Make.com pricing explained: how credits work and what a scenario really costs

Make.com pricing is based on credits, and most people only find out how many a scenario uses after it has been running for a week. A workflow that looked tiny in testing can quietly use thousands of credits a month, because Make charges for schedule checks that find nothing and for every item a loop processes, not just for "runs".

The direct answer: a Make scenario costs one credit each time its trigger checks for data, plus one credit for every module action on every item (Make calls them bundles) that passes through. Routers and error handlers are free. So the two numbers that drive your bill are how often the scenario is scheduled and how many items each run processes. The sections below show how to estimate both before you switch a scenario on.

How Make counts credits

Make moved from "operations" to credits in August 2025. According to Make's pricing page, as listed in October 2026:

• Most module actions cost 1 credit. Reading data from an app or webhook, searching, creating, updating or deleting records, transforming data with built-in tools, aggregating rows and iterating over rows all count.
• Routers and error handlers cost nothing. The Router module and the error handler directives (Rollback, Break, Resume, Commit, Ignore) do not use credits.
• Some AI features cost more. Make's credits help page explains that its built-in AI provider charges by tokens and model, and that some AI content extraction modules use more than one credit per operation.

Two rules sit underneath this and explain most surprises.

The trigger costs a credit every time it checks

A scheduled scenario uses a credit every time its trigger runs, even when there is nothing new. Make's own scheduling guide gives the arithmetic: a scenario checking every 5 minutes uses 288 credits a day, or 8,640 a month, before it does any real work. Hourly checks use 720 credits a month; once a day uses 30.

The trigger is charged once per run, however many items it returns.

Every later module runs once per bundle

If the trigger returns 20 new rows, each module after it runs 20 times, and each of those runs costs a credit. A scenario with five action modules that processes 20 rows uses 100 credits for that run, plus one for the trigger.

This is why loops are expensive. An Iterator that splits an invoice into its 30 line items multiplies every downstream module by 30.

Worked example: a client intake scenario

Take a common workflow for a 20 person firm: a web form submission creates a client record in the CRM, creates a folder in file storage, and posts a message to the team channel.

Version A: polling every 15 minutes.

• Trigger checks: 4 an hour, 24 hours a day, about 30 days = 2,880 credits a month, even if no forms arrive.
• Actions: 3 modules × 60 forms a month = 180 credits.
• Total: about 3,060 credits a month.

Version B: an instant webhook trigger.

• The webhook only fires when a form actually arrives, so there are no empty checks.
• Trigger: 60 credits. Actions: 180 credits.
• Total: about 240 credits a month.

Same result, roughly a twelfth of the credits. For low-volume business workflows, the schedule often costs far more than the work.

Now add a loop. Suppose each form can attach several documents, and the scenario iterates over them to upload each one, with an average of four files per form. The upload step now runs 240 times a month instead of 60. Small changes like this are where estimates go wrong.

Estimating a scenario before you build it

Use this formula and fill it in with your own numbers:

Monthly credits = trigger checks per month + (items per month × action modules per item)

Then add a margin for retries, testing and growth. Running the scenario manually a few times and checking the credits shown in each run's details is the most reliable check.

What happens when credits run out

Make's pricing page sets out the rules, as listed in October 2026:

• Scenarios stop until you add credits or upgrade.
• Incoming webhooks are queued up to your plan's queue limit and processed once scenarios resume, so check how large your plan's queue is.
• Polling scenarios look for records created since their last successful run when they restart.
• You can buy extra credits in bundles, or turn on auto-purchase of 10,000 credits on the Core, Pro and Teams plans.
• Unused credits do not roll over. They expire at the end of the term, and extra credits expire one month after purchase on monthly subscriptions.

For a firm, the stop-and-queue behavior is the risk to plan for. A client intake scenario that silently pauses on the 26th of the month is a missed enquiry. Turn on usage notifications and decide in advance whether auto-purchase is on.

Seven ways to use fewer credits

1. Use instant triggers (webhooks) wherever the app supports them, instead of polling.
2. Schedule what must be scheduled at a sensible interval. A daily report does not need a 5-minute check.
3. Filter early. Put a filter straight after the trigger so irrelevant items stop before they reach paid modules.
4. Search less. Repeated "search for a record" steps add up; store an ID once and reuse it.
5. Batch where the app allows it. One "create many rows" call costs less than one call per row.
6. Watch AI steps. Check which model and connection an AI module uses; Make's credits help page explains that its own AI provider charges by tokens.
7. Switch off test scenarios. Make's scheduling guide specifically warns that test scenarios left active keep using credits.

Make credits vs Zapier tasks

The units are not equal. Zapier does not charge for triggers, filters or paths, while Make charges for trigger checks but not for routers. A busy, simple workflow often costs fewer units in Zapier; a branching workflow with many steps often costs less in Make. Our comparison of Zapier vs Make for a 20 person firm covers the trade-offs beyond price, and our guide to how much Zapier costs for a team explains the task side.

If any scenario touches client, financial or health data, check with your compliance lead what Make's terms and data region allow before you build it.

How Flow Mesh helps

Our stack audit reviews the automation platforms you already pay for, including the scenarios using the most credits, and gives each tool a verdict: keep, configure, consolidate or replace. You get a written scope, and any rebuild is quoted in writing after the audit and carried out inside your own accounts, as described on our workflow automation page.

Key takeaways

• Make charges one credit per trigger check plus one per module action on every bundle; routers and error handlers are free.
• Polling schedules cost credits even when nothing happens: every 5 minutes is 8,640 credits a month before any work.
• Iterators multiply every module after them, which is where most estimates go wrong.
• Estimate monthly credits before switching a scenario on, and use webhooks and early filters to cut usage.
• When credits run out, scenarios stop, so set usage alerts and decide on auto-purchase in advance.