ContactBook a call

Operations

Who maintains a Zapier, Make or n8n setup after launch

Automations break after launch when apps change their APIs, logins expire or task limits run out. What to check each month, and who should own it.

Cover for “Who maintains a Zapier, Make or n8n setup after launch”, with the line: Automations rarely fail loudly, so someone has to be told when they stop.
Automations rarely fail loudly, so someone has to be told when they stop. Image: SBD.

A Zapier, Make or n8n setup needs a named owner after launch, because the apps it connects keep changing underneath it. Logins expire, APIs change fields, plans run out of tasks or operations, and a step can fail quietly for weeks before anyone notices the missing invoices or leads. The owner can be someone on your team with an hour a month, or an outside team on a care plan. What matters is that error alerts reach that person, and that they run a short set of checks every month.

This post is for operations leads and owners who had automations built, by a freelancer, an agency or a colleague, and now rely on them. It covers why automations break after launch, the monthly checks we run, how to set up error alerts on each platform, and how to decide who owns the work.

The short answer

Pick one named owner. Turn on error notifications in Zapier, set sensible error handling in Make, and add an error workflow in n8n, so a failure reaches that person within the hour. Once a month, check connections, task or operation usage, failed runs and any app that announced API changes. Keep a one-page list of every automation, what it does and who relies on it.

Why automations break after launch

An automation is a chain of calls to other companies’ software. Your builder tested it against those apps as they were on the day of launch. After that, each link can change without warning to you.

Connections expire. Most steps sign in to an app with OAuth, and the token behind that connection can stop working. Google’s OAuth documentation lists several reasons a refresh token stops working, including the user removing access, the user changing their password where Gmail scopes are involved, and six months without use. Projects left in Google’s testing status issue refresh tokens that expire after seven days, which catches out setups built quickly on a custom Google app. A connection made by a former employee will also stop working when their account is closed.

APIs change. Apps rename fields, retire endpoints and change limits. The automation platform usually updates its built-in app, but a custom HTTP step or webhook written against the old API will fail or, worse, keep running with a field that is now empty.

Limits run out. Zapier counts tasks, Make counts operations or credits, and n8n Cloud counts executions. Growth in volume, or one loop that runs more often than expected, can use a month’s allowance early. Zapier’s help centre says runs over the task limit are held, so nothing is lost, but nothing moves until someone acts.

Failures are quiet. A step that errors is visible in the platform’s history. A step that runs but writes the wrong value, or a filter that stops passing records because a dropdown option was renamed, shows no error at all. Those are the ones that cost the most, because people find them weeks later.

The process changes. Someone adds a new deal stage, a new product line or a new team. The automation still runs, but it now does the wrong thing for part of the work.

Set up error alerts on each platform

Before anything else, make sure a failure reaches a person. Each platform does this differently.

Zapier. Zapier’s help centre describes four settings for error notification emails: immediately (the default), immediately then an hourly summary, an hourly summary only, and never, which Zapier says is not recommended. Check that the setting is on for the account that owns the Zaps, and that the emails go to an inbox someone reads. Zapier also notes that no error email is sent when an error handler in the Zap has dealt with the failure, so if your builder added error handlers, make sure they alert someone too.

Make. Each scenario has a setting for how many consecutive errors it allows before Make deactivates it, and Make’s help centre says a scenario that starts with an instant trigger is deactivated after the first error. With “Store incomplete executions” switched on, a failed run is saved so it can be fixed and resumed. Make’s documentation warns that when the incomplete executions folder is full, failed data is discarded and cannot be recovered, so keep an eye on it. Add error handler routes on the steps that matter, with a final step that posts to Slack or email.

n8n. n8n lets you create an error workflow that starts with the Error Trigger node, then select it under each workflow’s settings. When a run fails, the error workflow receives the workflow name, the failed node, the error message and a link to the execution. One shared error workflow that posts to a team channel or creates a ticket covers every workflow you attach it to.

For quiet failures, add a check on the outcome as well as the run. A daily count of new records in the destination, compared with the source, catches the filter that stopped passing data.

The monthly checks

Once a month, the owner should spend about an hour on these:

  1. Failed and held runs. Open the history on each platform and read every failure since last month. Fix the cause, then replay the runs that need replaying.
  2. Connections. Look at the list of connected accounts. Reconnect anything flagged, and move any connection that uses a former employee’s login to a shared service account your company controls.
  3. Usage against the plan. Compare task, operation or execution use with the plan limit. If you are above about 80% by mid-month, find the step using the most before you upgrade.
  4. App changes. Skim the changelogs or emails from the apps your automations depend on, and note any deprecation dates.
  5. Self-hosted n8n. If you run n8n yourself, apply updates, check disk space and confirm the database backup ran.
  6. The automation list. Update the one-page list of what exists, what it does and who relies on it. Delete or switch off what nobody uses.
  7. A short note. Write three lines on what failed, what was fixed and what changed. That record is what lets someone else take over later.

Who should own it

There are three common answers, and each works in the right setting.

Someone on your team. Fine for a handful of automations that do not touch money or customers. They need admin access, the time each month, and enough confidence to read a failed run. The risk is that maintenance slips when they are busy or leave.

The original builder. Good if they are still around and offer support. Agree response times and a monthly fee in writing, and make sure every account is in your company’s name rather than theirs. Our post on what to do when your developer leaves covers why that matters.

A maintenance team. Sensible once automations handle orders, invoices, leads or payroll, where a day of failure costs real money. Our software maintenance service covers Zapier, Make and n8n, whoever built them. A named engineer handles alerts, fixes, platform and API changes and small requests, and sends a monthly report. Care plans start at $50 a month for one or two automations, and the Automation team plan at $999 a month works through a steady stream of small requests.

If something is already broken

If an automation is failing now, fix that first. Our fix broken automation service is built for that case. Once it is running again, set up alerts and the monthly checks above so the same thing does not happen quietly next quarter.

If you are taking over a setup you did not build, start with a review. The Automation Audit starts at $500 and maps what exists, what is fragile and what each fix costs. For a sense of the total monthly cost of looking after a setup, see Software maintenance cost. If platform pricing is part of the question, our comparison of n8n, Zapier and Make at 10k and 100k tasks and the Zapier tax in 2026 go into the numbers.

Questions we get asked

How often do automations break?

It depends on how many apps they touch and how often those apps change. A simple two-step Zap can run for years. A setup with custom API calls across several tools usually needs attention every month or two.

Will Zapier or Make tell me when something fails?

Both can email you about errors, if the settings are on and the email goes to someone who reads it. Neither will tell you when a step runs successfully with the wrong data, so add an outcome check for the important flows.

Is n8n harder to maintain than Zapier?

Self-hosted n8n adds server work: updates, backups and uptime. n8n Cloud removes most of that. The workflow maintenance itself is similar on all three.

Can you maintain automations someone else built?

Yes. We review them first, then quote a care plan based on what we found.

Sources and further reading

  • Zapier
  • Make
  • n8n
  • Software maintenance

Keep reading

Cover for “How to automate data entry, step by step”, with the line: Capture it once, check it, and write it where your team already works.

Explainer

How to automate data entry, step by step

How to automate data entry: find the sources, choose capture, add checks and review, write to the system of record, and measure the time saved.

Cover for “n8n vs Zapier vs Make at 10k and 100k tasks a month”, with the line: n8n counts whole workflow runs instead of steps.

Costs

n8n vs Zapier vs Make at 10k and 100k tasks a month

What Zapier, Make and n8n cost at 10,000 and 100,000 tasks a month, from each vendor’s pricing page on 14 Nov 2026, and when to pick which.

Cover for “The Zapier tax in 2026: what a 20-person team pays”, with the line: Three meters now run at once: tasks, AI steps and activities.

Pricing

The Zapier tax in 2026: what a 20-person team pays

Zapier now bills tasks, AI steps by model tier, and Agents in activities. A worked bill for a 20-person company, and what the same workflows cost on n8n.

Find out what is worth automating

About 95% of AI pilots never reach the profit line. We start with the work your team still does by hand and measure each build by the hours it saves.

See how we work

Cookie preferences

Choose what this site may use on your device. You can change this at any time from “Cookie settings” in the footer.

Strictly necessary

Keep the site working and remember the choices you make here. These cannot be switched off, and none of them track you.

Always on
What this includes (4)
NameProvider PurposeDuration
sbd-consentThis site Remembers your cookie choices12 months
Vercel Web AnalyticsVercel Inc. Counts page views in aggregate. Sets no cookie and stores nothing on your deviceNone
ticker-dismissed, sbd-note-offThis site Remembers that you closed a notice or language tipUntil the tab closes
sbd-voice-resume, sbd-voice-prefillThis site Carries a voice conversation across pages, only if you start oneUntil the tab closes

Analytics

Tell us which pages are read and where visitors leave, so we can improve them. IP addresses are anonymised and nothing is used for advertising.

What this includes (2)
NameProvider PurposeDuration
_gaGoogle Analytics (Google LLC) Distinguishes visitors with a random ID13 months
_ga_FPXDMQPRV1Google Analytics (Google LLC) Keeps the session state for this site13 months

Functional

Embedded services that set their own cookies. Without them the site still works; you get a link to open the service instead.

What this includes (1)
NameProvider PurposeDuration
Booking calendarCal.com Inc. Shows live availability and books the call on the booking pageSet by Cal.com

Advertising

Not used. This site has no advertising, retargeting or social media tracking, and we never sell data.

Not used

Questions about your data: hello@smallbaydigital.com