Make: automate your apps by dragging modules
Make is an online platform that connects your apps through scenarios you draw with the mouse, no code required. It is aimed first at freelancers and small businesses that want to automate repetitive tasks and add AI agents to them.
Data as of October 22, 2026
Key figures
| Developer | Make (groupe Celonis) |
|---|---|
| Type | No-code |
| Open source | No |
| License | proprietary |
| Suggested level | Beginner |
| Use cases | automation, workflows, data |
| Self-hostable | No |
| Business model | Freemium |
| Cloud offering | Yes |
| Multi-agent | Not disclosed |
| MCP support | Yes |
| Bring your own model | Yes |
| Official site | Official site · Documentation |
How scenarios work
Make is part of Celonis, a process-mining company, and is headquartered in Prague according to its official About page. The core idea is simple: a scenario is a chain of modules, each one an action in an app (Gmail, Google Sheets, a CRM, an online store and so on). You connect them on a canvas, add filters, routers and loops, and Make runs everything on a schedule or when an event happens. It is no-code automation at its most visual.
Concrete examples: pushing every new order into a tracking sheet, creating a task when a customer submits a form, posting content to several social networks, keeping two contact databases in sync.
Make AI Agents: AI in the same builder
With Make AI Agents, you build an AI agent in the same editor as your scenarios. The agent gets instructions and tools (your scenarios or modules) and decides which ones to use to handle a request: sorting inbound email, qualifying a lead, following up with a customer. Make emphasizes transparency, so you can see which decisions the agent made. The platform also offers an MCP server so compatible AI assistants can trigger your scenarios.
A word of caution: agents make mistakes. Keep a human in the loop for actions that involve money or customer relationships.
What it costs
As of October 22, 2026, Make bills in credits. The Free plan includes 1,000 credits per month. Make’s official comparison page then lists Core from $9 per month, Pro at $16 and Teams at $29, each starting at 10,000 credits per month; Enterprise is quoted on request. The pricing page notes that annual billing is cheaper than monthly and that credit volume can be scaled up. Extra credits cost 25% more than those included in your plan. AI agents are included in paid plans; if you connect your own model provider, that usage is billed separately.
Who it fits, and the alternatives
- Freelancer: Free, then Core, covers most simple needs.
- Small business: Pro or Teams to share scenarios and templates across colleagues.
- Technical team: good for moving fast, but there is no self-hosting and no access to the source code.
Strengths: quick to learn, a very visual interface, many ready-made connectors. Limits: credit usage can climb fast with chatty scenarios, and your data passes through Make’s cloud. To compare: n8n if you want to host it yourself, Zapier Agents if you already live in Zapier, and Relevance AI for teams of agents focused on sales and support.
Frameworks of the same type
- Gumloop · automation, workflows, data
- Zapier Agents · automation, web browsing, chatbot
Frequently asked questions
Is Make free?
There is a Free plan with no time limit and 1,000 credits per month as of October 22, 2026. That is enough to learn the tool and run a few small scenarios, not to support steady business use.
What is a credit in Make?
It is the billing unit: each module action (reading, searching, creating or updating data) consumes credits, usually just one. When you run out, scenarios stop until your credits renew or you buy extra ones.
Can I self-host Make?
No. Make is a hosted service only, under a proprietary license. If self-hosting matters to you, n8n and Dify offer editions you can install yourself.
Which AI model do Make AI Agents use?
On paid plans you can choose between Make's own AI provider or another provider through your own account. That choice affects both cost and where your data is processed.
Data sources
Data as of October 22, 2026
Report an error
Does a figure, a date or a fact look wrong? Tell us: every report is checked.