Make Problems & Issues
SaasVisual no-code automation platform (formerly Integromat)
Much cheaper than Zapier for multi-step automations and very capable visually, but credit consumption from polling and loops surprises users, webhooks and OAuth connections fail silently, scenarios auto-disable after errors or outages, and support is ticket-only and slow on billing disputes.
Should you use Make?
Try with cautionMake works for some teams, but users report recurring problems (score 62/100). Most reported issues: Why does Make use so many credits?; Make webhook not triggering the scenario; Make scenarios turn off automatically after errors. Not a good fit for: Developers wanting code and self-hosting; Teams needing strict data residency on their own servers. Users most often switch to Zapier, n8n, Pipedream.
Top problems
- Why does Make use so many credits?
- Make webhook not triggering the scenario
- Make scenarios turn off automatically after errors
Good for
- Automation agencies and freelancers
- Ops and marketing teams with multi-step workflows
- Small businesses moving off Zapier to save money
Avoid if you are
- Developers wanting code and self-hosting
- Teams needing strict data residency on their own servers
Common alternatives: Zapier, n8n, Pipedream
Overview
Make (make.com, formerly Integromat) is a visual automation platform where you build 'scenarios' that connect apps, APIs and AI models with routers, iterators, aggregators and error handlers on a drag-and-drop canvas. It bills per credit (formerly operations): most module actions cost 1 credit, including every polling check. It sits between Zapier (simpler, pricier) and n8n (more technical, self-hostable) and is popular with agencies, ops teams and freelancers.
Make Not Working? Common Errors & Fixes
Step-by-step fixes for the errors users hit most.
Make Webhook Not Triggering Scenario
A Make webhook that doesn't start the scenario usually means the scenario is off or scheduled rather than instant, the data structure is stale, or the webhook was deactivated. Turn the scenario on with 'Immediately' scheduling, re-determine the data structure and check the webhook's Logs and Queue.
Make Webhook 'Queue Is Full' or 410 Gone Error
Make returns 400 'Queue is full' when a webhook's queue hits its plan limit (667 items per 10,000 monthly credits, max 10,000) and 410 Gone when a webhook has had no scenario attached for over 5 days. Clear the queue and turn the scenario on, or reattach the webhook.
Make Using Too Many Credits (Operations)
Make burns credits fastest through polling triggers, which cost 1 credit per check even when nothing is found, and through loops that run many modules per bundle. Switch to webhooks, poll less often and filter right after the trigger.
Make Scenario Turned Off Automatically
Make disables a scenario's schedule after errors across several consecutive runs, and immediately on fatal errors like DataSizeLimitExceededError. Outages have also auto-disabled scenarios. Find the failing module in History, fix it, then switch the scenario back on.
Make 'Invalid Refresh Token, Please Reauthorize the Connection' Error
This error (often shown as invalid_grant) means the OAuth token for Gmail, Google Drive or Microsoft 365 expired or was revoked. Reauthorize on the Connections page; if it recurs about every 7 days, your custom Google Cloud app is still in testing mode and must be published.
Make RateLimitError (429 Too Many Requests)
RateLimitError is the HTTP 429 response from the connected app (or from Make's own webhook limit of 300 requests per 10 seconds). The app's limit is shared across all your scenarios, so slow the scenario down with sequential processing, sleeps or batching.
Make ExecutionInterruptedError (Scenario Timeout)
ExecutionInterruptedError means the run exceeded Make's maximum execution time: 40 minutes on paid plans, 5 minutes on Free. Remaining bundles aren't processed. Split the job into smaller chained scenarios and lower search limits.
Make Google Sheets Watch New Rows Not Working
Watch New Rows is a polling trigger that tracks row positions, not content. It skips rows added after a blank row, ignores rows re-added where rows were deleted, and won't fire if the scenario only runs on demand. Reset the starting point or use Search Rows with a status column.
Make BundleValidationError: Missing Value of Required Parameter
BundleValidationError means a module received a bundle with an empty required field or the wrong data type, usually because an earlier search returned nothing. Add a filter before the module, use ifempty(), or add a Skip error handler.
Make DataSizeLimitExceededError
DataSizeLimitExceededError is a fatal error tied to your plan's data transfer or file size limits; Make immediately disables the scenario regardless of the consecutive-errors setting. Reduce the data passed through the scenario or buy extra credits / upgrade.
Is Make Down? Scenarios Not Running
If many scenarios stop at once, check status.make.com. Recent incidents (Sept 2025, Dec 2025, March, July, Aug and Oct 2026) stopped executions, left webhooks unprocessed and auto-disabled scenarios that had to be re-enabled.
Signals & Issues
Patterns extracted from real user feedback — not raw reviews.
Every module action costs a credit, and polling triggers (Watch modules) spend a credit on every check even when nothing new is found. A trigger polling every minute burns roughly 44,000 credits a month on its own, and loops through iterators multiply usage per bundle.
On August 27, 2025 Make replaced operations with credits as its billing unit. Standard actions still cost 1 credit, but features using Make's built-in AI provider are billed by actual usage and can consume several credits per run, so AI-heavy scenarios got more expensive.
From November 6, 2025 extra credits cost 25% more than plan credits whether bought manually (previously no premium) or via auto-purchase (previously 30%). Core was capped at 300,000 credits per month, pushing heavier users to Pro.
One of the most common community threads: webhooks that only fire on 'Run once', stop working after weeks, or never receive data from the sender. Causes range from the scenario being off or scheduled instead of instant, to wrong data structure, to webhooks auto-deactivated after 5 days unused.
Make disables a scenario's schedule after several consecutive errors and immediately on fatal errors like data size limits. This protects credits but means automations silently stop until someone notices and re-enables them.
Incidents in Sept 2025 (all scenarios, ~2 hours), Dec 2025 (Cloudflare), March 2026 (US2 webhooks), July 2026 (~250 scenarios disabled for about 29 hours) and Oct 2026 (Scenario app failures) left automations stopped or webhooks unprocessed, often requiring manual re-enabling.
Gmail, Google Drive and Microsoft 365 connections regularly fail with 'invalid_grant' or 'Invalid refresh token, please reauthorize'. Some users hit it weekly because their Google Cloud app is in testing mode, and in some threads reauthorizing doesn't fix it.
Trustpilot reviewers describe weeks without replies and refunds only after pushing, including a case of credits charged multiple times due to platform lag. Community users note the support ticket form is the only direct contact channel.
Reviewers consistently say Make takes days to weeks before building complex scenarios comfortably. Concepts like bundles, iterators, aggregators, mapping and filters confuse non-technical users who expected Zapier-level simplicity.
Capterra reviewers report that errors are hard to trace and fix, and community threads show generic messages like BundleValidationError usually point to an empty value from an upstream module rather than the module that failed.
Paid plans stop a run after 40 minutes (5 minutes on Free) with ExecutionInterruptedError, leaving remaining bundles unprocessed. Large syncs must be split into chained scenarios or batched.
Instant triggers and bulk processing frequently hit third-party API limits (Webflow, Google Drive), and the app's limit is shared across all your scenarios. Make itself also caps incoming webhooks at 300 requests per 10 seconds.
Reviewers mention UI sluggishness, lag and occasional freezes when scenarios grow to many modules.
What Users Love
Far cheaper than Zapier for multi-step workflows
10,000 credits start at $9/month (annual), which buys much more automation than an equivalent Zapier plan for multi-step scenarios.
Powerful visual scenario builder
Routers, iterators, aggregators and filters on a visual canvas handle branching and bulk data that linear tools can't.
Large app library plus HTTP module
Thousands of prebuilt app modules and a generic HTTP/webhook module cover almost any API without code.
Built-in error handlers that don't cost credits
Error handler routes (Resume, Skip, Break, Rollback) and routers don't consume credits, and incomplete executions can be stored and retried.
Usable free plan for testing
The Free plan includes 1,000 credits a month and 2 active scenarios, enough to learn the tool and run light automations.
Pricing Plans
Storage: 1 MB data storage
- ✓1,000 credits/month
- ✓2 active scenarios
- ✓15-minute minimum interval
- ✓1 data store
Limitations: 5-minute max execution time, 5 MB file size, 7-day log retention.
- ✓10,000 credits/month (scalable up to 300,000)
- ✓Unlimited active scenarios
- ✓1-minute minimum interval
- ✓40-minute max execution time
Limitations: $9/month billed annually ($108/year). 100 MB file size, 30-day logs. Unused credits expire at end of term.
- ✓10,000 credits/month (scalable up to 8M)
- ✓Priority execution
- ✓Full-text execution log search
- ✓250 MB file size
Limitations: $16/month billed annually ($192/year).
- ✓10,000 credits/month (scalable)
- ✓Team roles and permissions
- ✓Shared scenario templates
- ✓500 MB file size
Limitations: $29/month billed annually ($348/year).
- ✓Custom credits
- ✓60-day log retention
- ✓1,000 MB file size
- ✓Invoice / wire transfer payment
Limitations: Custom pricing via sales.
Features
Automation
- Visual scenario builder
Routers, iterators, aggregators, filters
Triggers
- Instant webhooks
Queue size scales with plan credits (667 per 10k, max 10,000)
Reliability
- Error handlers
Resume, Skip, Break, Rollback, Commit; don't consume credits
- Incomplete executions storage
AI
- Custom AI provider connections
All paid plans since Nov 2025
Deployment
- Self-hosting
Cloud only
Developer
- Custom code in modules
Limited compared with n8n Code nodes
Who Should Use This?
Good Fit
Automation agencies and freelancers
Visual builder and low per-credit cost suit building many client workflows.
Ops and marketing teams with multi-step workflows
Branching and bulk processing are much cheaper than on Zapier.
Small businesses moving off Zapier to save money
Same integrations for a fraction of the price, if someone can learn the builder.
Neutral
Complete beginners wanting simple two-step zaps
Works, but Zapier is easier to learn for simple triggers and actions.
High-volume polling workloads
Every check costs a credit; switch to webhooks or costs climb quickly.
Avoid
Developers wanting code and self-hosting
No self-hosting and limited custom code; n8n fits better.
Teams needing strict data residency on their own servers
Make is cloud-only.
Regret Patterns
Common buyer's remorse scenarios reported by users.
Users set Watch triggers to every minute or every few minutes and burn most of the monthly allowance on empty checks.
A scenario hit consecutive errors or an outage, Make turned it off, and nobody noticed for days.
Beginners switch for price, then struggle with bundles, iterators, aggregators and data mapping.
Google and Microsoft connections expire with refresh-token errors, breaking scenarios until reconnected.
Users charged extra credits by lag or auto-purchase report long waits and resistance to refunds.
Scenarios using Make's AI provider consume variable credits, making costs less predictable than flat operations.
When It Breaks Down
Scenarios where this product tends to fail users.
Each check costs a credit; a 1-minute poll uses ~44,000 credits a month before doing any work.
Runs end with ExecutionInterruptedError and leftover bundles are not processed.
More than 300 requests per 10 seconds return 429, and full queues reject data with 'Queue is full'.
Rate limits are shared across scenarios, so 429 RateLimitError appears under load.
Refresh tokens expire about weekly, so connections fail with 'invalid refresh token'.
Webhooks received during incidents may not be processed and scenarios may be auto-disabled.
Alternatives Users Switch To
Zapier
9x mentionedEasiest automation tool with the largest app library and better beginner onboarding; trade-off is a much higher price per task, especially for multi-step workflows.
n8n
9x mentionedSelf-host for free with unlimited executions, bills per workflow run in the cloud and allows JavaScript/Python code; trade-off is a steeper learning curve and server maintenance.
Pipedream
5x mentionedDeveloper-first automation with code steps in Node.js/Python; better for API-heavy work, worse for non-coders.
Microsoft Power Automate
5x mentionedIncluded with many Microsoft 365 plans and strong for Microsoft-centric workflows; clunkier outside the Microsoft ecosystem.
Activepieces
4x mentionedOpen-source, self-hostable no-code builder with a simpler UI; smaller integration library.
IFTTT
3x mentionedVery simple consumer automations at low cost; not suitable for business workflows with branching or data processing.
Frequently Asked Questions
See how Make compares in our Best Automation Software rankings, or calculate costs with our Budget Calculator.
Work at Make or spotted something outdated? Suggest a correction — we verify every change against public sources. See our methodology.