How to Connect Semrush to Perplexity with CorpusIQ MCP¶

Semrush answers questions about search that no other surface in the business can answer. Which keywords a competitor holds, who links to them, what a term costs to bid on, and which pages are losing ground month by month.

The interface is built around choosing a report first. Most real questions start somewhere else, with a domain and a suspicion.

Connecting Semrush through CorpusIQ MCP puts that data in a Perplexity thread where the question comes first, read-only.

Why Connect Semrush to Perplexity?¶

Competitor keywords. "What organic keywords does competitor.com rank for that we do not?"
Own position. "Show me my domain's organic and paid traffic trend for the last six months."
Backlinks. "How does our referring domain count compare to competitor.com?"
Demand sizing. "What is the search volume and CPC for a term we are considering?"
Competitive set. "Who are our top organic competitors, and how far ahead are they?"
History. "Which of our pages lost the most organic traffic between these two snapshots?"
Read-only by scope. The API key reads report data. It cannot change a project, a campaign, or a listing.

How It Works¶

Connect Semrush to CorpusIQ with an API key from an account whose plan includes API access.
Add CorpusIQ to Perplexity as a custom remote MCP server, using the endpoint from your CorpusIQ account settings.
CorpusIQ queries domain overview, organic and paid keywords, the organic competitor list, backlink overview, keyword research, and domain history as the question requires.
Perplexity answers from those reports and names the domain and period it measured.

Direct MCP retrieval is read-only. It does not retain raw customer files or full connector response payloads; operational query logs are retained for up to 30 days.

Setup Steps¶

In Semrush

Open your profile and go to Subscription info, then API units.
Copy the API key. API access requires the Business plan or higher.

In CorpusIQ

Open your dashboard, find Semrush, and click Connect.
Paste the API key and save.

In Perplexity

Open Settings and find the MCP servers section.
Add a custom MCP server and paste the endpoint:

https://mcp2.corpusiq.io/mcp

Connect and approve the CorpusIQ authorization page once.
Start a new thread and ask a competitor question.

Semrush bills each report in API units. A narrow question, one domain and one metric family, costs less than a broad one and usually answers more precisely.

Example Perplexity Questions¶

Competitor keywords

"What organic keywords does competitor.com rank for that we do not?"
"Which competitor pages rank in the top three for our head terms?"
"Show me the paid keywords competitor.com bids on."

Own organic position

"Show me my domain's SEO overview for the last six months."
"Which of my pages lost the most organic traffic between these snapshots?"
"What is our estimated organic traffic split by country?"

Backlinks

"How does our referring domain count compare to competitor.com?"
"Which sites link to both domains?"
"Has our authority score moved in the last quarter?"

Demand and planning

"What is the search volume, CPC and difficulty for a term we are considering?"
"Which related terms have volume above a threshold and difficulty below one?"
"How seasonal is this term across twelve months?"

Cross-source

"Does the organic traffic in Semrush match the sessions Search Console reports for last quarter?" (requires the Google Search Console connector)
"Which Semrush keyword gaps map to pages our sitemap already covers?" (requires a sitemap read)

Security¶

Read access only. The key reads report data. There is no path from Perplexity that edits a project, a campaign, or a listing.
Revocable at any time. Regenerate the key in Semrush and the connection ends immediately.
Account-scoped. Access follows the key. A Semrush account you have not connected stays out of reach.
Unit budget stays yours. CorpusIQ only spends units on the queries you ask for. There is no background crawl.
Scoped direct-MCP retention. Raw customer files and full connector payloads are not retained. Operational logs keep query text and bounded outcome summaries for up to 30 days.
TLS in transit between Semrush, CorpusIQ, and Perplexity.

Comparison: Semrush Reports vs Semrush in Perplexity¶

Aspect	Semrush reports	CorpusIQ MCP with Perplexity
Input	Choose a report and a domain	A question about a domain
Output	Tables you then interpret	An answer with the domain named
Across domains	One report per domain	Asked as a comparison
Cross-source	Export and reconcile by hand	Asked across connectors
Ad-hoc questions	Build a new report	Just ask
Write access	Full, by you	None, read only

Reports are built for tracking a plan. The connector is for the question that starts with a suspicion.

FAQ: Common Questions¶

Why does Semrush say the API key is invalid?

API access requires the Business plan or higher, and the account must have API units left. Both conditions fail with the same message.

Will asking questions burn my API units?

Yes, each report costs units. Ask narrow questions: one domain, one metric family, one period. That is cheaper and more precise than a broad sweep.

Why is there no data for a domain?

Semrush data lags reality, and very new domains may not be indexed yet. Check the domain in the Semrush interface before assuming a connection problem.

Can it change anything in Semrush?

No. The key reads report data and the connector has no write tool. Projects, campaigns, and listings are untouched.

Does this work only in Perplexity?

No. Semrush is exposed to ChatGPT, Claude, and Perplexity through the same endpoint on the same subscription.

Where is the Semrush data stored?

CorpusIQ does not retain raw customer files or full connector response payloads. Operational logs may persist for up to 30 days. See the [security documentation](/hermes/security/read-only-and-security).

Next steps: Connect Semrush to Perplexity now

Part of the MCP knowledge base at corpusiq.io - connect 40+ business tools to AI.

This Hermes repo is one of the largest structured collections of public AI, automation, business, and technology documentation. Content remains attributed to original authors and repositories. Indexed and organized by www.CorpusIQ.io.