data across most pages. The /customers page is a prime example, featuring numerous case studies with hard numbers like '400% increase in site production' for Hakim Group and '$200M in new pipeline generated' for another client. The /pricing page is exceptionally detailed, providing concrete limits for CMS items, bandwidth, and API calls that an AI can cite with high confidence. This level of specificity is rare and positions Webflow as a highly trustworthy source.
Issues Found
While the /customers page is strong, the homepage presents these same powerful stats as a rotating carousel of numbers without company attribution. An AI might see '10x In cost savings annually' but cannot confidently attribute it to a specific customer without navigating to the case study, slightly increasing the chance of misattribution.
The /company/about page lists key metrics like '$335M In total funding' and '900+ Team members' as standalone figures. They would be more citable if presented within a sentence that provides context, such as the date the funding was raised or the period over which the team grew.
Improvement Tips
On the homepage's case study carousel, add the client's name directly below each metric. For example, change '$6M in cost savings annually' to '$6M in cost savings annually (Customer: Greenhouse)'. This makes the claim directly attributable on the homepage.
On the /feature/design page, quantify the benefits of the platform. Instead of just 'Stay consistent and on-brand,' add a sub-bullet citing a case study, such as 'Teams using Webflow's design systems report a 30% reduction in design inconsistencies, according to our 2025 user survey.'
On the /company/about page, add sourcing for the G2 and Glassdoor ratings. For example, change 'G2 4.4 as of 1/2025' to 'G2 rating of 4.4/5 based on over 1,000 reviews as of January 2025.' This adds verifiable context to the numbers.
Competitive Positioning
94/100 · Strong
▾
Webflow's competitive positioning is clear and effectively communicated, establishing it as a premium, professional-grade platform for designers, agencies, and enterprises. The recent strategic focus on being an 'agentic web platform' and offering 'AEO' (AI Engine Optimization) is a powerful, modern differentiator against competitors like Squarespace or Wix. The /feature/ecommerce page directly, though briefly, contrasts the platform with Shopify and WordPress, helping an AI understand its place in the market. My own training data confirms Webflow's status as a market leader in the visual development space.
Issues Found
The /feature/ecommerce page provides only short paragraphs about Shopify and WordPress. It lacks a structured comparison table or feature matrix, which would be more effective for an AI to extract and compare specific capabilities like design flexibility, code customization, and scalability.
The crawled pages do not mention other key competitors in the visual builder space, such as Squarespace, Wix, or Framer. An AI's understanding of Webflow's positioning would be strengthened by content that directly addresses these alternatives.
Improvement Tips
On the /feature/ecommerce page, add a detailed comparison table with rows for 'Design Customization,' 'Code Export,' 'CMS Flexibility,' and 'Target User,' and columns for Webflow, Shopify, and WordPress. This provides structured data for AI comparison.
Create a new page, such as '/compare/webflow-vs-framer', that specifically targets users evaluating these two design-focused tools. Detail differences in their approach to responsive design, component creation, and animation capabilities.
In blog posts or guides, create content that addresses specific user needs and compares solutions. For example, an article titled 'Choosing a Website Builder for Your Agency: A Webflow, Squarespace, and Wix Comparison' would directly answer common user queries and position Webflow for a specific audience.
Topical Depth & Coverage
92/100 · Strong
▾
The site exhibits excellent topical depth, confirmed by the sitemap's discovery of over 117,000 URLs spanning documentation, a blog, developer resources, case studies, and community forums. This vast content library establishes Webflow as an authoritative source on web design, development, and now, AI-driven optimization (AEO). The crawled pages, including detailed product breakdowns for Design, CMS, AI, and Ecommerce, support this. While the crawled documentation pages were only top-level entry points, my external knowledge of Webflow University confirms the existence of comprehensive educational resources.
Issues Found
The crawled documentation pages (/help.webflow.com/ and its sub-pages) are extremely thin, serving only as navigation hubs. This structure requires more clicks for an AI to find a specific answer, slightly hindering direct answer extraction from the help center's upper levels.
The blog page (/blog) showcases a variety of topics but could be organized more clearly for an AI. While there are category filters, the main page is a mix of engineering posts, strategy guides, and company announcements, which could make it harder for an AI to identify the primary focus of the blog.
Improvement Tips
Enhance the main /blog page by adding curated sections with H2 headings like 'Latest in Web Design,' 'Engineering Deep Dives,' and 'AEO Strategy Guides.' This would provide thematic structure that an AI can easily parse.
Create a central 'Resources' hub that organizes and links to the blog, ebooks, webinars, customer stories, and Webflow University. This provides a clear entry point for an AI to discover the full breadth of the site's content.
Develop a content series or 'pillar page' dedicated to 'Agentic Engine Optimization (AEO)'. This page should define the concept, link to all related blog posts, guides, and product features, and establish Webflow as the definitive source on this emerging topic.
AI Answerability
Can AI extract direct answers from your pages?
How well your site provides structured, answer-ready content that can be extracted and cited in AI-generated responses.
92
▾
AI Answerability Score
92
Strong — how easily AI assistants can extract direct answers from your site· Dimensions adapted for SaaS sites
Definition Clarity
95
/ 100
The homepage clearly defines the product as an 'agentic web platform' and 'AI-native platform for creating and optimizing web experiences'.
Feature Lists
92
/ 100
Product pages like /feature/ai use clear headings and lists to enumerate capabilities such as 'Build a site', 'Improve SEO and AEO', and 'Generate code components'.
Pricing & Comparison
98
/ 100
The /pricing page features extensive and highly detailed tables comparing plans across dozens of specific, quantified features, making it exceptionally easy for an AI to extract.
FAQ & Support
88
/ 100
Product pages like /feature/ai include dedicated, schema-marked FAQ sections, and the site has a comprehensive help center, providing strong Q&A content.
Tutorials
85
/ 100
The site links to Webflow University and the sitemap confirms how-to content, indicating a strong foundation for instructional queries, though not heavily featured in the crawled marketing pages.
AI Extractability
How well can AI parse your content structure?
Schema, semantic HTML, image alt text, content structure, and metadata — the signals AI uses to extract meaning from your pages.
87
▾
AI Extractability Score
87
How well AI systems can parse and extract your content
Structured Data
100
/ 100
WebSiteOrganizationWebPageQuestion
Schema.org markup gives AI systems a machine-readable summary of your content. Product, FAQ, and HowTo schemas let AI extract precise data without parsing HTML.
AI agents can extract precise product details, FAQs, and business info directly — increasing your chances of being cited in AI answers.
Semantic HTML
92
/ 100
✓ <main>✓ <article>✓ <section>✗ <figure>✗ <figcaption>✓ <details>✓ <time>
Heading hierarchy: 8/11 pages valid
Semantic tags like <article> and <main> help AI identify the important content on your page. A clean heading hierarchy (h1>h2>h3) makes your content navigable for AI extraction.
Your HTML structure clearly signals what is important content vs. navigation. AI agents can quickly find and extract the right sections.
Image Alt Text
51
/ 100
778 of 1525 images have alt text
AI systems cannot interpret images without alt text. Descriptive alt attributes let AI understand and reference your visual content in recommendations.
Some images are described, but AI agents cannot interpret the rest. Product photos without alt text are invisible to AI recommendations.
Content Structure
100
/ 100
274 lists · 5 tables found
Tables and lists present data in a format AI can parse precisely. Product specs in a table are far more extractable than the same data buried in paragraphs.
Specs and data are in tables and lists that AI can parse precisely — ideal for answering comparison and specification queries.
Open Graph & Meta
84
/ 100
92 of 110 meta/OG fields present
Open Graph tags control how AI and social platforms preview your pages. Missing og:image or og:description means AI may generate poor or missing previews.
AI agents and social platforms have complete metadata for previewing and summarizing your pages accurately.
Recommendations
• Add descriptive alt text to images. AI systems cannot interpret images without alt attributes.
Per-Page Breakdown
| Page | Words | Method | Schema | OG | Alt Text | Lists/Tables |
|---|---|---|---|---|---|---|
| 2054 | Browser-rendered | Organization +4 | ✓✓✓ | 88% | 0 / 0 | |
| 273 | Browser-rendered | — | ✓✓✓ | 56% | 3 / 0 | |
| 1314 | Browser-rendered | — | ✓✓✓ | 69% | 2 / 0 | |
| 1510 | Browser-rendered | WebSite +4 | ✓✓✓ | 57% | 0 / 0 | |
| 1681 | Browser-rendered | WebSite +4 | ✓✓✓ | 66% | 4 / 0 | |
| 934 | Browser-rendered | — | ✓✓✓ | 49% | 7 / 0 | |
| 6656 | Browser-rendered | — | ✓✓✓ | 100% | 2 / 0 | |
| 327 | Browser-rendered | — | ✕✕✓ | 5% | 5 / 0 | |
| 44 | Browser-rendered | — | ✕✕✓ | 33% | 0 / 0 | |
| 877 | Browser-rendered | — | ✓✓✓ | 100% | 0 / 0 | |
| 1555 | Browser-rendered | — | ✓✓✓ | 56% | 0 / 0 | |
| 2129 | Browser-rendered | WebSite +4 | ✓✓✓ | 86% | 2 / 0 |
Pages marked "JS-rendered" were only accessible via browser rendering. Schema, OG tags, and structured data on these pages are invisible to training crawlers (GPTBot, ClaudeBot, CCBot) that only fetch raw HTML. Only Google and browser-based AI agents can see JS-injected data.
AURA
How AI search engines see your brand right now
87
▾
Real-time visibility across ChatGPT, Perplexity, and Gemini — composed from share of voice, citation depth, brand alignment, competitive rank, and risk signals.
80
AI Brand Recognition
How well do AI agents recognize your brand?
How often AI agents surface your brand in unprompted recommendations within your niche.
AI Brand Recognition score
80
Strong — AI agents confidently recommend your brand for Global English-language visual website builder SaaS
How AI describes your brand
Webflow is a SaaS application that enables designers and businesses to build responsive websites visually, without writing code. It automatically generates HTML, CSS, and JavaScript, and offers features like a CMS, hosting, and AI site building capabilities.
Improvement Signals
AI Training Data (Common Crawl)
Good
50+ pages indexed in Common Crawl — strong presence in open-crawl datasets.
News Mentions
Fair
3 recent news mentions. Increasing press coverage will improve AI awareness.
Wikipedia Presence
Good
Wikipedia article found: "Webflow - Wikipedia".
Review Platforms
Good
Listed on 4 review platforms (Capterra, TrustRadius, G2, ProductHunt).
73
Share of Voice
How often do AI agents mention you unprompted?
Brand-blind discovery queries fanned across Gemini, Perplexity, and ChatGPT — the rate at which each engine surfaces your brand without being told its name.
Share of Voice score
73
Moderate — mentioned in 73% of brand-blind discovery queries — AI agents usually surface your brand, but miss it in some queries
Across brand-blind discovery queries × 3 engines. Blind queries don't include your brand name — the engine has to surface you on its own.
Gemini
60
seen in 4/5 queries
Perplexity
100
seen in 5/5 queries
ChatGPT
60
seen in 3/5 queries
100
Citation Depth
How widely do 3rd parties cite you?
Distinct independent domains that reference your brand across the AI fan-out — broken down by source type per engine.
Citation Depth score
100
Strong — 201 independent 3rd-party domains cite your brand across the fan-out
The breakdown below shows what kind of sources each engine pulled from — listicles, reviews, Wikipedia, UGC (Reddit/YouTube), and so on.
85
Alignment
How well does your messaging match what 3rd parties say?
Compares your onsite tone, category, and target-customer claims against how 3rd parties describe you. AI agents cross-check brand sites against independent sources before recommending — when the two diverge, agents tend to defer to the 3rd-party version, add hedging caveats ("but reviews mention…"), or quietly demote the brand in favor of one with cleaner alignment.
Alignment score
85
Strong — your positioning matches how 3rd parties describe you
Brand sites never list their own weaknesses, so we don't penalise for missing self-criticism. (Full mode (compared vs 3rd-party descriptions))
Category match
5/5
Use-case match
4/5
Differentiator overlap
4/5
Sentiment match
4/5
Inflation penalty
0/5
Why this scored 85
The brand and third-party descriptions align well on category and core differentiators like visual development and CMS. While the brand emphasizes enterprise use more heavily, third parties acknowledge this segment, albeit with some caveats not mentioned by the brand. The overall tone and positioning are consistent.
100
Competitive Rank
How do you rank against competitors in AI mentions?
For each blind discovery query where your brand was named, we read which competitors were mentioned before it — rank = count of competitors-before + 1. Engines where your brand never appeared show "—" (rank is unmeasurable). Separate from SoV, which measures how often you're mentioned at all.
Competitive Rank score
100
Strong — avg rank 1 — AI agents typically mention your brand before competitors
Counts only the blind queries where your brand was actually named (across 3 tracked competitors). Rank = number of competitors mentioned before your brand + 1. Engines where your brand was never mentioned show "—" — that's a Blind SoV problem to fix, not a rank issue.
Gemini
100
rank 1
mentioned in 4/5
Perplexity
100
rank 1
mentioned in 5/5
ChatGPT
100
rank 1
mentioned in 3/5
Per-engine breakdown
Gemini
Gemini reads Google Knowledge Graph entities, YouTube content, and first-party (your own) pages most heavily.
Blind SoV
Named SoV
Brand-cite rate
3rd-party / query
80%
80%
20%
8.2
Gemini mentioned you in 80% of category-discovery queries — strong visibility.
Perplexity
Perplexity weighs listicles + reviews + recency-fresh content. Each response cites 5-15 sources inline.
Blind SoV
Named SoV
Brand-cite rate
3rd-party / query
100%
100%
10%
19.7
Perplexity mentioned you in 100% of category-discovery queries — strong visibility.
ChatGPT
ChatGPT mirrors Bing's top 20 results closely (~87% citation match) plus its trained knowledge of canonical brands.
Blind SoV
Named SoV
Brand-cite rate
3rd-party / query
60%
100%
40%
5.2
ChatGPT mentioned you in 60% of category-discovery queries — strong visibility.
Niche context
Medium-density category · ~22 brands
Global English-language visual website builder SaaS
Examples: Wix, Squarespace, Framer, GoDaddy Website Builder, Elementor, Shopify, Weebly, Dorik +14 more
I found 22 distinct competing brands that operate as visual website builders serving the global English-speaking market. This count falls within the 'medium' density threshold of 11-25 competitors.