The problem
How it works
The official tutorial starts a Make scenario whenever Gmail receives a message with a Leads label. A Make AI Agent (New) module receives the sender, subject and body, and has three tools: Google Calendar free and busy times, Make AI Web Search to research the company behind the email domain, and Gmail to create a draft. Its instructions tell it to score each lead high, medium or low against set criteria, decide whether a meeting is worth proposing, offer two or three free slots if so, and always finish by creating a draft reply. A final Slack module posts a link to the draft in a private channel. You test by replaying earlier runs and checking the Reasoning tab for tool calls and errors.
Results and limitations
This is a tutorial, not a reported result. Make’s product page adds that each agent decision is shown in a Reasoning panel and that you can set rules, add manual approvals or stop the agent at set points.
Before you try it
The agent only creates drafts; keep it that way, so a person reads and sends every reply. The tutorial is built for people who emailed you; do not point it at bought or scraped contact lists, and check anti-spam and data protection rules before any follow-up. Email contents go to the AI provider you connect, so check its data terms. Company research from web search can be wrong, and the sample scoring rules and Berlin working hours are examples to replace with your own. Results depend on your setup and were not independently reproduced by Top 200 Repos.
Tools and repositories
No ranked repository is involved. This use case still stands on its own source.
- Tools
- Make AI Agent (New) · Gmail · Google Calendar · Make AI Web Search · Slack
- Prerequisites
- A Make account with an AI provider connection, a Gmail label for lead emails and a private Slack channel.
Try it yourself: Lead qualification rules for your own business
Fill the fields to adapt the prompt. Empty fields stay as visible [placeholders].
I sell [What you sell] to [Ideal customer]. Write qualification rules for an agent that reads inbound sales emails. Define what makes a lead high, medium or low priority, using only facts from the email and public company information. Say when the agent should suggest a meeting, using my working hours of [Working hours and time zone], and when it should reply with information only. The agent must create a draft for my review and never send email itself. Add rules for spam, students, competitors and anyone who asks not to be contacted, who should get no sales follow-up.Test with sample emails you write yourself before connecting a real inbox.
Sources
- Sales outreach AI agent use case (opens in a new tab)
Project documentation · help.make.com · Undated · Source checked 12 Nov 2026, 08:08 UTC
What this source supports
- Scenario triggers on Gmail messages with a Leads label and passes them to a Make AI Agent (New) module
- Agent tools are Google Calendar free/busy, Make AI Web Search and Gmail draft creation
- Instructions score leads high, medium or low and decide whether to propose a meeting
- The agent proposes two or three free slots and always creates a draft reply
- A Slack module posts a link to the draft in a private channel
- Testing uses earlier runs and the Reasoning tab to inspect tool calls and errors
- Make AI Agents (opens in a new tab)
Vendor post · make.com · Undated · Source checked 12 Nov 2026, 07:54 UTC
What this source supports
- A Reasoning panel shows each agent decision and the tools it used
- Users can set rules, add manual approvals or stop the agent at specific points
- Agents work alongside fixed scenario logic