Make.com: what it is, how credits work, and who it fits

What Make is

Make is a visual automation platform, found at make.com. You build scenarios: chains of modules connected on screen. The first module, the trigger, watches an app or receives an event. The modules after it act: add a row to a spreadsheet, send an email, update a customer record.

An example: each time someone fills in a form on your website, the scenario creates the contact in your CRM, sends a confirmation email and notifies the team.

On the vendor side, the About page states that Make is part of Celonis and places its headquarters in Prague. The legal name of the company and its founding year are not recorded.

What Make does in practice

On the number of connected apps, the vendor’s pages contradict each other: More than 1,000 apps according to the pricing FAQ; "3,500+ apps" on the plan cards of the same page source.

The limits that matter day to day depend on the plan:

• number of scenarios: 2 active scenarios (Free); unlimited on paid plans;
• frequency: the minimum interval between two scheduled runs is 15 minutes (Free); 1 minute (Core, Pro, Teams, Enterprise);
• run time: up to 5 minutes per run on the free plan, 40 minutes from Core up;
• code: "Make Code App": not included in Free; 2 credits per second of execution on paid plans.

Routers, which split a scenario into several branches, and filters are available on the free plan and use no credits.

What Make bills

The basic math is a multiplication. A scenario with 4 modules (a trigger and 3 actions) that processes one item uses 4 credits. Started 200 times a month by a webhook, meaning only when an event arrives, it uses 200 × 4 = 800 credits.

The catch is the scheduled trigger. When a scenario checks at a regular interval whether there is anything new, every check uses 1 credit, even if it finds nothing:

• every hour: 24 × 30 = 720 credits a month for the trigger alone;
• every 15 minutes: 96 × 30 = 2,880 credits a month;
• every minute: 1,440 × 30 = 43,200 credits a month.

The free plan includes 1,000 credits. A single scenario scheduled every 15 minutes, the minimum interval on that plan, goes past it on checks alone, before it has processed anything.

When the app supports it, a webhook trigger avoids the empty checks. The guide what is a webhook? explains the difference, and the guide credits, tasks, executions compares the units used by different tools.

The plans in brief

Make sells two things at once: a feature level and a volume of credits.

• Free: 1,000 credits a month, 2 active scenarios, no access to the Make API.
• Core: $16 per month ($11 per month billed annually) for 10,000 credits, unlimited active scenarios, one-minute interval.
• Pro: from $28 per month, with priority execution and custom variables.
• Teams: from $51 per month, with teams and team roles.
• Enterprise: custom pricing.

The pricing page mentions no trial period for the paid plans: you start on the free plan. The full table and the volume levels are further down, with the date recorded. The Make pricing page works through three scenarios in credits, and the cost calculator compares one scenario across several tools.

Hosting and data

On self-hosting, the pricing page answers: No: no self-hosted plan on the pricing page; AWS hosting (EU, North America) for every plan source.

On where data is stored, the help center says: Data center in the US or the EU, chosen when the organization is created and not changeable afterwards source. That choice is made once: decide before you create your organization.

One caveat, stated by the vendor itself: an EU data zone does not guarantee that all AI processing stays in the EU. Maia and Make AI Web Search are processed in the United States, and with Make’s AI provider, the region depends on the model you choose.

AI and agents

An AI agent is a program that lets a language model decide which actions to take. For Make, the help center says: "Make AI Agents" (beta), every plan: 1 credit per operation, plus token-based credits with Make's AI provider source. MCP, the protocol that lets an AI assistant call outside tools, is covered: "MCP server" included in every plan according to the pricing page.

What an agent uses depends on where the model comes from:

• with “Make’s AI Provider” (every plan): 1 credit per operation, plus credits based on the AI tokens used;
• with a connection to your own AI provider (paid plans): 1 credit per operation, with tokens paid directly to the provider.

In the first case, you can no longer work out credit usage from the module count alone: it also depends on how long the exchanges with the model are.

Which setup for your situation

• You have one or two short scenarios, started by an event: the free plan may be enough. Count the modules, multiply by the number of runs, compare with 1,000.
• You need to check an app at a regular interval: count the checks first. At one check every 15 minutes, you are already past the free plan.
• Your scenarios are long and run often: do the math before you subscribe, because every extra module multiplies usage.
• You need to keep the tool on your own servers: Make is not built for that.

Alternatives to Make

The difference that weighs most is the billing unit. n8n counts one execution per complete run of a workflow, however many steps it has, and offers a free edition you host yourself. Zapier counts tasks, meaning successful action steps; its triggers use none, polling included.

The head-to-heads n8n vs Make and Make vs Zapier run the math on the same scenario. For the big picture, see the automation tools comparison.

What the documentation does not say

The vendor’s legal name and its founding year are not recorded. Neither are the languages of the interface and of support: none of the pages read states them.