Lesson 8 – What Should You Automate First?

The question everyone asks

“What should I automate first?” It’s the most common question new clients ask me. And the answer is almost never what they expect. They usually want to automate everything at once – solve all their processes in one go. But that’s the fastest way to get overwhelmed, give up, and conclude that automation “isn’t for them.”

My answer is always the same: start with the process that hurts the most. The one you hate doing. The one that eats up the most time. The one that, if you could just make it disappear, would genuinely make your daily life easier.

The three-question test

Before you automate anything, ask yourself three simple questions:

Do I do this at least three times a week? Is it the same process every time – the same steps, the same sequence, the same logic? Can the whole thing be described as a set of rules, requiring no individual judgment calls?

If the answer to all three is yes, it’s worth automating. If any answer is no, it might still be worth automating, but it’s not where you should start.

A real client story: the nutrition consultant

One of my clients is a nutrition consultant. Before we worked together, their client intake process looked like this: send a paper-based assessment form (or a Word document) to new clients. Wait for them to fill it out. They either didn’t fill it out, didn’t fill it out thoroughly enough, or the whole thing dragged on for weeks. Then the consultant would manually read through the answers, figure out which services to recommend, write up a proposal, format it, and send it as a PDF.

The whole process was slow, inconsistent, and draining.

Here’s what we built in Make.com – six modules, nothing more:

We converted the paper assessment into a clean, simple online form that takes minutes to complete. When someone submits it, the data flows through a webhook into Make.com. OpenAI’s ChatGPT analyzes the responses based on the consultant’s criteria – the same criteria they used to evaluate clients manually, now encoded as a detailed prompt. The AI generates a structured output, which gets parsed through JSON and forwarded to a PDF generator that produces a beautiful, professional quote document. That PDF recommends specific services based on the client’s individual needs. And every submission gets logged in a Google Sheet automatically – no data is ever lost.

The result: the assessment period went from weeks to minutes. Client response time accelerated dramatically. The consultant stopped spending hours on manual evaluations and proposal writing. And the Google Sheet became a goldmine of data – showing how often the form gets filled out, what types of clients are coming in, and which services are most frequently recommended. Business intelligence, generated as a side effect of automation.

Start small, then grow

Don’t start with a 45-module sales funnel. I can teach you that too – but it’s not where you begin.

Start with something like an automated quote. A contact form response. An invoice archiver. A booking follow-up. Pick the one task that passes the three-question test and causes you the most friction. Build it. Learn from it. Keep refining the prompts, tweaking the logic, adjusting the output.

Over a few months, that first simple workflow will mature into something genuinely powerful. And by the time you understand the benefits it brings to your business, you’ll be ready for the next step – a more complex process, maybe even that big sales funnel. But you’ll build it with confidence, because you’ll have the foundation.

It’s worth starting with small steps. Start with those three questions.

This lesson is part of the free introductory module of the “Automate with Make.com” course. Start building: Make.com