How to automate tasks without code using AI

Every automation, however complex it looks, is built from three pieces:

1. The trigger: the event that starts it. Example: “a new email arrives with the word invoice”.
2. The AI action: the step that thinks. Example: “the AI reads the email and pulls out the amount and the date”.
3. The result: where the information ends up. Example: “add a row to my spreadsheet”.

Seen this way, automating stops being scary: you’re just saying when this happens, do that.

1. Go to https://zapier.com or https://make.com and create a free account (you can use your Google email).
2. Zapier is simpler and connects with over 6,000 apps: good to start. Make is more visual and its free plan goes further.
3. You don’t have to decide forever: the concepts are identical on both.

The free plan is more than enough for your first automations. You only pay once you use it every day.

1. Create a new flow (in Zapier it’s called a Zap, in Make a Scenario).
2. Pick the trigger app and connect it (it will ask you to sign in to, say, Gmail).
3. Choose the exact event: “new email”, “new row in the sheet”, “form submitted”.
4. The tool lets you test the trigger with real data to confirm it’s picking up the information correctly.

Start with something small and concrete: the task you repeat most and enjoy least.

Here’s the magic. Add an AI step so it understands the information, not just moves it:

1. Add an action and search for ChatGPT or Claude (or the “AI by Zapier” block).
2. Connect it with your account (or your API key if it asks).
3. Write a clear instruction using the trigger’s data. For example:

Read this email and return only the invoice amount and the due date, in this format: amount | date.

4. Test the step: you’ll see the AI’s response with real data.

The AI can summarize, classify (“is this a complaint or a question?”), draft a reply or extract data. That’s the difference between moving information and understanding it.

1. Add the final action: where the result goes. For example, “add a row in Google Sheets” or “send an email”.
2. Connect that app and drag the data the AI extracted into the right fields.
3. Run a full end-to-end test and check the result comes out as you expect.
4. When it all fits, turn on the flow. From then on it works on its own, without you touching anything.

Tip: leave the flow running for a few days and review it. Tweaking the AI instruction once or twice is normal.

If you don’t know where to start, steal one of these:

• Emails to sheet: every invoice that arrives, the AI pulls the amount and date and saves it in a spreadsheet.
• Tidy inbox: the AI classifies each incoming email (urgent, client, spam) and adds a label.
• Quick replies: when a common question comes in, the AI drafts a reply and leaves it ready for you to review.
• Daily summary: every afternoon, the AI summarizes the day’s messages and sends you an email with the important bits.

When you’re ready for the next step, check out How to create your first AI agent without coding and How to automate marketing reports with ChatGPT.

Problem	Fix
The flow doesn’t start	Check the trigger is well connected and the event is the right one
The AI returns something odd	Make your instruction more specific and tell it the exact format you want
I ran out of free tasks	Combine several steps into one flow or switch to Make (larger free plan)
Data lands in the wrong column	Go back to the last step and drag each field into place again