Your sales team shouldn't be spending half their day deciding whether a lead is worth pursuing. AI can do that qualification in under 3 seconds — and route warm leads directly to the right rep while cold leads go into a nurture sequence. Here's exactly how to build it.
What This System Does
The workflow we're building connects four things: your lead capture form (or CRM intake), Make.com as the orchestrator, Claude AI for scoring and summarization, and your CRM for storage and routing. When a new lead comes in, the system automatically:
- Enriches the lead with company size, industry, and tech stack data
- Sends the lead profile to Claude AI for a quality score (1–10) and a 2-sentence qualification summary
- Routes score 7–10 leads directly to your sales team with a Slack alert
- Drops score 1–6 leads into an automated nurture email sequence
- Logs everything in your CRM with the AI score, summary, and routing decision
What You'll Need
- Make.com account (free tier works for testing; Pro recommended for production)
- Anthropic API key (for Claude AI access)
- A CRM — HubSpot, Salesforce, or Airtable all work
- A lead capture form (Typeform, Webflow, or your own)
- Slack workspace (for hot lead alerts)
Step 1: Set Up the Make.com Scenario Trigger
Create a new scenario in Make.com and add a Webhook trigger. Copy the webhook URL — this is where your form will send new leads. If you're pulling from an existing CRM, use the CRM's 'Watch New Records' trigger instead.
Test the trigger by submitting a sample lead. You should see the data appear in Make.com's data inspector on the right side. Pay attention to which fields map to company name, job title, email, company size, and the prospect's message — you'll need all of these for the AI prompt.
Step 2: Enrich the Lead (Optional but High-Value)
Add a Clearbit or Apollo.io module after the trigger to enrich the lead with company data — employee count, industry, annual revenue, and technologies used. This data dramatically improves the AI's scoring accuracy because it can factor in company fit, not just the lead's self-reported information.
Map the enriched fields into Make.com variables for use in the next step. If enrichment fails (the company isn't in the database), set default values so the scenario doesn't break.
Step 3: Score the Lead with Claude AI
Add an HTTP module configured for the Anthropic API. Set the method to POST and the URL to `https://api.anthropic.com/v1/messages`. Add your API key in the headers as `x-api-key`.
The prompt is the most important part. Here's the exact prompt template we use for B2B SaaS companies — adapt the ICP criteria to match your business:
You are a lead qualification expert for AutomateUs, a B2B automation agency. Score this lead from 1–10 based on fit with our ICP (Ideal Customer Profile). Our ICP: 5–200 employee companies, operations/marketing roles, US/UK/CA/AU, experiencing manual workflow pain. Lead data: - Name: {{lead.name}} - Company: {{lead.company}} - Title: {{lead.title}} - Employees: {{enriched.employees}} - Industry: {{enriched.industry}} - Message: {{lead.message}} Return JSON: { "score": <1-10>, "summary": "<2 sentence qualification summary>", "reason": "<main scoring factor>" }
Parse the JSON response using Make.com's JSON parser module. Map `score`, `summary`, and `reason` to variables for the next steps.
Step 4: Route Based on Score
Add a Router module with two paths. Path 1 filters on `score >= 7` (hot leads). Path 2 is the default (everyone else).
In Path 1 (hot leads): Add a Slack module to post a message to your `#hot-leads` channel with the lead's name, company, score, and AI summary. Then add a CRM module to create the contact with 'Hot Lead' status and assign to the appropriate sales rep based on territory or round-robin logic.
In Path 2 (nurture): Add the lead to your email nurture sequence in HubSpot or ActiveCampaign with 'Nurture' status. The AI summary goes into the contact notes field so reps have context if the lead ever re-engages.
Step 5: Log Everything in Your CRM
Regardless of which path a lead takes, create or update their CRM record with: AI quality score, AI qualification summary, enrichment data, routing decision, and timestamp. This creates an audit trail and gives your team full context on every lead — whether they're in the hot list or the nurture sequence.
Results You Can Expect
One of our clients — a B2B SaaS company with 40 inbound leads per week — implemented this system and saw their sales team's qualified-lead-to-meeting rate jump from 22% to 61% within 90 days. The reason: reps were spending zero time on unqualified leads and 100% of their time on prospects that actually matched the ICP.
The system processes each lead in under 5 seconds and scales infinitely. Whether you receive 10 leads per week or 1,000, the infrastructure handles it without adding headcount.