To do competitor research with ChatGPT, give it live data and ask it to find the pattern. On its own, ChatGPT only knows what was in its training data and what it can find with web search, so it can't tell you a rival's traffic, the keywords they rank for, or the ads they ran this week. Connect a data source such as Manifold MCP, Similarweb or Semrush, and ChatGPT can pull those numbers and turn them into a competitor brief in one chat.
This guide covers four questions most teams want answered about a competitor: how big they are, what they rank for, what they advertise, and what people say about them.
Can ChatGPT do competitor analysis?
ChatGPT can do the analysis part well: comparing positioning, summarizing reviews, spotting messaging patterns, and turning raw data into a SWOT or a battlecard. What it lacks by default is the data. Ask it "how much traffic does competitor.com get?" without a connected tool and you'll get a guess or a refusal.
That's why the useful setup is ChatGPT plus a live data connection through MCP (Model Context Protocol), the standard that lets an AI assistant call outside tools. Once connected, you ask in plain English and ChatGPT picks the right lookup.
Which tools give ChatGPT competitor data?
- Manifold MCP is a hosted MCP server for ChatGPT, Claude and Cursor. It is hosted, needs no API keys of your own, and you pay per call. For competitor research it covers domain overviews and traffic estimates, the keywords a site ranks for, keyword gaps, search competitors, backlinks, the public ad libraries of Meta, TikTok, LinkedIn and Google, and Reddit, LinkedIn, X, YouTube, TikTok, Instagram and Facebook posts.
- Similarweb offers an MCP connector for Claude and other assistants covering traffic and engagement, traffic sources, audience, keyword and app data. Similarweb says it needs a Similarweb account with API access. Strong if you need deep web and app analytics and already have a contract.
- Semrush has Traffic Analytics for estimating a site's visits and top pages, and Advertising Research for the keywords competitors bid on, their ad copies and ad history. It also offers a Semrush MCP for ChatGPT and Claude, included in its Semrush One and SEO Classic plans. Strong if your team already uses Semrush daily.
Similarweb and Semrush are full platforms with their own apps and reporting. Manifold MCP is built for asking questions inside the AI app you already use and paying per lookup, rather than for dashboards.
How do I research a competitor's website traffic?
- Connect a data source to ChatGPT (for Manifold MCP: add
https://mcp.manifoldmcp.com/mcpas an MCP app in developer mode and sign in). - Ask for an overview of each competitor: estimated organic traffic, ranked keyword count, top pages and authority.
- If you're sizing a whole market, compare many domains in one call.
Use Manifold MCP to compare these competitors: competitor-a.com,
competitor-b.com, competitor-c.com and our site example.com.
For each, show estimated monthly organic and paid traffic, number of
ranking keywords, and domain rank. Then rank them and tell me who is
growing fastest based on their top pages.
Manifold MCP's bulk traffic tool estimates organic and paid traffic for up to 1,000 domains per call, which is useful for sizing a category.
Traffic numbers from any third-party tool are estimates. Use them to compare sites with each other, not as exact visit counts.
How do I find the keywords my competitors rank for?
Two lookups do most of the work:
- Ranked keywords shows everything a domain or a single page ranks for in the top 100, with position, URL, volume and estimated traffic.
- Keyword gap shows keywords a competitor ranks for that your site doesn't rank for at all.
Run a keyword gap between example.com and competitor-a.com in the US.
Show the 100 biggest gaps by volume with their ranking URL and keyword
difficulty. Then group them into content themes and suggest the 5 pages
we should build first.
Not sure who your search competitors even are? Ask for the domains that share the most keywords with you. They're often different from the companies your sales team names.
Who are example.com's search competitors? List the 20 domains that share
the most keywords with us, with the shared keyword count and their
average position.
How can I see my competitors' ads?
There are free official places to look, and faster ways to analyze what you find.
Free, manual options:
- The Meta Ad Library lets you search all ads currently running across Meta technologies by keyword or advertiser.
- Google's Ads Transparency Center is, in Google's words, a searchable hub of ads served from verified advertisers across Search, YouTube and Display, showing which ads ran in a region and when an ad last ran.
These are great for a quick look. They get slow when you want to compare several competitors, read every ad's copy and landing page, and summarize the angles.
In ChatGPT: Manifold MCP's ad library tools read the public ad libraries of Meta, TikTok, LinkedIn and Google. Each ad comes back with its headline, body text, first and last date shown, whether it's active, its call to action and the destination URL, plus impressions and spend where the library publishes those ranges.
Show me every active ad Attio is running on Meta. For each one, give the
headline, the main promise, the call to action, the landing page and how
long it has been running. Then summarize their top 3 messaging angles and
which ads have run longest.
Ads that have run for a long time are usually the ones that work, so sort by first-shown date and study the oldest active ads closely.
For paid search: Semrush's Advertising Research covers the Google keywords competitors bid on and their ad history, which ad libraries don't show.
How do I find out what customers say about competitors?
Reddit is where people ask for recommendations and complain honestly. Ask ChatGPT to search posts and comments that mention a competitor, then summarize the praise and the pain points.
Search Reddit for posts and comments from the past year mentioning
competitor-a. Summarize what people like, what they complain about, and
which alternatives they recommend. Quote the 5 most useful comments with
links.
Reddit search is ranked and never complete, so use it to find examples and themes, not to count mentions.
What does a full competitor brief look like?
Once the data is in the chat, ask for one document that pulls it together:
- Size and growth: traffic estimates and top pages.
- Search strategy: top keywords, biggest gaps against you.
- Paid strategy: live ads, angles, offers and landing pages.
- Voice of customer: Reddit praise and complaints.
- Recommendations: three things to copy, three to avoid, three gaps to own.
On cost: Manifold MCP starts with 500 free credits and no card, then top-ups from $10 or Pro from $20 a month. Most ad library calls cost 1 credit, while bulk traffic estimates cost more, so check prices before large runs.
Questions people ask
Can ChatGPT do market research?
Yes, for synthesis. ChatGPT can summarize a market, compare positioning and draft personas. For anything that needs current numbers, like traffic, keyword volumes or live ads, connect a data source so it isn't guessing.
How do I see what ads my competitors are running?
Search the Meta Ad Library and Google's Ads Transparency Center for free. To analyze many ads at once in ChatGPT or Claude, connect a tool that reads those libraries, such as Manifold MCP, and ask for a summary of angles, offers and landing pages.
Is Similarweb or Semrush better for competitor research?
They overlap. Similarweb is known for web and app traffic and audience data; Semrush is known for SEO and paid search research. Both offer MCP connections for AI assistants tied to their paid plans. If you want pay-per-call access without a subscription, Manifold MCP is an alternative.
How accurate are competitor traffic estimates?
All third-party traffic numbers are modeled estimates, not analytics data. They're reliable for comparing sites and spotting trends, less so for exact visit counts.
Can I track competitors automatically with ChatGPT?
ChatGPT runs lookups when you ask, so it's best for on-demand research. For scheduled monitoring, use a platform with alerts built in. Semrush's Traffic Analytics, for example, includes traffic alerts and weekly competitor monitoring reports.