Notes/06
AI AutomationDec 2025·5 min read

Make.com vs n8n for serious automation work

We've built production workflows on both. Here's an honest comparison — what each platform is actually good for, where they break, and how we choose.

We run production automation on both Make.com and n8n. The "which one should I use?" question comes up every time we start a new project with a client. Here's the unfiltered answer.

Make.com: visual-first, hosted, fast to build

Make's strength is speed. The scenario builder is genuinely pleasant to work in — modules snap together, the data mapping interface is clear, and the error handling UI is the best in the category. For a team that needs to build fast and iterate quickly, Make wins.

The downside is pricing at volume. Make bills on operations — every module execution in a scenario counts. A moderate workflow processing 10,000 leads a month with 15 modules per run costs 150,000 operations per month. That's $29/month at the Growth plan, which is fine, but complex orchestration can balloon the count quickly.

n8n: code-adjacent, self-hostable, no ceiling

n8n is for teams comfortable in code or infrastructure. The node-based editor is less polished than Make's, but the escape hatches are better — you can write raw JavaScript or Python in a Code node, reference environment variables, and deploy the whole thing on your own server for a flat monthly cost regardless of execution volume.

For high-volume or sensitive-data workflows, n8n's self-hosted option is the right call. We run our internal lead scoring pipeline on n8n hosted on a €20/month Hetzner VPS. At 40,000 scenario runs a month, that's economically significant.

How we choose

Client-facing projects where the client's team will need to maintain the workflow: Make. The visual interface is approachable without technical training.

High-volume or data-sensitive internal pipelines: n8n self-hosted. The economics are better and we control the infrastructure.

Anything connecting to a local database or internal API: n8n, because Make's cloud can't reach private network resources without additional tunneling.

The honest caveat

Both platforms are good enough that the choice rarely matters as much as the quality of the workflow design. A well-structured Make scenario will outperform a poorly designed n8n workflow every time.

G
Georgiy
Founder, propagandas.studio · Vienna, AT
Work with us →

← Previous

Attnex: building an ad intelligence platform from scratch

All posts