If you're starting out with automation in your business, sooner or later you'll face the same question: n8n or Make? Both tools promise to automate your processes without you needing to write any code. And both tools deliver on that promise – but for completely different kinds of businesses.
Here's an honest comparison, drawn from what we experience day in and day out in the Bad Tölz-Wolfratshausen area and beyond.
The biggest difference between n8n and Make isn't about features. It's about how the two tools count.
According to Justinkeirath (2026): Make counts every single operation – meaning every module that runs inside a workflow. n8n, on the other hand, counts a workflow execution as a whole, regardless of how many steps are involved. That sounds like a technical detail, but it quickly becomes a cost problem.
A concrete example: you have a workflow that receives a new customer enquiry, writes it to your CRM, sends an email, creates an entry in Google Sheets and triggers a Slack message – that's five operations in Make. In n8n: one execution. According to ConnectSafely.ai (2026), n8n can be 10–20x cheaper than Make for complex workflows with many steps.
For simple workflows with just a few steps, Make is often cheaper and quicker to set up. That's the honest picture.
Our verdict for Criterion 1:
This is the point where most comparisons stop. And yet that's exactly where things get interesting.
Pure automation – moving data from A to B, triggering forms, sending notifications – you can do with either tool. But as soon as you want to bring AI into your processes – for example to classify incoming enquiries, automatically draft quotes or intelligently evaluate customer data – you need a tool flexible enough to talk to APIs like the Claude API or OpenAI.
This is where n8n has the edge. The self-hosted version gives you full control over which AI models you connect – whether OpenAI, Mistral or a locally running model via Ollama. You're not tied to a specific provider, and your data doesn't necessarily leave your own system. For trades businesses or service providers in Bavaria that handle sensitive customer data, that's not a trivial point.
Make does offer AI integrations as well, but connecting your own models or local instances is considerably more restricted.
What this means in day-to-day work:
A plumbing business from Holzkirchen automated their quoting process using a combination of n8n and the OpenAI API. Incoming enquiries via the contact form are automatically categorised, relevant data is pulled from the CRM and a draft quote is generated – the employee reviews it and sends it off. Result: 70% less manual effort in quote preparation, with no developer permanently involved. If you can picture something like this for your own trades business, you'll find a concrete example from our practice at /projects/ki-angebotserstellung-handwerk.
A strong opinion we're happy to stand by: Most automation tools are marketed as "AI-ready" – but AI-ready doesn't mean the same thing as AI-capable. Anyone who seriously wants to integrate AI into their workflows needs a tool that doesn't fall apart at the first complex API request.
This is the question that should be asked before making a decision, but rarely is.
Make is cloud-only. You sign up, build your workflow, done. No server, no updates, no technical knowledge required. That's a real advantage for small businesses that don't have anyone to look after infrastructure.
n8n comes in a cloud version (similar to Make) and as a self-hosted option. The self-hosted version is significantly cheaper – according to Exponentieller Wandel (2026), switching from Zapier Pro (around €220/year for 1,500 tasks/month) to n8n self-hosted brought infrastructure costs down to roughly €221.76/year – for two complete server instances with unlimited executions. That's a different price bracket altogether.
But self-hosted means: someone has to set up the server, configure backups and roll out updates. That's not rocket science, but it requires either in-house know-how or a partner who takes care of it.
A real downside you should know about: n8n self-hosted sounds temptingly cheap – and it is. But if your workflow goes down one day and nobody in your business knows how to restart a Linux server, you've got a problem. The cost savings can quickly be eaten up by downtime or external help. That's not a reason against n8n, but it is a reason to honestly answer the question of who will operate it before you get started.
Softstorm observation: In 8 out of 10 of our automation projects, the first bottleneck wasn't the choice of tool – it was the question of who in the business will maintain the workflows long-term. Sorting that out before you start saves you most of the headaches.
If you want to know which tool concretely fits your business in the Wolfratshausen, Miesbach, Geretsried or Rosenheim area – and how to build automation that still runs smoothly in two years' time – a short conversation is often worth more than any comparison article.
We analyze your processes, identify automation potential and show you which digital systems really make sense.
