Visitors are getting frustrated on /
17% of clicks on / are rage or dead clicks (32 of 184, 6 people) — a sign of broken, unresponsive or misleading UI.
Track whether ChatGPT, Gemini, Perplexity, Google AI Overviews and Claude recommend you, with a confidence interval on every number, from $79/mo month-to-month.
| BusinessMCP | Profound | |
|---|---|---|
| Entry price | Free (AI recall); grounded engines from $79/mo | $99/mo (1 engine) – $399/mo (3 engines), billed yearly |
| Engines measured (web-grounded) | Up to 5: AI Overviews, Gemini, Perplexity, ChatGPT, Claude | 1 (Starter), 3 (Growth), up to 9 (Enterprise) |
| Web-grounded probes (not model recall) | Yes, from Growth ($79); Free/Starter = AI recall only | Yes, all tiers |
| Confidence intervals on every rate | 95% Wilson interval, always shown | No |
| Share of voice vs tracked competitors | Yes | Yes |
| Cited third-party domains list | Yes | Yes |
| AI-crawler hit tracking (GPTBot, ClaudeBot...) | Yes | Yes (crawler analytics on higher tiers) |
| Bundled with analytics + CRM + AI business analyst | Yes | No |
Profound is the reference product in AI visibility: it tracks brand mentions and citations across ChatGPT, Perplexity, Google AI Overviews and, on enterprise, up to nine answer engines, with daily runs, agent credits and a shopping-and-conversation research layer. Its public pricing page lists two self-serve tiers, both billed yearly: Starter at $99 per month (ChatGPT only, 50 prompts, one seat) and Growth at $399 per month (three engines, 100 prompts, three seats). Enterprise is custom, and published teardowns put those deployments at $2,000–$5,000+ per month.
BusinessMCP measures the same thing — is your brand recommended or cited when someone asks an AI assistant the questions your buyers ask — and prints a 95% confidence interval beside every rate so a small week is never mistaken for a trend. Grounded engines start on the $79 Growth plan (Google AI Overviews + Gemini), Scale adds Perplexity + ChatGPT, Business adds Claude, and every plan is month-to-month. The trade you make is breadth and cadence: five engines at most, measured weekly, alongside the analytics, CRM and AI business analyst that come with the platform.
Profound’s Starter is $99 per month billed yearly and covers ChatGPT alone; three engines cost $399 per month, also yearly. That is $4,788 up front for the plan most reviews recommend. BusinessMCP’s grounded engines start at $79 per month with no contract, and the free plan already measures AI recall.
Visibility is a binomial proportion measured from a handful of answers. BusinessMCP computes a 95% Wilson interval for every rate and shows it ("42% ±11"), pools by trials rather than averaging percentages, and only calls a change conclusive when the intervals say so. Profound reports point estimates.
Profound is a standalone visibility suite. BusinessMCP puts citation rate, AI-crawler hits (GPTBot, ClaudeBot, PerplexityBot) and the human referrals those crawls eventually send — the AI channel in your first-party analytics — on one dashboard, and the AI business analyst can reason across all three.
Both tools show which third-party domains the engines cite. BusinessMCP goes one step further by mining the engines’ own answers for the competitor names they recommend instead of you, so the entities you track are the ones the models actually name.
Real data, not a mockup
Recorded on our own site, PII scrubbed — the same product you get on the free plan.
Today's insights
4 from the daily analysis · Sep 2217% of clicks on / are rage or dead clicks (32 of 184, 6 people) — a sign of broken, unresponsive or misleading UI.
You rank on page 1–2 (pos 5–20) for 18 queries with real impressions — e.g. "skills directory" at position 11. Small on-page moves could push these into the top 3.
Across your tracked prompts, AI answers name HubSpot in 48% of grounded answers vs your 22% — and 1 other tracked brand also out-rank you. The sources those answers cite are the fastest lever.
visitors moved from 118 to 186 week-over-week.
Runs every morning, on every plan. Real cards from our own workspace.
| Plan | Price | Notes |
|---|---|---|
| Starter | $99/mo | billed yearly; ChatGPT only, 50 prompts, 1 seat, 100 agent credits |
| Growth | $399/mo | billed yearly; ChatGPT + Perplexity + Google AI Overviews, 100 prompts, 3 seats, daily runs |
| Enterprise | Custom | up to 9 platforms, SSO/SAML, SOC 2, dedicated Slack; teardowns report $2,000–$5,000+/mo |
| BusinessMCP Growth | $79/mo | month-to-month; AI Overviews + Gemini grounded, 15 questions, weekly, with analytics + CRM |
| BusinessMCP Scale / Business | $199 / $499/mo | adds Perplexity + ChatGPT (25 questions), then Claude (40 questions) |
Profound self-serve prices are shown "billed yearly, 2 months free" on its pricing page; the enterprise range is from published 2026 teardowns, not from Profound. BusinessMCP Free and Starter include no web-grounded engine and report AI recall, not a citation rate.
An alternatives page you can trust has to cut both ways. Honestly:
For measuring whether the major assistants recommend and cite you, and for finding the domains they cite instead, yes — with intervals Profound does not show. Profound goes wider (up to nine platforms on enterprise), runs daily, and adds agent credits and conversation research. If you need those, stay.
Profound’s three-engine Growth plan is $399 per month billed yearly ($4,788/yr). BusinessMCP’s Scale plan, which grounds four engines (AI Overviews, Gemini, Perplexity, ChatGPT), is $199 per month with no contract, and includes analytics, a CRM and the AI business analyst. Growth at $79 grounds the first two.
Visibility rates move slowly and each probe costs real money, so we spend the budget on more samples per engine per week rather than one sample per day. Every question is measured once per ISO week per engine; adding a question probes it immediately rather than waiting for the next week.
Those plans include no web-grounded engine, so we ask a model without web search whether it knows your brand. That measures training-data recall, which a brand younger than the model’s cut-off can never score on, so we never present it as a citation rate. Grounded measurement starts on Growth or with your own model key.
No. Every self-serve BusinessMCP plan is month-to-month; annual billing is optional and gives two months free. Profound’s published prices are billed yearly.
Export your tracked prompts from Profound and paste the unbranded ones into BusinessMCP (Growth allows 15, Scale 25, Business 40 — pick the questions a buyer would actually ask, not variations of one).
Add the competitors Profound shows you losing to; BusinessMCP will also propose the brands the engines name in their own answers.
Run both for two weekly cycles and compare the cited-domain lists. Expect BusinessMCP’s rates to carry wider intervals on fewer samples — that is the honest picture, not a bug.
Connect your own OpenAI, Gemini or Anthropic key if you want those engines grounded on a plan below the one that includes them; BYOK engines are never plan-gated.
Every price on this page traces to a public source, last verified September 15, 2026. Spotted a change? Tell us and we'll re-verify.
Companies running MCP servers with us
Makers and companies publishing a server in the BusinessMCP directory.
Grabbit MCP
grabbit.sh
Find Reddit buyer-intent threads, inspect subreddit rules, read full…
FinancialData.Net MCP Server
financialdata.net
Get real-time stock prices, fundamentals, institutional trading insights, and…
API2Cart MCP
api2cart.com
Unified MCP server for 70+ eCommerce platforms.
Polyrama
github.com/Polyrama
Prediction-market terminal and MCP for live odds, market search, trader analytics…
Twitter/X API by SocialData
socialdata.tools
Read-only Twitter/X data for AI agents: search tweets, look up profiles…
MachineTranslation.com MCP
machinetranslation.com
Agents can translate text through 22 leading AI models at once and get back only…
PolymarketScan
polymarketscan.org
Free no-auth Agent API plus an independent record of Polymarket wallets and prints.
Humanizer PRO MCP
texthumanizer.pro
Hosted OAuth MCP for rewriting authorized draft text, scanning AI-style signals…
Process Street MCP Server
process.st
Connect AI agents to Process Street workflows, tasks, runs, data sets, and…
Magic Hour MCP
github.com/magichourhq
Generate and edit video, images, and audio through 44 Magic Hour tools from…
Trendos
trendos.com
AI search visibility analytics and opportunities.
ContextStream
contextstream.io
Shared project memory for AI coding agents across Cursor, Claude Code, Codex…
Asyntai – AI Support Agent for Websites
github.com/asyntai
Agents run a website's Asyntai AI support agent: manage its knowledge base and…
iLook
ilook.fit
Privacy-first, non-medical AI face-analysis tools for consented photos, with clear…
HelpMyAgent
helpmyagent.com
Business data, intelligence and procurement APIs designed for AI agents.
FoxForm
foxform.app
Agents can build forms, quizzes and calculators where each answer scores points…
Google Ads MCP Server
get-ryze.ai
Solana Sniper Bot MCP
github.com/Solara-Snipe-Bot
Autonomous Solana meme-coin sniper with 48 AI MCP tools and multi-source launch…
padel.how Racket Reviews
padel.how
CorpusIQ
corpusiq.io
Business operations platform exposing 65+ MCP tools for data, commerce, marketing…
CareClinic Health Tracker
careclinic.io
Track symptoms, medications, and mood from your CareClinic account, then review…
Vibe Prospecting
vibeprospecting.explorium.ai
B2B company and contact intelligence for prospecting workflows.
Install one cookieless script next to Profound and compare on your own numbers. Free plan, no credit card, cancel anytime — the opposite of an annual contract.