Help for owners
My Make scenarios run for too long and stop before they finish
Short answer
Make has a scenario run duration limit that depends on your plan, and a run that goes over it does not create an incomplete execution to resume. Long runs come from large batches, slow connected apps and loops. Splitting the work into smaller runs, and storing failures, is the usual direction.
Does this sound like your situation?
- A scenario that handled a few records now handles hundreds and stops partway.
- Some records were processed and the rest were not, with no clear error.
- The run history shows a very long run before it ends.
- The same records are picked up again on the next run.
- I cannot find the failed run in Incomplete executions.
Why does this happen?
The run is longer than the plan allows
Make documents a scenario run duration limit, and says you can check the limit for your plan on its pricing page. It also states that a scenario that runs longer than that limit does not create an incomplete execution, so there is nothing stored to resume.
Confirmed by help.make.com
One run is asked to do too much
A scenario that reads a long list and then handles each item in turn takes longer as the list grows. A flow built for a handful of rows can cross the duration limit once the business grows.
A connected app is slow or rate limited
Each step waits for the other app to answer. Make has a guide on rate limit errors, and a slow or throttled app stretches every run.
Confirmed by help.make.com
Incomplete executions are off by default
Make states that incomplete executions are disabled by default and have to be enabled in the scenario settings. With them off, a failed run leaves no stored copy to retry.
Confirmed by help.make.com
The same records are read again
A scenario that stops before it marks records as done will pick them up on the next run. That creates doubles when the first half already reached the other system.
What can I do about it today?
- Open the run history and sort by durationNote the longest runs, how many records each one handled and where the last step stopped.
- Do not press Run once repeatedlyEach press can process the same records again and create doubles in your accounting, CRM or store.
- Check your plan's run duration limitLook at Make's pricing page or ask your account owner, and compare it with your longest run.
- List what was processed and what was notCheck the destination system for the last record that arrived. This is the dividing line for a catch-up.
- Switch nothing off yetIf the scenario feeds orders or payments, ask the person who built it before changing the schedule.
When should I bring someone in?
- Records are missing or doubled after a stopped run.
- The scenario is business critical and runs near the limit every day.
- You cannot tell which records already reached the other system.
How do I handle it?
- Measure where the time goesI read the run history and find which step and which app account for most of the duration.
- Split the work into smaller runsProcess a limited batch per run, and mark each record as done only after the other system confirms it, so the next run continues where this one stopped.
- Add an error routeI use Make's documented error handlers and the incomplete executions option so a failed record is stored, not lost. I check which errors Make does not store.
- Guard against doublesEach record carries a unique ID, and the destination is checked before a create step.
- Catch up the unfinished recordsI send the remaining records in small batches and compare counts on both sides at the end.
The price depends on what I find, so I quote in writing after a short discovery call. Tell me what is happening and I reply within 24 hours.
What do owners ask about this?
Will Make let me retry a run that timed out on duration?
Make states that a run longer than the duration limit does not create an incomplete execution. Plan on finding the unfinished records yourself or through a check built into the scenario.
Is a higher plan the answer?
The limit depends on your plan, so it may help, but a flow that grows with the business will meet the next limit later. Splitting the work is the lasting step.
Where do I see the limit?
Make points to its pricing page for the run duration limit of each plan.
Why are my records doubled after a timeout?
The scenario likely created records in the other system and stopped before it marked them done, so the next run created them again.
Sources
How was this page made?
I drafted this page with AI assistance from the official documents listed under Sources. Found a mistake? Tell me.