Ooty connects your marketing platforms to your AI assistant. Ask ChatGPT, Claude, or Gemini a question and get an answer pulled from Google Analytics, Search Console, Google Ads, Meta, YouTube, Amazon, HubSpot, and more. No exports, no dashboards, no tab switching.
Three steps. Five minutes. No technical background required.
If you're new to connecting AI assistants to external tools, our guide to MCP for marketers covers the fundamentals. Otherwise, let's get you set up.
Setup in 3 Steps
From zero to AI-powered marketing analytics in under 10 minutes
<10mintotal setup time
1
Sign up & choose products
~2 min
Create your account at ooty.io and pick the MCP products you need. Each connects to specific marketing platforms.
Pick from 7 MCP productsEach product = specific APIs14-day money-back guarantee on every plan
2
Connect your accounts
~3 min
Go through the familiar OAuth consent screens for each platform. Google, Meta, YouTube, Amazon -- standard flows you already know.
Standard OAuth consent screensMinimum scopes onlyCredentials encrypted server-side
3
Paste one URL into your AI assistant
~1 min
Copy your MCP endpoint URL and license key into ChatGPT, Claude, or Gemini settings. One URL, one key. That's the entire client-side setup.
One URL per productOne license key for allWorks on any machine
Step 1: Choose your products
Head to ooty.io/pricing and create an account. You'll see seven products, each connecting your AI assistant to a different set of marketing platforms.
Ooty SEO ($49/mo) connects to Google Search Console, PageSpeed Insights, Knowledge Graph, and the Indexing API. Keyword tracking, page-level performance, technical audits, site speed monitoring, and indexing management. If you've ever spent 20 minutes pulling a Search Console export into a spreadsheet to find your top-declining pages, this replaces that workflow.
Ooty Ads ($59/mo) connects to Google Ads and Meta Ads. Campaign performance, ROAS, cost-per-acquisition, budget analysis, and cross-platform spend comparisons. Pull data from both platforms in a single conversation.
Ooty Social ($39/mo) connects to Instagram, LinkedIn, X, and Reddit. Engagement metrics, content performance, cross-platform comparisons, and audience growth trends.
Ooty Video ($19/mo) connects to YouTube Analytics and the YouTube Data API. Watch time, audience retention, subscriber growth, and video-level performance. Ooty's most affordable product, built for creators and teams who publish video regularly.
You keep seeing "MCP" in AI threads and newsletters. Everyone talks about it like you already know what it means. Here is the short version, then we will set it up.
MCP in 30 seconds
MCP (Model Context Protocol) is an open standard that lets AI assistants conn
Charts have always lived outside AI conversations. You run an analysis, get a table of numbers, and then open a separate tool to visualize it. Ploti changes that. It is a free, open-source MCP server that renders 43 chart types as interactive widgets directly
Most marketing teams use AI the same way: copy data from one dashboard, paste it into ChatGPT, Gemini, or Claude, ask a question, and hope the answer is useful. The AI does its best with a screenshot or a CSV export, but it is working with a snapshot, not your
Ooty Analytics ($39/mo) connects to Google Analytics 4, Search Console, and Chrome UX Report (CrUX). Traffic sources, conversion paths, user behaviour, and Core Web Vitals. If you already have Ooty SEO and want deeper traffic analysis, Analytics is the natural complement.
Ooty Commerce ($29/mo) connects to Amazon PA-API, Keepa, and Rainforest. Product research, price tracking, BSR trends, competitive analysis, and review monitoring. Built for Amazon sellers and e-commerce teams.
Ooty CRM ($39/mo) connects to HubSpot and Pipedrive. Deal pipeline analysis, contact management, activity tracking, and sales forecasting.
You don't need all seven. Most people start with one or two and add more later. If you want everything, the All-Access bundle at $99/mo covers every product and saves significantly compared to buying individually.
Step 2: Connect your accounts
For each product you chose, you'll go through a standard OAuth consent screen. Same "Sign in with Google" or "Connect your Facebook account" flow you've done before. About 60 seconds per product.
Google services (Analytics, Search Console, Ads, YouTube, PageSpeed): Google's standard consent screen asking you to grant read access. If you manage multiple Google accounts, make sure you select the one with access to the properties you want to analyse.
Meta (Instagram, Facebook Ads): Meta's Business Login flow. Your Instagram account needs to be a Professional or Business account, not personal.
LinkedIn, X, Reddit: Each platform has its own OAuth flow. You'll be redirected to the platform, approve access, and come back to Ooty.
HubSpot, Pipedrive: Standard CRM OAuth flow. You'll need admin or appropriate permissions on your CRM account.
Amazon (Commerce): Amazon's Product Advertising API. This requires an Amazon Associates account.
Five things worth knowing about security:
We only request the permissions each product needs. Ooty Analytics asks for Analytics and Search Console read access. Ooty Ads asks for Ads read access. We never request Gmail, Calendar, or anything outside what the product uses.
All access is read-only. Ooty pulls data from your platforms. It doesn't create campaigns, publish posts, delete anything, or modify your accounts.
Your credentials are encrypted on our servers. OAuth tokens are encrypted with AES-256-GCM and stored in our database. They never sit on your machine. For full security details, see our architecture post.
You can disconnect at any time. Revoking access removes the encrypted tokens from our servers. You can also revoke directly from the platform side.
No data is stored permanently. When you ask a question, Ooty fetches fresh data from the platform APIs in real time. We don't build a copy of your analytics database.
Step 3: Add one URL to your AI assistant
Once your accounts are connected, you get an MCP endpoint URL and a license key. Here's the setup for ChatGPT, Claude, or Gemini.
One URL per product. One key for all of them. No local process to run, no packages to install, no environment variables. The same key works on any machine.
For a deeper look at the architecture under the hood, see our technical walkthrough.
What to ask first
Restart your AI assistant after adding the config. MCP servers load at startup, so the restart is necessary for tools to appear.
What You Can Ask
Real questions you can ask Claude once connected -- no dashboards, no learning curve
SEO
SEO & Search
"Show me pages losing traffic this month"
"Check Core Web Vitals for my top 10 pages"
"What keywords am I ranking for that I'm not targeting?"
Ads
Paid Ads
"Compare ROAS across my active campaigns"
"Which ad groups are spending budget without converting?"
"Show me cost-per-lead trends for the last quarter"
Social
Social Media
"What content performed best on Instagram this week?"
"Compare engagement rates across my social channels"
"Show me posting frequency vs engagement correlation"
Video
Video Analytics
"Which videos have the highest watch-through rate?"
"Show me audience retention drop-off points"
"Compare this month's views to last month"
Analytics
Web Analytics
"What were my top traffic sources last month?"
"Show me conversion paths for organic visitors"
"Which landing pages have the highest bounce rate?"
Commerce
E-Commerce
"Track price history for competitor products"
"Show me BSR trends for my category"
"What products are trending in this niche?"
Here are starter queries for each product to get you going.
Ooty SEO: "Show me my top 10 pages by organic traffic in the last 30 days, and flag any that dropped more than 20%." Follow up with "why did page X drop?" to dig into impressions, CTR, and position changes.
Ooty Ads: "What's my ROAS across all active campaigns this month? Which campaigns are underperforming?" Follow up with "break down the underperforming campaign by ad set."
Ooty Social: "What content performed best on Instagram this week? How does engagement compare to my average?" Follow up with "what do the top posts have in common?"
Ooty Video: "Which of my recent YouTube videos have the highest audience retention rate?" Follow up with "what's the average view duration for my top 5 versus bottom 5?"
Ooty Analytics: "What were my top traffic sources last month? Which ones are growing and which are declining?" Follow up with "for declining sources, show me the landing pages they're hitting."
Ooty Commerce: "Show me price history and BSR trends for the top 5 products in my category." Follow up with "which has the most volatile pricing?"
Ooty CRM: "Show me all deals that have been in the same stage for more than 14 days." Follow up with "for each stalled deal, show me the last activity date and assigned owner."
Before & After
The same cross-platform analysis -- two very different workflows
Traditional Workflow
1Log into Google Analytics
2Export data to spreadsheet
3Log into Search Console
4Export more data
5Log into Google Ads
6Export campaign data
7Cross-reference in spreadsheet
8Write up findings
~45 minutes of manual work per analysis
With Ooty + Claude
1Ask Claude your question
2Get an answer with data from all sources
~30 seconds -- same data, same analysis, one conversation
90x
faster than the manual workflow
Getting more out of it
Ask follow-ups. The real value is in conversations, not single queries. Context carries across the conversation, so each follow-up builds on the previous answer.
Combine products. With SEO and Analytics both connected, your AI can cross-reference Search Console data with GA4 data in one conversation: "Show me pages with high impressions but low clicks, and check if any also have poor engagement metrics." Our MCP marketing tutorial walks through multi-product workflows.
Set up a weekly check-in. Start a conversation each Monday with: "Give me a summary of anything unusual in my SEO and ads data this week." Save a prompt template so you can run it consistently.
Be specific about time ranges. "Last 30 days" works better than "recently." Your AI uses whatever time range you specify in the API calls, so precision helps.
Request specific formats. Want a table? Ask for a table. Want bullet points? Say so. The AI is good at formatting, but explicit instructions get better results.
Troubleshooting
Tools not showing up
Restart your AI assistant after adding the MCP config. MCP servers load at startup. In ChatGPT and Claude, closing and reopening the app is sufficient. If tools still don't appear, check your JSON config for syntax errors (trailing commas, missing brackets).
Authentication errors
Double-check your license key. It should be pasted exactly as shown in your Ooty dashboard, with no leading or trailing spaces.
Common mistakes:
Missing the "Bearer " prefix (note the space after "Bearer").
Invisible characters copied from an email client.
Using an expired key from a cancelled subscription.
No data returned
Verify your accounts are connected at ooty.io/pricing. If you connected Google Analytics but haven't granted access to the correct property, API calls return empty results.
For Google services, make sure you selected the correct account during OAuth.
For Meta, make sure your Instagram is a Professional or Business account.
For Analytics, confirm you granted access to the specific GA4 property you want to query.
Slow responses
First-call latency is slightly higher due to session creation. Subsequent calls should be 400-800ms. If responses are consistently slow:
Check your network connection. Ooty's servers are in Europe (London region).
Try a simpler query first. Large date ranges or many metrics take longer.
If a specific tool consistently times out, email us with the tool name.
OAuth token expired
Tokens refresh automatically, but occasionally a refresh fails if the upstream platform revokes the token. Go to your Ooty dashboard, disconnect the affected platform, and reconnect it. Takes about 30 seconds.
Something else?
Write to hello@ooty.io. We reply to everything, usually within a few hours on weekdays.
Frequently asked questions
Do I need any technical knowledge?
No. If you can copy-paste a URL and a key into a settings screen, you can set up Ooty. Nothing to install, no code to write, no command line.
Which AI assistant should I use?
Ooty works with ChatGPT, Claude, and Gemini. Use whichever you already prefer. The data and tools are identical regardless of assistant.
Can I use Ooty with multiple assistants at the same time?
Yes. Your license key works across all three. Set it up on ChatGPT for quick questions and Claude for deeper analysis if you like.
Is my data shared with the AI provider?
Your data passes through the AI assistant's conversation, so it's subject to that provider's data handling policies. Ooty itself doesn't store your query results or conversation history.
What happens if I cancel?
Your license key stops working at the end of your billing period. All OAuth tokens are deleted from our servers. No data is retained.
Can I try it risk-free?
We offer a 14-day money-back guarantee on all plans. Try it, and if it's not useful, email us for a full refund.
Can my whole team use one account?
Each license key is for individual use. Team members should each have their own Ooty account with their own connected platforms.
How often is the data updated?
Ooty fetches data in real time from platform APIs when you ask. There's no caching or delay. You see the same data you'd see in the platform's own dashboard.
What to try next
Build a morning dashboard prompt. Write a prompt that pulls key metrics across all connected products: "Pull my top SEO pages, ad spend, and social engagement for the last 7 days. Flag anything that changed more than 15% from the week before."
Use Ooty for reporting. Instead of building slides manually, ask your AI to summarise performance in a format you can paste into a report.
Explore cross-platform analysis. The queries that save the most time span multiple tools. "Show me my top organic landing pages, check which ones have the highest bounce rate in GA4, and cross-reference with their PageSpeed scores."
Read the architecture post. Curious about how Ooty handles authentication, data flow, and security? Our architecture deep-dive covers everything.
