Make (formerly Integromat, now owned by Celonis) is a no-code automation platform that builds workflows on a visual canvas: each app is a circle, and data flows from one to the next. Besides connecting apps, it has AI steps for reading, summarizing and classifying text, and agents that decide for themselves which tool to call. Its main difference from Zapier is that it handles complex logic, such as branching, looping over a list and transforming data, more easily and more cheaply.

What is Make good for?#

  • Processing store orders: it picks up WooCommerce orders, branches by city or amount, and runs a separate job for each group.
  • Extracting data from text: it reads incoming emails or invoices with an AI step and saves structured output (such as JSON) to a spreadsheet.
  • Publishing content across channels: from one WordPress article, it generates short versions for several social networks and queues them for posting.
  • Cleaning data: it goes through the customer list in an Excel file row by row, removes duplicates and standardizes columns.
  • A support-reply agent: an agent that suggests an answer to a customer's order-status question by looking at the orders sheet.

Strengths and weaknesses#

Strengths:

  • A visual canvas that shows the data flow clearly and makes debugging easy.
  • Strong tools for working with arrays, JSON and HTTP without writing code.
  • Usually cheaper than Zapier at high volume.

Weaknesses:

  • Harder to learn than Zapier; concepts such as iterators and aggregators take time.
  • Agents and advanced AI features are not in the free tier.
  • Services without a ready-made connector require the HTTP module.

Free tier and pricing#

The free tier gives a few active scenarios with a monthly credit cap and a limited run interval, enough for learning and light jobs. Paid plans unlock more credits, faster runs and AI agents. Cost is calculated by credit usage. Check the official site for current pricing.

Language support#

The interface is in English, but text in other languages moves through scenarios without trouble, and the AI steps understand instructions in many languages; output quality depends on the model you pick. Test non-English prompts on real data before relying on them.

Getting started in three steps#

  1. Create an account and pick a ready-made template close to your need.
  2. Connect the apps and use the run-once option to watch the data path on the canvas.
  3. When you are confident, set the schedule and activate the scenario.

Make alternatives#

  • Zapier: simpler with more ready-made integrations, but more expensive.
  • n8n: logic similar to Make, with the option to self-host for free.
  • Lindy: for people who want a ready-made email and meeting assistant rather than designing workflows.

If you want structured data from AI inside your scenarios, getting structured output (JSON) from AI will help, and to get started, read automating repetitive work with AI: where to begin.