.87a004c7.png&w=3840&q=75)
The first thing I ever automated was embarrassingly simple — and it started with money I didn't know I was losing.
I was spending on AI tools and subscriptions across multiple services. The costs were real but scattered: invoice emails arriving at different times, different amounts, different billing cycles. Nothing was in one place. I had no clear picture of what I was actually spending until I sat down to check — and by then, enough time had passed that the number was always a surprise.
So I built a system. It read incoming invoice emails, extracted the relevant details, and updated a spreadsheet automatically. The spreadsheet did the cost analysis. I could see everything in one view — what I was spending, on what, and where the money was quietly accumulating.
No traditional code. Mostly drag, connect, test, break, fix, and run again. The first version was built in an afternoon — not perfect, but useful enough to change how I looked at the rest of my work.
At first, I thought I had automated a spreadsheet update. What I had really removed was a recurring decision: check it now, leave it for later, or keep carrying it in the back of my mind. The task itself wasn't the cost. The mental residue was.
Most People Don't Have an Automation Problem
They have an observation problem.
The tools are genuinely ready. No-code platforms have made workflow automation accessible to anyone willing to spend an afternoon building something. AI has made the logic layer even simpler — you can describe what you want in plain language and have something working in hours. The barrier is lower than it has ever been.
And yet most people automate nothing — or automate the wrong things first.
The reason is not technical. It is that most people move through the same repeated tasks every week without stopping long enough to notice the pattern. The task is so routine it becomes invisible. It does not feel like a problem worth solving because it does not feel like a problem at all. It just feels like work.
Insight
The first skill in automation is not building workflows. It is developing enough operational self-awareness to notice what you are doing on repeat.
The Five-Minute Rule
Look at the last two weeks of your working life. What did you do more than three times that followed the same steps every time?
Not the big, complex processes. Not the strategic work. The small, repeated, slightly boring tasks that you do without thinking — because they are so routine they don't feel worth solving.
That is your starting point.
A good first automation should be:
- Repeated often — at least three or four times a week, ideally daily
- Rule-based — if X happens, do Y. No judgment calls required.
- Low-risk — if it breaks, nothing critical fails
- Clear enough to describe — you can explain it in two sentences to someone who has never seen it
- Annoying enough to matter — it interrupts your flow even when it only takes minutes
The reason most people start with the wrong thing is that they want their first automation to be impressive. They want to automate something complex, something visible. That instinct is exactly backwards.
The best first automation is the one nobody notices — because it just quietly disappears from your day.
One Warning Before You Build
Automation does not fix unclear processes. It amplifies them.
If the task is ambiguous when you do it manually — if you have to make judgment calls each time, if the steps change depending on context, if the output varies based on something unstated — do not automate it yet. A bad process automated is just a bad process running faster and failing in ways that are harder to see.
Before you build anything, clean the process first. Decide what should trigger it, what information matters, what the output should be, and when a human needs to step in. If you cannot describe the workflow clearly in plain language, you are not ready to automate it yet.
Build the Ugly First Version
Once you have identified the task and described it clearly, the build is more straightforward than you expect — but messier than the tutorials suggest.
Before I opened any tool, I had to admit something: I did not have an automation problem. I had an unclear-process problem. The subscription tracker only worked once I had defined exactly what an invoice email looked like, what fields mattered, and what the spreadsheet needed to show. The automation was the last step, not the first.
A few things that actually helped:
Pick one tool and stay there. For a first automation, the tool matters less than the habit you are building — learning to see work as triggers, steps, conditions, outputs, and exceptions. n8n, Make, and Power Automate all work. Pick the one your environment supports and build.
Use AI to close the gaps. You will get stuck. A connection that does not work as expected, a field that does not map correctly, a logic step that is not obvious. Describe the specific problem to an AI assistant and ask for the precise step. You do not need to understand the tool comprehensively. You need to understand enough to build what you are building right now.
Build the simplest version first. The first version should do exactly one thing. Not the full workflow you eventually want. One trigger, one action, one output. Get that working and watch it run a few times before adding anything else. Every layer of complexity you add before the foundation is stable is a layer that can fail without warning.
Expect to break things. The first run will probably fail. The second version will be better. The third will be the one you actually use. That is not a problem — that is how useful systems get built.
Tip
A good first automation should feel almost too small to matter. That is usually why it works.
What You Are Actually Removing
When the subscription tracker worked for the first time, I did not save five minutes. I removed a decision from my day.
The task itself was not the cost. The cost was the mental interruption — the context switch, the moment of "I should check that before I forget," the low-grade awareness that something routine was waiting. Multiply that across every small repeated task in a week, and you are not losing hours. You are losing focus in small increments that compound into something significant.
That is what automation actually does at the personal level. Not dramatic efficiency. Quieter days. More mental space for the work that actually requires judgment.
The point is not to remove yourself from the work. The point is to remove the repeated decisions that no longer need you.
Closing
Find the five minutes you repeat every day.
Not because five minutes will change your life.
Because once a system handles it for you, you start seeing the rest of your work differently.
Resources to Start With
If you want to go deeper, these three YouTube channels are the most practical free education currently available on workflow automation and AI agents — filtered for people who actually build rather than just explain:
Nate Herk — n8n specialist, non-technical background. Step-by-step with no skipped steps. Start here if you want to learn n8n from someone who learned it the same way you will.
David Ondrej — Full builds from start to finish in 20–40 minute videos. Good for seeing how real workflows are structured before you try to build your own.
Jono Catliff — Business operator turned automator. Blueprint-style walkthroughs: here is the problem, here is the workflow, here is how to build it.
One honest note: passive watching does not build automation skills. Watch one video, then build the thing the video described. That cycle will teach you more in a week than a month of courses.
Written from experience, not expertise. More on LinkedIn.