Connections

AI agent with Make automations

An AI agent collects a business enquiry and passes the details to your Make scenario through an approved connection for the next step.

The customer context

We need a quote for three branches.

One request may need a record and a team notification. Define each action and its failure handling before treating the process as complete.

What remains after the conversation

Through Make, the AI agent saved enquiry M-520 and created a sales task with the branch quantities, deadline and contact.

Telp assistant Website
  1. Customer

    We need a quote for three branches.

  2. Telp

    How many printers per branch, what deadline, and where should we email the quote?

  3. Customer

    One per branch, three in total, by next month. [email protected]. Please send the enquiry.

  4. Telp

    Enquiry M-520 is saved in the sales register and a quote task is assigned to sales. It includes three printers, the deadline and your email.

From a question to the next step

  1. 01

    Define the destination and required information.

  2. 02

    Configure the Make scenario and permitted actions.

  3. 03

    Test success, missing fields and failed delivery.

FAQ

Frequently asked questions

How does Make connect a conversation to several business apps?

A configured Make scenario receives selected conversation data and performs agreed steps, such as recording a contact and notifying the team. Each step must use a supported connection and permitted data.

Is connecting a Make account enough to get started?

No. You need to define when the scenario runs, which fields it receives and what each app should do. The complete workflow is tested on a real example before it is enabled for customers.

Does the Telp subscription include Make costs?

No. Costs and limits for your Make account are separate. Additional steps and more frequent runs can affect your usage in Make.

What if the first Make step succeeds but the next one fails?

A partially executed scenario is not a completed job. Before launch, agree how your team sees a failed step and continues the process; the agent tells the customer only the confirmed result.

Want to validate this workflow with your own examples?

Tell us the channel, current process and first result you want to improve. We will suggest a focused rollout path and the cases your team should test.