Until recently, enrichment workflows meant switching tabs, managing API keys, running scripts, or building one-off workflows just to append basic firmographics.
What’s different now is that enrichment is becoming conversational, real-time, and embedded directly inside the environments where you’re already thinking and working.
Instead of exporting lists, switching platforms, running batch jobs, and waiting for files to come back, we’re going to show you how you can enrich company data directly inside Anthropic’s Claude using ZoomInfo’s MCP. We’ll take a simple CSV file with nothing more than company names and websites and, in real time, transform it into a fully enriched dataset with employee headcount, revenue, headquarters details, and funding information.
What makes this powerful is the architecture. Claude orchestrates the workflow, ZoomInfo’s MCP securely retrieves authoritative company data, and the result is compiled back into a clean, downloadable file — all within the same interface.
Let’s get into how LLMs and trusted data infrastructure can come together to eliminate friction and how ZoomInfo’s data layer can now follow you anywhere across your AI and application stack.
Starting with a Basic Company File
Inside Claude, we begin by uploading a simple CSV file. That file contains only two columns: company name and company website. There’s no employee data, revenue, headquarter details, or funding information.
This is the reality for most go-to-market teams. You have a starting point, but it’s incomplete. To prioritize, route, segment, or analyze those accounts, you need richer context.
Instead of exporting the file and running enrichment somewhere else, we stay directly inside Claude.
Requesting Enrichment from ZoomInfo
Once the file is uploaded, we prompt Claude to enrich the companies using ZoomInfo. Specifically, we’re going to ask for employee headcount, revenue, headquarters city and state, last funding date, and total funding raised.
Claude extracts the company names and websites from the CSV. Then, behind the scenes, it initiates a request to ZoomInfo using our MCP server. That request is sent directly into ZoomInfo’s systems, the companies are matched, and the enrichment process begins.
What’s happening in real time is important. Claude is not guessing or hallucinating data. It is calling ZoomInfo through the MCP connection, retrieving authoritative company records, and bringing that information back into the workflow.
As the process runs, Claude compiles the enriched results into a clean CSV and Excel file. The output includes the original company name and website, along with the newly appended fields for headcount, revenue, headquarters location, last funding date, and total funding.
In less than a minute, the dataset is fully enriched and ready to download.
The Output: A Fully Enriched File
On the right side of the Claude interface, you can see the updated table. Employee headcount is now populated. Revenue is filled in. Headquarters city and state are included. Funding data is added directly into the columns.
From there, you can download the file as an Excel document and immediately use it for segmentation, prioritization, routing, or analysis.
There’s no batch processing delay, no context switching, and no need for a separate enrichment platform. It all happens inside the same environment where you’re already thinking and working.
It’s Not Just Companies
The exact same process works for people enrichment.
If you have a list with just a name and company name, or even just an email address, you can ask Claude to enrich it using ZoomInfo’s MCP connection. The workflow is identical. Upload the file, specify the fields you want, and Claude sends the request through ZoomInfo to retrieve the enriched data.
This transforms Claude from a generative tool into a connected data engine.
The Bigger Opportunity: ZoomInfo Anywhere
The most powerful part of this workflow isn’t just that it works inside Claude. It’s that the MCP servers and APIs we’ve built are portable.
You can bring this workflow into ChatGPT. You can bring it into Gemini. You can integrate it into your own internally built, vibecoded applications. You can embed it into your CRM. You can connect it to your marketing automation platform. ZoomInfo’s data, insights, and enrichment capabilities can follow you anywhere across your application hierarchy.
The LLM provides the interface and orchestration, and ZoomInfo provides the trusted data layer. Together, they eliminate the friction between idea and execution.
Enrichment in Under a Minute
In this demo, the full company enrichment process takes around 45 seconds. A raw CSV goes in, and a fully enriched Excel file comes out.
By coupling the power of a large language model with ZoomInfo’s MCP infrastructure, you now have rich, authoritative company data available anywhere you work. Enrichment is no longer a separate workflow. It’s embedded directly into your AI environment.
