Five numbers to check on your website before you spend another dollar on ads

The most common request we get from a growing service business is some version of “we need more leads, can you run ads for us?”

Sometimes the answer is yes. Often it isn’t—not because advertising doesn’t work, but because the site those ads would point at is already turning away most of the people who reach it. Paying for more visitors to a page that converts at 0.8% is an expensive way to discover you had a website problem.

Before you raise a budget, check these five numbers. All of them are available in tools you already have, and together they tell you whether you have a traffic problem or a conversion problem.

1. Your form conversion rate

Where to find it: in GA4, compare form submission events against sessions for the same period. If form submissions aren’t set up as an event, that’s the first thing to fix—and it’s worth knowing that a lot of sites have been quietly not recording them since their last redesign.

What’s healthy: for a regional service business with intent-driven traffic, one to three percent of all sessions is typical, and four to eight percent on a dedicated service page is good. Under one percent across the board usually means something is broken rather than underperforming.

What to do about it: a rate under one percent points at mechanics before design. Test the form on a phone. Count the fields. Check whether the submit button gives any feedback when pressed. We’ve found forms that threw a silent error for every visitor using one browser, for months, with nobody the wiser.

2. The gap between mobile and desktop

Where to find it: the same conversion rate, split by device category.

What’s healthy: mobile usually converts a bit lower than desktop. That’s normal. A gap of more than about half is not.

What to do about it: most service businesses now get 60 to 75% of their traffic from phones, so a big mobile gap means the majority of your visitors are having the worse experience. Usually it’s one of three things: a form that requires too much typing, a phone number that isn’t tappable, or a page that takes so long to load on cellular data that people give up before they see it. Open your site on your own phone, on data, not on office WiFi. That’s the version most of your customers get.

3. Time to first human contact

Where to find it: not in any analytics tool. Submit your own form and start a timer. Do it at 10 AM and again at 7 PM.

What’s healthy: a real reply within an hour during business hours, and an automatic acknowledgment immediately at any hour.

What to do about it: this is the number with the most upside on the whole list, because it’s usually the worst and it’s usually free to fix. Buyers of urgent services contact several companies at once and go with whoever answers first. If your average is four hours and your competitor’s is 20 minutes, no amount of ad spend closes that gap. We wrote about this in more detail in what your website should be doing while you’re on a job site.

4. Cost per closed job—not cost per lead

Where to find it: your ad platform gives you cost per lead. The number you want requires one more step: take the leads from a given channel over a quarter, ask your team how many became paying work, and divide the spend by that count.

What’s healthy: it depends entirely on your job size, which is the point. A $90 lead is expensive for a $400 service call and cheap for a $40,000 project.

What to do about it: nearly every business we’ve done this exercise with finds that one channel produces cheap leads that rarely close and another produces expensive leads that almost always do. The platform dashboards will never tell you this, because they can’t see what happened after the form was submitted. Once you know, the budget decision usually makes itself.

This is also the number that exposes a tracking gap. If nobody can tell you which leads came from where, that’s the real finding, and it’s worth fixing before the next campaign rather than after.

5. The pages people actually arrive on

Where to find it: GA4 landing pages report, sorted by sessions. Search Console will show you which queries brought them.

What’s healthy: there’s no benchmark here. The insight is in the comparison between where people land and where you’ve spent your attention.

What to do about it: most owners assume the homepage is the front door. Usually it’s third or fourth, behind one service page and something unexpected—a location page, an old blog post, a job listing. If 40% of your organic entrances land on a service page you haven’t touched in three years, that page deserves more of your next budget than a homepage redesign does.

Reading the five together

The pattern tells you where the money should go:

• Traffic is low, conversion is healthy. You have a visibility problem. Ads or SEO are the right answer, and you’ll get a return on them.
• Traffic is fine, conversion is under one percent. You have a website problem. More traffic makes it more expensive, not better.
• Both look fine, but the phone isn’t ringing more. You likely have a measurement problem, and the leads are arriving somewhere nobody is counting.
• Conversion is fine, cost per closed job is terrible. You have a targeting problem. The site is doing its job; you’re buying the wrong visitors.

The trap worth naming

Don’t optimize a number you can’t connect to revenue. Bounce rate, time on page, impressions, and follower counts all move around and none of them pay anyone’s salary. If you can’t draw a line from a metric to a job on the schedule, it’s context at best.

Pick the two numbers on this list that connect most directly to closed work, check them monthly, and make budget decisions on those.

If you’d like us to run these five for your site and tell you what we see, we’re glad to. It takes us about an hour and you’ll get the findings whether or not you work with us.