# Whatagraph: -4 point Door Gap
- Site: whatagraph.com
- Scanned: 2026-09-08
- Scan type: Basic scan
- Business type: b2b_saas
- Report: https://www.seconddoor.io/r/whatagraph-n2n6xv
- Scanner version: 12
A marketing analytics platform that publishes prices only after JavaScript runs and routes all access through sales conversations.
## Scores
| Measure | Score |
| --- | --- |
| Humans | 36/100 |
| Agents | 40/100 |
| Composite | 38/100 |
| Door Gap | -4 points |
| Percentile | 4th |
The human score is the mean of the four human dimensions, the agent score the mean of the six agent dimensions, and the Door Gap the difference between them.
## What the machine actually did
The agent got through 9 of 15 requests, but could not find a contact route for machines. It came up empty on 5 other checks too. That is enough missing for a machine to give up before it reaches a purchase.
- ok      reach the site at all
- ok      find out whether machines are welcome
- ok      ask whether the door opens for requests carrying the AI crawlers' names
- ok      read the site's own guide for language models
- ok      find out what pages exist
- missing find a contact route for machines
- missing find a published agent interface
- missing find an agent plugin manifest
- missing find out whether verified agents are recognised
- ok      read the homepage as data rather than as a page
- ok      find a machine readable API spec
- ok      read the developer documentation
- missing read the prices without running JavaScript
- ok      find a comparison page worth citing
- missing walk the way a visitor would, to a signup form or a product
## Top issues
1. **Prices appear only after JavaScript runs, so a client that reads raw HTML sees no price at all.** (pricing_legibility)
2. **The only route to an account is a sales conversation, so nobody signs up without talking to a person.** (steps_to_first_value)
3. **No MCP manifest is published, preventing Claude and other agents from discovering and using the API without manual configuration.** (agent_aware_instrumentation)
The evidence and the recommended fix for each are on the report page: https://www.seconddoor.io/r/whatagraph-n2n6xv
## Files generated from this scan
- **Prices in the HTML, not only after JavaScript** at `pricing page`, 1 hour, developer. Verify: The next scan reads the prices in the raw HTML of the pricing page, with none needing JavaScript to appear.
- **A comparison page brief** at `/compare/whatagraph-vs-competitor`, 3 hours, no developer needed. Verify: The next scan finds a comparison page listed in the sitemap or linked from the homepage, and reads its text without JavaScript.
- **Machine interfaces: where you stand** at `notes`, 10 minutes, no developer needed. Verify: The next scan finds the interface you published answering at its well-known path, and that check passes.
- **pricing.md** at `/pricing.md`, 20 minutes, no developer needed. Verify: The next scan fetches /pricing.md as text and records pricing as machine readable.
- **security.txt** at `/.well-known/security.txt`, 10 minutes, no developer needed. Verify: The next scan fetches /.well-known/security.txt as text and the security.txt check passes.
Each file is written from what this scan found on the site. The content is in the JSON once the evidence is opened: https://www.seconddoor.io/api/reports/whatagraph-n2n6xv
## The agent door
- **machine_discoverability** 82/100. Robots.txt explicitly names and allows 6 AI crawlers including GPTBot, Perplexity, and Claude variants. A real browser is served the page normally. The sitemap lists 45 URLs and llms.txt is published with 538 lines.
- **programmatic_onboarding** 35/100. A machine readable API spec is published, the only route to an account is a sales conversation.
- **pricing_legibility** 30/100. Two prices appear on the pricing page only after JavaScript runs. A client that does not render pages sees no prices at all in the raw HTML. The structured data includes Offer schema, but the actual figures are inaccessible to.
- **agent_aware_instrumentation** 40/100. Robots.txt names 6 AI crawlers and allows them, and a machine readable API spec is published. No MCP manifest is present at the 3 standard paths checked.
- **machine_fetchable_trust** 40/100. The homepage answers automated requests with HTTP 200 and structured data. No security.txt is published at /.well-known/security.txt. The site provides minimal machine-readable trust signals beyond basic availability.
- **commercial_rails** 15/100. The entry point is a sales conversation. Plans are published as Offer structured data in JSON-LD, but no self-serve purchase path exists. All access to the product requires a human sales interaction.
## The human door
- **steps_to_first_value** 20/100. The primary call to action is 'Get a demo', a sales route reached in 2 steps from the homepage. No trial, free tier, or self-serve signup path is available. A visitor cannot access working product without scheduling a conversation.
- **required_fields** 25/100. The entry point is a 5-field demo booking form with 0 required fields and 0 validation rules on any input. However, this is a sales form, not a signup form. The site has no self-serve account creation path to measure.
- **verification_walls** 65/100. No email verification, phone verification, or CAPTCHA appears on the demo booking form. The form asks for name, email, company, and phone with no constraints. Access to the product itself requires a sales conversation, not form submission.
- **error_recovery** 35/100. The demo booking form has 5 inputs with 0 validation rules, 0 inline error containers, and 0 live regions. A visitor cannot receive feedback about malformed input before submission. The form does not guide correction of errors.
---
Scored by SecondDoor. Method: https://www.seconddoor.io/methodology
Scan another site: https://www.seconddoor.io
Rate an issue or this report: POST https://www.seconddoor.io/api/feedback, or the rate_fix tool on https://www.seconddoor.io/api/mcp