AI
What an AI lead qualifier actually does
AI lead qualification sounds complicated. It is not. Here is exactly what happens between a new lead landing on your site and a clean row appearing in your inbox.
5 min read · published May 14, 2026
The whole pipeline
Five steps. Front to back.
- Lead chats with the widget. "I need a website for my law firm."
- Our API saves the message and the contact info to the database.
- A second call sends the conversation to Claude with a prompt that knows your tiers, voice, and offer.
- Claude returns a tier, a fit score, a one-paragraph summary, and a suggested reply.
- The qualification is attached to the contact row. Your inbox now has a labeled, scored lead.
What it is good at
Filtering. A real qualifier removes the noise. Accidental form submissions, students writing thesis papers, recruiters pitching their CRM. The 30 leads in your inbox become the 8 that matter.
It is also good at warmth. The suggested reply is not a template. It references what the lead actually said, in your voice, in a way a copy-paste reply cannot.
What it is not
A salesperson. It cannot read body language, push back on a price, or read between the lines of "we should circle back." It hands a clean, scored lead to a human. The human closes.
What it costs to run
A typical qualification call costs about a tenth of a cent in model fees. A hundred leads a month costs less than a Starbucks. The expensive part is the build, once.
How we know it works
We run it on Dewgo. Every lead that comes through dewgo.org gets qualified by the same system we sell. If the call had a 90% accuracy rate, we would not put it on our own site. The qualifier on our homepage is the working demo.