Most advice about conversion rates jumps straight to fixes: move the button, shorten the form, add testimonials. Those fixes often help, and we covered the most common ones in our article on websites that get traffic but no leads. But for a Vancouver business that wants to improve its site steadily rather than guess, the work starts one step earlier, with measuring what is actually happening.

This guide is about that process: deciding what counts as a conversion, tracking it properly, finding where visitors drop off, and changing things in an order that tells you what worked.

Step 1: Decide what counts as a conversion

A conversion is any action that moves a visitor towards becoming a customer. For most service businesses, the important ones are:

  • Phone calls, including taps on a phone number on mobile
  • Contact or quote form submissions
  • Online bookings completed through your booking system
  • Chat conversations that turn into an enquiry
  • Purchases, for businesses that sell online

There are also smaller signals, such as visits to your pricing page, clicks on directions or a downloaded brochure. They are useful for understanding behaviour, but keep them separate from the actions that bring in work, or your conversion rate will look better than your business does.

Step 2: Track it properly

You cannot improve what you are not measuring, and many small business sites measure very little. The basics:

  1. Set up key events in Google Analytics 4. GA4 calls the actions you care about key events. Mark form submissions, booking confirmations and phone-number taps as key events, rather than relying on page views alone.
  2. Track phone calls. If a large share of your enquiries come by phone, a site that only tracks forms will badly undercount. At minimum, track taps on your phone number. Call tracking services can go further if calls are central to your business.
  3. Track bookings at the confirmation step. If your booking system runs on another domain, make sure the completed booking is recorded, not just the click to the booking page.
  4. Connect your ad platforms. If you run Google Ads or social campaigns, send the same key events to those platforms so each campaign is judged on enquiries, not clicks.
  5. Filter out the noise. Spam form submissions and test entries inflate the numbers. Add basic spam protection and exclude your own team's visits where you can.

Before changing anything on the site, give the tracking a few weeks to collect a baseline.

Step 3: Calculate it the useful way

The simple formula is conversions divided by visitors. A single site-wide number hides most of what you need to know, so break it down:

  • By page. Which service pages produce enquiries, and which get visits but no action?
  • By traffic source. Organic search, paid ads, social, referrals and direct visits behave very differently.
  • By device. If mobile visitors convert far less than desktop visitors, the mobile experience is the first place to look.

Be careful with small numbers. If a page had 40 visits and 2 enquiries last month, one extra enquiry changes the rate by a large amount. Look at longer periods before drawing conclusions about low-traffic pages.

Why industry benchmarks mislead

It is tempting to compare your rate with a published average for your industry. Those averages mix very different businesses, traffic sources and definitions of a conversion, so a "good" number for one site can be meaningless for another. A clinic that counts booked appointments and a retailer that counts newsletter signups are not measuring the same thing. The most useful benchmark is your own site, before and after a change.

Step 4: Find where visitors drop off

Map the path a visitor takes to become an enquiry. For a typical service business it looks something like this: arrive on a page, read about a service, go to the contact or booking page, start the form, submit it.

Then look at how many people make it through each step. GA4's exploration reports can show the paths people take, and a free tool such as Microsoft Clarity shows heatmaps and anonymized session recordings. The step with the biggest drop is usually where the most valuable fix is. A contact page that many people reach but few submit points to the form. Service pages that get visits but almost no clicks onward point to the content or the call to action.

Step 5: Prioritise, then change one thing at a time

You will usually find more problems than you can fix at once. Rank them by two questions: how many visitors does this affect, and how hard is it to fix? Start with high-impact, low-effort changes on your highest-traffic pages, and fix anything broken, such as forms that fail on mobile, before anything else.

Then change one significant thing at a time and give it enough time to show an effect. If you redesign the homepage, rewrite three service pages and change the form in the same week, you will not know which change helped or hurt. Allow for seasonality as well: a December dip in enquiries may have nothing to do with the change you made.

A/B testing, where two versions of a page run side by side, gives the clearest answer, but it needs enough traffic to reach a reliable result. For many small Vancouver businesses, careful before-and-after comparisons are the realistic option.

Common measurement mistakes

  • Counting thank-you page views without deduplicating, so refreshes and repeat visits inflate the number.
  • Ignoring phone calls, which can make a site look like it converts poorly when it does not.
  • Judging success by bounce rate, which counts someone who found your number and called you as a bounce.
  • Comparing periods of different lengths or seasons, and reading normal variation as a result.
  • Optimizing for a micro-conversion, such as clicks to a pricing page, while actual enquiries stay flat.

Where to start this week

  1. List the actions on your site that bring in real business.
  2. Check that each one is tracked as a key event in GA4, including phone taps.
  3. Look at last month's numbers by page, source and device.
  4. Find the step in your enquiry path with the biggest drop.
  5. Pick one change to make there, and note the date you made it.

If you would rather have someone do this with you, our conversion rate optimization service starts with exactly this kind of measurement, and tells you honestly whether the fix is a few changes or a rebuild.