Google Analytics 4, usually called GA4, is free and already installed on most business websites. Most owners I talk to have logged in once, stared at a wall of charts and never gone back. That's a shame, because a properly set up GA4 can answer a handful of questions that directly affect how you spend money.
GA4 should answer five business questions: where your leads come from, which pages turn visitors into leads, where people give up, whether calls or forms matter more, and whether your last change helped. If your setup can't answer those, the fix is to track fewer, better things and test them, not to add more reports.
The usual story goes like this. Someone installs GA4, turns on the defaults, maybe adds a few events because a plugin offered them, and walks away. An event is just a record that something happened on your site, like a button tap or a form submission. The account fills up with data, and none of it lines up with a decision you actually need to make.
I do it the other way around. Start with the questions, then track only what those questions need. Here are the five I'd want answered for almost any service business in Houston.
Question 1: Where do my leads come from?
Traffic is easy to count. Leads are what pay the bills. GA4 can show you leads by source, meaning Google search, Google Ads, your Google Business Profile, social, email or referral sites, but only if a lead is recorded as a lead.
That means a specific event fires when a form is actually submitted successfully. Google even suggests a standard name for it, generate_lead. Mark it as a "key event," which is GA4's name for the actions that matter most. Keep that list short. If everything is a key event, the label means nothing.
One honest limit: GA4 can tell you where form fills came from. It can't tell you which of them became paying customers. For that, you need the lead source carried into whatever you use to track customers, which I cover in tracking leads when the sale happens by phone.
Question 2: Which pages actually turn visitors into leads?
Some pages attract a lot of visits and produce nothing. Others get modest traffic and generate most of your inquiries. You want to know which is which before you spend money on a redesign or new content.
In GA4, look at leads by landing page, meaning the first page someone saw on the visit. A quiet service page that produces steady inquiries deserves more attention than a popular blog post that never produces one. That kind of comparison tells you where to put your effort.
Question 3: Where do people give up?
If a lot of people start your contact form and few finish it, the form is the problem. Too many fields, a broken step on phones, a confusing error message. You'll only see this if you track two moments: when someone starts the form and when they finish it.
A form_start event plus your generate_lead event gives you a simple drop off rate. If it's bad, fix the form before you buy more traffic. Sending more people into a leaky form just burns ad money faster. This is a big part of our forms and tracking work.
Question 4: Are calls or forms the bigger channel?
Plenty of Houston customers would rather call than type. If you only track forms, you're blind to a big share of your leads. At minimum, track taps on your phone number link on mobile. That's a phone_click event, and it tells you someone tried to call.
A tap isn't a completed call, though. For real call data by source, you'll want call tracking, which I explain in call tracking without hurting your local SEO. Once both are in place, you can see whether your ads mostly drive calls while organic search mostly drives forms, and adjust.
Question 5: Did the change we made help or hurt?
You changed the homepage headline. You moved the form higher. You launched a new site. Did it work? Without clean tracking, you're relying on gut feel and whoever complains loudest.
With the right events in place, you can compare the weeks before and after a change. That comparison isn't a lab experiment. Seasons, weather and ad spend all move the numbers. But it's far better than guessing, and it keeps you from undoing a good change because one slow week spooked you.
A checklist for a GA4 setup you can trust
Here's what I look for when I audit a business's analytics.
- A
generate_leadevent that fires only after a successful submission, never on a failed one. - A
form_startevent on every lead form. - A
phone_clickevent on every tap to call link. - Only two or three key events. Leads and calls, usually.
- Google Tag Manager in charge of the tracking, with tags named clearly enough that the next person can understand them. Tag Manager is a free Google tool that lets you add and change tracking without editing your site's code every time.
- Every change tested in Tag Manager's Preview mode and GA4's DebugView before and after publishing.
- A cookie consent setup that actually holds tags back until a visitor chooses, if your situation requires consent. That's a legal call, so get advice.
- One simple weekly view someone on your team actually opens.
That last point matters more than it sounds. A beautiful dashboard nobody reads is decoration. One short weekly view showing leads by source, leads by page and form drop off will change more decisions than a dozen polished reports.
Get the answers, skip the noise
You don't need to become an analytics person. You need a setup built around your questions, tested once properly, and reviewed on a schedule. If you want the more technical walkthrough of naming rules, consent settings and Tag Manager housekeeping, I wrote the longer version on my site.
We build tracking this way on every site we launch, and we can fix an existing setup as part of a redesign or on its own. If you're not sure your numbers can be trusted, start with the free Lead Leak Check. We test your forms, phone speed, tracking, Google basics and what AI answers say about you.
Adapted for business owners from an essay by Hector Herrera.



