Company Intelligence & Lead Qualification Agent
Give it a companys domain and it researches the companys site and recent news, scores it against your ideal customer profile, and writes an account brief with sources for every finding.
Agent profile
Watches for
A company domain is submitted
Runs
On demand
Connects to
- Company websites
- Exa web research over MCP
- Research database
- Review interface
Hands off to a person
It recommends; it doesn’t decide. A salesperson reviews the brief, its sources, and the reasoning behind the score before pursuing an account.
The challenge
A sales team needed to decide quickly whether an account was worth pursuing: who the company is, what it sells, which industries and regions it serves, its size, the technology it appears to use, recent news, growth signals, and which roles would likely buy. Gathering that by hand took too long to do for every account.
What we built
Validate
Confirms the domain belongs to a real, active company before any research starts.
Read the site
Extracts the company description, products and services, industries, locations, customer types, leadership and departments, and technologies that can reasonably be identified.
Research beyond it
Uses Exa over MCP to find recent news, acquisitions, new products and locations, hiring, AI and automation initiatives, platform changes, and expansion into new markets.
Structure the findings
Collects everything into a predictable structure, separating facts from inferences and unknowns.
Score the fit
Compares the company with a defined ideal customer profile: size, industry, market, digital maturity, technology, growth activity, and likely need for web, automation, AI, integration, or marketing work. The score explains itself.
Write the brief
A summary, key findings, recent signals, ICP fit, reasons to pursue or not, roles to contact, talking points, and the sources behind each finding.
Under the hood
- MCP
- Exa
A digital asset management workflow that ingests images, manages their metadata, grades each one against a rubric, makes the library searchable, and keeps every version.
A Plaid integration that pulls accounts, balances, and transactions into a documented PostgreSQL schema, keeps it current without duplicates, and answers questions about it with SQL and semantic retrieval.
Give it a company’s domain and it researches the company’s site and recent news, scores it against your ideal customer profile, and writes an account brief with sources for every finding.
