# Make.com Operations Limit Exceeded? How to Find the Cause Fast

**By StackTrekker Editorial** · Published October 9, 2026
Make.com counts one operation for every module that runs, and the count does not depend on whether the module changed anything. The monthly allowance runs out early for a single reason: a scenario is running far more often than the data actually changes. Three settings do almost all of the damage. A polling trigger on a short interval fires on a timer and spends an operation every time, even when the source has nothing new. A filter placed after a module lets every bundle reach a step that already counted. An error handler that retries the whole flow runs the earlier steps a second time. You do not need a larger plan to stop this. Open the usage view, sort scenarios by operations, then lengthen the interval, move the filter to the first step, or replace the polling trigger with a webhook. Most of the waste disappears at the trigger.

## What counts as an operation in Make.com?

An operation is one module execution. Every time a module runs, Make counts one operation against your account, whether the module creates a record, updates one, or only passes data through. A trigger that polls a source counts too, so the first module in the chain is never free. That makes the number at the end of the month a multiple of two things: how many times the scenario ran, and how many modules each run touched. A three-module scenario that runs 500 times spends 1,500 operations, not 500.

The distinction that catches teams out is between a module running and a module doing useful work. Make counts the run, not the result, so a step that finds nothing and a step that writes a row cost the same. The [Make Help Center](https://help.make.com/operations) defines operations and how each module is counted.

## Why did your operations run out before the month ended?

When the allowance disappears in days rather than weeks, the schedule is almost always the cause. Make charges per run, so the interval on the trigger sets the pace of the spending. Every 15 minutes is 96 runs a day, which is about 2,880 runs in a month. Give the scenario three modules and that is nearly 8,600 operations a month spent to watch a source that may have produced ten new items.

A form that receives a few responses a week does not need to be checked 96 times a day. The same trap appears with scenarios that were built for a one-off job and then left switched on. A scenario you ran by hand to import a backlog keeps running on its schedule until you switch it off. Before you change anything, run the arithmetic for the busiest scenario: runs per month multiplied by the module count. The result usually points straight at the cause.

## Which settings burn the most operations?

The table below lists the four settings that turn a cheap scenario into an expensive one, and the change that caps each of them.

| Setting | Why It Burns Operations | How To Cap It |
|---|---|---|
| Short Polling Interval | The Trigger Runs On A Timer Whether Or Not The Source Changed | Set The Longest Interval The Work Allows |
| A Filter After A Module | The Module Before It Already Ran And Counted | Move The Filter To The First Step |
| A Repeating Error Handler | Each Retry Runs The Earlier Modules Again | Fix The Cause And Stop The Retries |
| A Left-On Test Scenario | A Scenario Built For One Job Keeps Running | Switch It Off When The Job Is Done |

Polling is the biggest lever because it multiplies. Slowing an interval from 15 minutes to an hour cuts the run count to a quarter at once, and moving to a webhook removes the idle runs entirely.

A filter in the wrong place stops a bundle, not a run, so the step ahead of it is already paid for. Move the filter onto the trigger's connection and the later modules never execute for filtered-out items. If you prefer to compare the logic with a different tool, the way [Zapier filters placed after an action](/zapier-filter-not-working/) still let the earlier step run follows the same rule.

## How do you find the scenario that used them?

Start where the numbers are. Open your account menu and find the usage area, which breaks operations down by scenario for the current period. The scenario at the top of that list is your culprit, not the one you suspect.

If the usage view only shows a total, open the History tab on each scenario and read the run count, then multiply it by the number of modules on the canvas. A scenario with a large run count and a short interval is spending on the timer. A scenario with a small run count but a high module count is spending inside a single run, which usually means a loop or an iterator is processing far more items than you expected.

Write the scenario name down before you change anything, then compare the next period against this one. Our [Make.com tutorial](/make-com-tutorial/) shows where the History tab sits and how to read a single run.

## How do you stop it from happening again?

Four changes remove most repeated overspend, and none of them needs a plan upgrade.

1. Replace polling with a webhook where the source supports it. A webhook fires only when an event arrives, so an idle source costs nothing.
2. Set the interval to the longest value the work tolerates. A weekly report does not need a 15-minute check.
3. Move every filter onto the trigger, before any action runs.
4. Switch scenarios off when they are not needed, and use an error handler that stops rather than retrying a failure on a loop.

If the scenario writes to Notion, half the cost sits in the destination. A slow or rate-limited write can make a run retry, and a property-type mismatch is a common trigger for that. Our [Notion sync error fix](/notion-sync-error-fix/) covers the write side, while a data mismatch that stops a later module is closer to the checks in our [Make.com bundle validation error guide](/make-com-bundle-validation-error/).

Self-hosted tools such as [n8n count executions rather than operations](https://docs.n8n.io/build/understand-workflows/understand-executions), which is why teams that self-host stop watching a monthly cap. If you stay on Make, the habit that matters is the monthly review: once a period, sort scenarios by operations and look for a run count that has no matching change in the data.

## FAQ

**Why did Make.com use up my operations in a few days?**

A scenario on a short polling interval runs on a timer whether or not the source has new data, and every run spends one operation for each module it touches. Every 15 minutes is 96 runs a day, so a three-module scenario reaches about 8,600 operations a month even when nothing changes. Lengthen the interval or move the trigger to a webhook.

**Does a scenario spend operations when nothing changes?**

Yes. Operations are counted per module run, not per result. A polling trigger that finds no new items still runs and still counts, so an idle scenario on a short interval keeps spending until you switch it off or slow it down.

**Do triggers count as operations in Make.com?**

A polling trigger counts as one operation each time it runs, because the trigger is itself a module. A webhook trigger only runs when an event arrives, so it spends an operation only when there is real data to hand to the next module.

**How do I see which scenario used the most operations?**

Open the usage area of your account, which breaks operations down by scenario for the current period, and read the top entry. If only a total is shown, open each scenario's History tab, read the run count, and multiply it by the number of modules on the canvas.

## Sources

- [Make Help Center: operations](https://help.make.com/operations)
- [Make Help Center: schedule a scenario](https://help.make.com/schedule-a-scenario)
- [Make Help Center: fix errors and warnings](https://help.make.com/fix-errors-and-warnings)
- [n8n documentation: understand executions](https://docs.n8n.io/build/understand-workflows/understand-executions)
- [Notion Help Center](https://www.notion.com/help)

---

Source: https://stacktrekker.com/make-com-operations-limit-exceeded/