My take is a bit different after months of experimentation - while n8n is great, I wouldn't limit myself to just one tool.
Here's why: When I started automating my own business processes (and believe me, I tried everything), I discovered that the real magic happens when you mix different approaches based on your needs. Sometimes it's n8n, sometimes Make, and sometimes good old custom code D:
Instead of focusing on tools, I'd suggest starting with a simple question: What's the smallest thing you can automate today that would make your life easier? Then pick the tool that fits that specific need.
Quick tip: Start with a $4 droplet like they suggest, but keep some budget for experimenting with other tools too. You never know which one will click with your workflow!
Fantastic thoughts Pawel, thank you for commenting.
I also agree. I don't use n8n exclusively. I sometimes code myself (especially with these new vibe coding tools it's super easy to build microservices for simple use cases) or just leverage someone else's open source repo for something. Sometimes I also use Make or Zapier, depending on the situation.
What my experience has been however is that most people getting started with no-code have near zero technical background.
They don't know how to think about problems like an engineer.
They don't know what is an API or a for loop or what data types are there.
Then it gets easy to confuse no-code with no-think and that's a good recipe for disappointment.
Most of the time sticking to one tool until you start hitting walls is a good idea. But if you're already technical enough to know all these things, then yes. You'll likely end up building your own preferred stack.
If you just create a new Droplet from within Digital Ocean's platform, you won't have Docker on it.
You can check if docker is installed by typing in this command:
docker --version
if it returns some version number, you have Docker. if it doesn't understand what you want, you don't have Docker.
you can install Docker on your virtual machine, but if you're not technical, the easier option is to destroy your droplet and create a new one through the link I just gave you
Awesome, I’m definitely going to follow these steps today and see how it goes. Looking forward to getting everything set up.
Great stuff, thanks for that David. Got stuck / errors a few times, but with some assistance from ChatGPT, figured it out 🙏
That's awesome to hear Stephen, hope you'll like the upcoming workflows I'll be sharing more about in February!
I appreciate the Make alternative! N8N FTW.
I'm getting more and more frustrated with Make and n8n is awesome. running it on your own server is just a cherry on top!
My take is a bit different after months of experimentation - while n8n is great, I wouldn't limit myself to just one tool.
Here's why: When I started automating my own business processes (and believe me, I tried everything), I discovered that the real magic happens when you mix different approaches based on your needs. Sometimes it's n8n, sometimes Make, and sometimes good old custom code D:
Instead of focusing on tools, I'd suggest starting with a simple question: What's the smallest thing you can automate today that would make your life easier? Then pick the tool that fits that specific need.
Quick tip: Start with a $4 droplet like they suggest, but keep some budget for experimenting with other tools too. You never know which one will click with your workflow!
I recently wrote about finding that automation sweet spot here: https://thoughts.jock.pl/p/automation-guide-2025-ten-rules-when-to-automate
Have you tried any automation tools yet? What's been your experience?
Fantastic thoughts Pawel, thank you for commenting.
I also agree. I don't use n8n exclusively. I sometimes code myself (especially with these new vibe coding tools it's super easy to build microservices for simple use cases) or just leverage someone else's open source repo for something. Sometimes I also use Make or Zapier, depending on the situation.
What my experience has been however is that most people getting started with no-code have near zero technical background.
They don't know how to think about problems like an engineer.
They don't know what is an API or a for loop or what data types are there.
Then it gets easy to confuse no-code with no-think and that's a good recipe for disappointment.
Most of the time sticking to one tool until you start hitting walls is a good idea. But if you're already technical enough to know all these things, then yes. You'll likely end up building your own preferred stack.
Hi David, thanks for posting. I'm (blindly ;-0) following the instructions and got as far as this step in the web console:
sudo docker volume create caddy_data
sudo docker volume create n8n_data
When i type them in (either one) the console says, "sudo: docker: command not found"
What can i do? Thanks, Andy
It's likely that Docker is not installed on your virtual machine.
Did you create the Droplet through this link?
https://marketplace.digitalocean.com/apps/docker
There's a button called "Create Docker Droplet".
If you just create a new Droplet from within Digital Ocean's platform, you won't have Docker on it.
You can check if docker is installed by typing in this command:
docker --version
if it returns some version number, you have Docker. if it doesn't understand what you want, you don't have Docker.
you can install Docker on your virtual machine, but if you're not technical, the easier option is to destroy your droplet and create a new one through the link I just gave you
Yep. This happened to me just now. :D