n8n or Make.com for a small business: how we choose

Both tools connect your apps and run the work automatically. Most comparisons put their features side by side, which is not how the decision gets made. Four questions decide it.

1. How many things happen per month?

Make charges by operation, and an operation is roughly one step in one run. A workflow with eight steps, running fifty times a day, is 12,000 operations a month. That maths is what tips businesses off Make, not its features.

n8n charges by workflow execution on its cloud plan, so a run counts once regardless of how many steps it has. For high-volume, many-step work, that is a completely different bill.

Below a few thousand operations a month, price is not the deciding factor and you should ignore it.

2. Who maintains this after we finish?

This is the one that matters most and gets discussed least.

Make is visual in a way non-technical people can genuinely read. An office manager can open a scenario, see which step failed, and often fix it. That is worth real money over a year.

n8n expects more comfort with data structures and the occasional snippet of code. It is far more powerful in the same breath. If nobody on your team will open it, that power turns into a dependency on whoever built it.

3. Can it be self-hosted?

n8n can run on your own server. That matters when data cannot leave your infrastructure, or when volume makes any per-run pricing painful. It also means you own updates, backups and downtime.

Make is hosted only. For most small businesses that is an advantage, not a limitation.

4. What do your tools already support?

Both have hundreds of connectors and both can call any API. The question is which of your tools has a deep, well-maintained connector rather than a thin one. Check the specific services you depend on, not the total count on the marketing page.

How it usually lands

Make suits a small team, moderate volume, mixed tools and a non-technical person who needs to see what is happening. Most small business automation we build sits here.

n8n suits high volume, self-hosting requirements, heavy data manipulation, or a technical team who will keep extending it.

Both is normal at a certain size: Make for the business workflows your team touches, n8n for the heavy processing behind them.

What actually breaks, on either one

Choosing the tool is the easy part. These are the failures we see in the wild, regardless of platform:

• No error handling. One failed upload stops the run, silently, and nobody notices for four days.
• Unsafe loops. When several files arrive at once, a step that is not safe to repeat delivers the first and loses the rest. We fixed exactly this on a live inspection-report system: a calendar step inside the loop was halting the attachments after the first one.
• No alert on failure. If a broken automation looks identical to a quiet week, you will find out from a customer.
• Matching on the wrong field. Product codes and IDs are often named differently in two systems. Match on something genuinely stable across both.
• Credentials in one person’s account. When they leave, the automation dies with their login.

An automation without error handling and a failure alert is not finished, whichever tool built it.