Connect your real apps.

Time to build. We'll make the classic first automation: a new enquiry comes in, and it lands in your CRM, your inbox and a spreadsheet without you touching a thing. Four apps, one scenario. Either Make or Zapier works fine here, and if you're still weighing them up our guide comparing n8n, Zapier and Make lays out the trade-offs. The two skills this lesson teaches, connecting an app and mapping its data, are the ones you'll use in every flow you ever build, so it's worth doing slowly the first time.

What a connection actually is

Before a tool like Make can read your Gmail or write to your CRM, you have to let it. A connection is that authorised link: you sign in to the app once, grant access, and from then on the scenario can act on your behalf without you logging in each run. You'll set up one connection per app: one for the form, one for Gmail, one for Google Sheets, one for the CRM. It takes a minute each and you only do it once. A sensible habit: connect the account you actually want it acting as, not your personal login, so the trail is clean.

Pick the four apps for this flow

We'll keep them ordinary, the everyday tools a small business already runs:

• A form as the trigger. A Google Form, a Typeform, or your website's contact form. When someone submits it, the scenario fires.
• A CRM to hold the lead. HubSpot has a generous free tier that's great to learn on; Pipedrive and others work the same way.
• Gmail to send an acknowledgement back to the person.
• Google Sheets as a simple running log, so you can eyeball every enquiry in one place.

If your real CRM isn't connectable yet, build the flow with a spreadsheet standing in for it. The shape is identical, and you'll swap the real app in later.

Mapping: the skill that unlocks everything

This is the heart of it. When the form trigger fires, it hands the scenario a bundle of fields: name, email, message, maybe a phone number. Mapping is telling each action which of those fields goes where. In your "create CRM contact" step, you drop the form's name into the CRM's name box and the form's email into the CRM's email box. In your "send email" step, you map the form's email into the "to" field so the reply reaches the right person. In your "add a row" step, you map every field into its column.

In Make this is delightfully visual: you click into a field and pick the piece of data from a list of everything the steps above produced. Zapier does the same with its "insert data" picker. You are never typing the customer's details. You're wiring a pipe so that whatever comes in through the form flows to the right place automatically. Get the mapping right once and it works for every enquiry after.

Watch out for the small stuff

A couple of things trip people up the first time, and both are easy:

• Required fields. If your CRM insists on an email and the form lets it be blank, the step can fail. Make the field required on the form, or add a filter that only continues when an email is present.
• Matching formats. A date or a dropdown sometimes needs to be in a particular shape for the next app to accept it. If a step errors, the format is the usual suspect.

Test with one safe record

Never switch a new flow on for the world and hope. Send one test enquiry through it, using a test name and your own email, then follow that single record all the way down. Did the contact appear in the CRM with the right name and email? Did the acknowledgement land in your inbox? Did a tidy row show up in the sheet? Both Make and Zapier let you run the scenario once on demand and show you exactly what each step did, with the real data, so you can see where anything went sideways. If a field landed in the wrong box, fix the mapping and run it again. Two or three test runs and you'll trust it. That five-minute habit is the difference between an automation that quietly works and one that quietly mangles every lead.

Once your test record sails through cleanly, you've got a real, working lead-to-CRM flow. Turn it on and the next genuine enquiry will run the whole chain by itself. Next, we make it clever: an AI step that reads the enquiry and sorts it for you.