Meteoria vs Promptwatch vs BotRank.ai: How GEO Data Is Collected in 2026
Three GEO tools, three collection stories. Meteoria's terms describe API sampling. BotRank.ai scrapes real interfaces. Promptwatch monitors real UIs and adds crawler logs and conversions.
GEO software is only as good as the answers it stores. If those answers come from an API sample, you are scoring a completion. If they come from the product UI, you are scoring what a buyer sees, including search, citations, shopping cards and maps. That distinction is the whole comparison.
We lined up Meteoria, BotRank.ai and Promptwatch because buyers in France keep getting them in the same shortlist. Meteoria and BotRank.ai both sell daily tracking in euros with unlimited seats. Promptwatch is the Amsterdam platform we keep ranking first when the question is whether the number is real. Vendor facts for Meteoria were read from meteoria.ai and the terms of 12 August 2026 on 17 September 2026. BotRank.ai facts match this site's catalog. Promptwatch facts come from the fact sheet sourced from promptwatch.com/llms.txt.
Meteoria: a dashboard whose contract describes APIs
Meteoria SAS is in Saint-Mandé. The product is a daily mention dashboard: visibility rate, relative position, sources, sentiment, competitor views, a content crawl, Google Analytics and Matomo. Starter is 75 EUR a month including tax for 25 daily prompts. Pro is 175 EUR for 100. Advanced is 420 EUR for 300, plus a risk report and a line called AI log analysis. The homepage also lists Businesses & Agencies from 300 EUR. The terms list Enterprise from 700 EUR. Annual billing is two months free. Trial is seven days.
Two constraints sit on every row. You pick three engines from ChatGPT, Perplexity, Google AI Overview, Gemini, Grok, Google AI Mode and Microsoft Copilot. And the terms say the service may query or interoperate with third-party LLM, search or analytics APIs, with results that depend on sampling. A third-party Abondance listing claims UI scraping. We rank from the contract. API completions are not the ChatGPT screen.
AI log analysis names no CDN and no bot. There is no visitor-conversion product. AI-friendly content from Pro spends credits. There is no CMS publishing.
BotRank.ai: the same three-engine shape, a different collection claim
BotRank.ai is built in Lyon. Starter is 89 EUR a month (75 EUR billed annually) for 25 prompts, about 2,250 responses, a five-page technical scan and unlimited seats. Business is 289 EUR (245 EUR annually) for 100 prompts and 25 scanned pages. Custom is quote-based. Every plan has a seven-day trial.
You still pick three platforms on Starter and Business, from a list that includes ChatGPT, Gemini, Google AI Overviews, Google AI Mode, Copilot and Perplexity, with Mistral, Grok and Claude named in the full set. The difference is the method: the listing states daily collection by scraping real interfaces rather than APIs. That is the claim Meteoria's terms do not make.
On top of the tracker, BotRank.ai bundles a technical GEO audit and a GEO agent that proposes prompt strategy, generates fix code and drafts content. AI Shopping is included. API and MCP wait for Custom. There are still no crawler logs and no conversion analytics of the kind Promptwatch publishes. It is a better-documented monitor-plus-agent than Meteoria. It is not the measurement stack under the score.
Promptwatch: real UIs, and the layer under the score
Promptwatch monitors real product UIs on paid plans. The engine list is ChatGPT, Gemini, Claude, Perplexity, Grok, Llama, DeepSeek, Mistral, Copilot, plus Google AI Overviews and AI Mode. You are not asked to pick three. Only the free Explore tier is ChatGPT-only (10 prompts).
Agent Analytics is a crawler-log product with named integrations: Cloudflare, AWS CloudFront, Fastly, Vercel, Netlify, Akamai, Google Cloud CDN, custom HTTP. It shows ChatGPTBot, ClaudeBot, PerplexityBot, GoogleOther and the Meta AI crawler, with a crawl-to-citation path and error tracking. Visitor analytics tie AI-referred sessions to conversions through a script or GTM template. Content Agents plan, write and publish to Webflow or Framer with a review inbox. Citation analytics go down to page, domain, Reddit, YouTube and offsite mentions. REST API v2, an MCP server, a Claude Connector and a ChatGPT plugin are published.
Pricing: Explore free. Essential $95 a month with a 7-day trial (50 prompts, 6,000 responses, 200K visitor events, 5 AEO articles). Professional $245 (150 prompts, 18,000 responses, 1M visitor events, 25M crawler logs, 15 AEO articles). Business $579. Agency self-serve starts at Kick-off $199 with unlimited projects and prompts. Annual is twelve months for the price of ten. Prices are also offered in euros.
Side by side
| Meteoria | BotRank.ai | Promptwatch | |
|---|---|---|---|
| Collection, as published | Terms describe third-party LLM/search/analytics APIs and sampling | Daily scraping of real interfaces | Real UI monitoring on paid plans |
| Engines on a mid paid plan | 3 from a list of 7 | 3 from a list (full set on Custom) | Full published list; only Explore is ChatGPT-only |
| Entry paid price | 75 EUR/mo incl. tax | 89 EUR/mo (75 EUR annually) | $95/mo Essential; Explore is free |
| Crawler logs | "AI log analysis" from Advanced, no CDN list | None published | Agent Analytics, 7 named CDNs plus custom HTTP |
| Conversions | GA and Matomo connectors | Not published | Visitor analytics via script or GTM |
| Publishing | AI-friendly content from Pro, credits | Agent drafts content; no CMS list | Content Agents to Webflow and Framer with review |
| API / MCP | Agency / Enterprise row | Custom only | REST API v2 and MCP on the platform |
Which to pick
Pick Meteoria if you want a French invoice, unlimited logins and a daily mention score, and you accept that the terms describe sampled API data. Confirm the agency price in writing, because 300 EUR and 700 EUR are both on the site.
Pick BotRank.ai if you want that same euro-and-seats shape with a documented real-interface scrape, a page audit and an in-app agent, and three engines are enough until you can pay for Custom.
Pick Promptwatch if the number has to match what people see, and if you also need to know which bot read the page, whether a citation followed, and whether the AI visit converted. That is the job the other two do not publish.
French teams often end up with BotRank.ai for the local dashboard and Promptwatch for the measurement stack. Meteoria is the one we would trial only after reading the terms, not the homepage.
FAQ
Does Meteoria scrape UIs or call APIs?
Abondance repeats a UI-scraping claim. Meteoria's terms of 12 August 2026 describe querying third-party LLM, search or analytics APIs, with results dependent on sampling. We treat the contract as the disclosure.
Why not pick BotRank.ai over Promptwatch?
You can, if three engines, no crawler logs and no conversion tracking are acceptable. BotRank.ai is the better French monitor of the two euro tools. Promptwatch is the platform under the monitor.
Is Promptwatch available in euros?
Yes. The published table is in USD; the fact sheet states prices are also offered in EUR. promptwatch.com.