Counting the enquiries you already get

There's a question we ask on every first call, and it's not a trick, although it lands like one.

How many enquiries did the website bring you last month?

The answer is almost never a number. It's "a fair few", or "it's been quiet", or, most often, a pause and then "I'd have to go back through my emails". Once in a while someone says twelve, and I write that down, because someone who knows it's twelve is already running a different business from someone who doesn't.

This isn't a criticism. Nobody sets out to not count. The counting just never gets set up, because the website launched, the enquiries started arriving in an inbox, and an inbox feels like it's already doing the job.

It isn't.

Why the inbox isn't a count

Your inbox holds form submissions mixed with supplier emails, spam that got through, replies on old threads, and the three messages you moved to a folder and forgot. Reconstructing a month from it takes an hour and produces a number you don't quite trust.

Worse, it's silently incomplete. It has nothing about the people who rang instead. Nothing about the ones who filled in half the form and left. And nothing at all about where any of them came from, which is the part that decides where your money goes next year.

The gap that does the most damage, though, is the one that looks like good news. A quiet month reads as a quiet market. Sometimes it's a quiet market. Sometimes the form has been failing since a plugin update in March and nobody noticed because a form that breaks doesn't send you an email saying so. We've written about that failure mode in the context of connecting your site to other tools, and it's the reason we argue for an unconditional plain-email copy of every submission going to a shared address, forever, no matter what else it's wired into.

The version that takes ten minutes

Before any analytics at all: a label in your email, a shared spreadsheet, or a note on your phone. One row per enquiry. Date, name, how they found you if you know, what happened in the end.

That's it. That's the whole system, and for a business getting fifteen enquiries a month it's honestly better than most analytics setups, because it captures the phone calls and it captures the outcome. A conversion you can't connect to a job is only half a number.

Do it for one month before you do anything clever. It will tell you things immediately. Usually that the volume is lower than it felt, or that a third of it comes from one source nobody was paying attention to.

Then make the website count itself

Once you've got a baseline, get the site counting form submissions on its own, because the manual version decays the moment things get busy.

In Google Analytics 4 the mechanism is a key event. Google's own explanation of key events is short: an event that measures something particularly important to your business, which you mark as a key event so it gets reported separately. Form submitted, quote requested, booking made. Mark it, and every report will now split by source, so you can finally see whether the enquiries came from search, from the map pack, from ads, or from the one Facebook post that did unexpectedly well.

Two traps here, and both are common enough that they're the first thing we check on an existing setup.

The first is counting the wrong moment. A lot of setups fire the event when someone reaches a thank-you page, or worse, when the submit button is clicked. Neither proves the message arrived anywhere. If the send fails, you get a conversion recorded and no enquiry, which is the worst of both worlds: a number that says everything is fine while the leads evaporate. Fire it on a confirmed send, and test it by submitting the form yourself and following the message all the way to the inbox.

The second is double counting. A page refresh, a back button, someone submitting twice because the first one seemed slow. Not fatal, but it inflates a small number a lot, and small numbers are what you've got.

The phone calls

Analytics will never see them, and for most Australian service businesses they're the majority of the enquiries.

You can solve this properly with call tracking, a separate number on the website that forwards to your real one, which gives you a clean count of web-driven calls. It works and it's not expensive, but it adds a number to maintain and a slightly odd number in your Google Business Profile if you're not careful about which one goes where.

Or you can do what I'd suggest to most small operators: ask. Two words at the end of the call, "how'd you find us", noted in the same spreadsheet. It's less rigorous and more honest than a tool that tracks the calls but can't tell you which ones turned into work.

What the number is actually for

Not for a dashboard. Nobody's life improved from a dashboard.

It's for three decisions. Whether the site is getting worse, which you can't know without a baseline. Where the enquiries come from, which tells you what to keep doing. And whether your problem is traffic or conversion, which are different problems with different fixes, and people spend money on the wrong one constantly.

It also settles the testing question. Below a few thousand visitors a month you can't run a meaningful A/B test, because the sample sizes don't arrive in any useful timeframe, which is the argument we made in full when you can't run a test. What replaces testing is counting plus conversation: the monthly number, and ringing the last five people who got in touch to ask what nearly stopped them. The five phone calls tend to produce better changes than the analytics do. I don't entirely know why. Something about hearing someone hesitate.

If the count says you have traffic and not enquiries, the leaks are usually in a short and predictable list, and we've got that list in improving your conversion rate.

This week

Send yourself a test enquiry through your own contact form, right now, and follow it to wherever it's meant to land. That's five minutes and it's the single highest-value thing in this post, because a meaningful share of the sites we audit have a form that isn't delivering to someone.

Then start the spreadsheet. One row, date and name. Next month you'll have a number, and every conversation about the website gets easier from there.

If you'd rather have the measurement set up properly and watched, that's part of how we handle analytics and conversion.