Why can website traffic rise without more leads?

Why visits and enquiries drift apart

A rise in visits feels like progress, and sometimes it is. But a visit only turns into a lead when several things go right in a row. The right person lands on a page that suits them. They find a quick way to get in touch. That contact gets recorded. And it reaches a business that can say yes. Break any link in that chain and the visit count can climb while the phone stays quiet.

For a local service business, the causes sort into four main groups. Intent: the new visitors wanted to learn something, not hire someone. Tracking: people are calling or sending forms, but the reports don't see it. Service fit: the pages attract searches for jobs or places you don't take. Capacity: enquiries arrive but get missed, answered late or turned away. Two smaller causes sit between them. One is friction on the page itself. The other is enquiries that moved to a different channel, such as the call button on your Google Business Profile.

The table below sets them out in the order we'd check them, cheapest first. Most of the checks use free Google tools you probably already have: Search Console for what people searched, Google Analytics 4 (GA4) for what they did on your site, and your Business Profile's Performance page for calls made straight from Google. None of them needs a developer, though fixing what they find sometimes does.

One warning before you start. Resist the urge to redesign. A new homepage changes dozens of things at once, so even if leads pick up afterwards you won't know why. Find the cause first, change the one thing it points to, then measure again.

Step one: find out which traffic actually grew

Total visits hide the story. Before you touch a single page, split the rise by where it came from and where it landed. Growth spread evenly across your service pages means something very different from growth on one old blog post, or from searchers in another country. Work through these steps with the same two date ranges each time, so every view compares like with like.

Two quirks are worth knowing while you read these reports. GA4 automatically excludes traffic from known bots and spiders. Google says you can't switch that off, and you can't see how much it removed, so a sudden spike from an odd source still deserves a look. And Search Console includes clicks and impressions from AI Overviews and AI Mode inside the ordinary Web search type, so at the time of writing (October 2026) they don't appear as a separate line. The guide to Google AI impressions explains how to read them.

Note the anonymised queries too. Search Console leaves some queries out of the query list to protect user privacy, though they still count in the chart totals unless you apply a query filter. So the rows in the Queries tab will never add up to the headline figure, and any view with a query filter applied will show fewer clicks than the unfiltered chart. That's normal, not a fault.

By the end of this step you should be able to finish one sentence: the extra visits mostly came from these searches, to these pages, from these places. If you can't, the rest of the checks will be guesswork.

1. In Search Console, open the Performance report and use the date filter's Compare option to set this period against the one before it, or against the same months last year. The table then shows a Difference column you can sort by. Only one comparison runs at a time.
2. On the Queries tab, add the Branded or Non-branded query filter where your property has it. Google says the filter isn't available for sites with a low number of impressions, and that some queries can land in the wrong group. Branded growth comes from people who already know your name. Non-branded growth is new reach, and that's where intent matters most.
3. Switch to the Pages tab with the same comparison to see which URLs gained the most clicks. Write down the top five.
4. Check the Countries tab. The country dimension groups data by where the search came from, so a jump from outside Australia shows up here straight away.
5. In GA4, open the Traffic acquisition report. Its default view groups sessions by channel. Add Session source / medium as a secondary dimension to see the exact sources behind each channel.
6. Open the Landing page report in GA4 (Reports, then Engagement, then Landing page) and read sessions beside key events for each page. A page with rising sessions and flat key events is the one to look at first.

Step two: check what the new visitors wanted

Once you know which pages grew, read the queries that sent people to them. The question is simple. Was this person trying to hire someone, or trying to learn something? Both are worth having. Only one of them tends to pick up the phone today.

Sort your top growing queries into the groups below. The examples are hypothetical and use an electrician, but the same split works for a plumber, a dentist or a landscaper. You don't need every query; the top twenty or so from the Pages you noted in step one will show the pattern.

A guide that starts ranking for a 'how do I' question can lift total visits a lot while bringing in very few customers. That's not a failure. Plenty of those readers will need you later, and some need you now but don't know it yet. The fix is a clear path from the guide to the service: a sentence that says when the job needs a licensed tradie, and a link to the page that sells it. Then judge guides on whether readers move on to a service page, and judge service pages on enquiries.

Google's own guidance on AI features adds a twist. Google says it has seen that clicks from results pages with AI Overviews are higher quality, with people more likely to spend more time on the site. Treat that as Google's observation, not your data. Your own key events by landing page will show whether it holds for your site.

Step three: test that every lead is being counted

This is the check people skip, and it often explains a report that says traffic is up and leads are down. If a form update broke the thank-you event, or the phone number in a new header is plain text instead of a tap-to-call link, the enquiries still come. The report just can't see them, and you end up fixing a page that was working fine.

In GA4, a key event is an event you've marked as important to the business, such as a form submission or a tap on your phone number. Marking an event only counts from that point forward; it doesn't change past data, and standard reports can take up to 24 hours to show it. A standard property can mark up to 30 events as key events, which is plenty for a local business. Run these tests whenever the site, the form tool or the tracking changes.

Keep personal details out of the tracking while you're in there. Google's Analytics policy says no data may be passed to Google that it could use or recognise as personally identifiable information, and lists email addresses, personal mobile numbers and names as examples. It warns that these often leak through page URLs and titles, such as a thank-you page address that carries the form fields. Send the service type, never the customer.

Then add the enquiries that never touch your website. Your Business Profile's Performance page counts calls as the number of times a customer clicked the call button on your profile, and website clicks as clicks on your website link. If profile calls rise while website forms fall, the leads may have moved rather than vanished. The guide to call tracking and conversion reconciliation covers how to line these counts up with Google Ads.

1. List every way a customer can reach you from the site: each phone link, each form, the email link, and any booking tool or chat.
2. Turn on debug mode for your own device with Google Tag Assistant, then open DebugView in GA4. Google says its Seconds stream shows the events logged in the last 60 seconds.
3. On your own phone, tap each phone link and send each form once, using an obvious test name. Watch for the matching event to appear.
4. Check each event is marked as a key event, and that one test submission fires one event, not two.
5. Confirm the form email lands in an inbox someone reads every working day, and that replying goes to the customer, not back to the website.
6. Compare a month of key events with your call log and inbox. They won't match exactly, because some people dial from memory or from your Business Profile, but a big gap points to a tracking fault.

Step four: check the traffic is for work you sell

Traffic is only useful if it's for jobs you want, in places you go. Older sites collect pages for services the business dropped years ago, suburbs it no longer travels to, or a one-off job type that happened to rank. Those pages can pull in visits all year while every enquiry they produce ends in a polite no.

Go through the pages you noted in step one and sort each into one of the rows below. Be honest about the last row. A job you can do but don't want more of still costs you time every time it rings.

This is also where your enquiries need sorting, not just your pages. A form from the wrong suburb or for the wrong service still registers as a key event. If you count every form as a lead, a rise in poor-fit enquiries can look like success in the report and feel like a waste of time on the tools. The trade guides on lead qualification, such as the one for electricians, set out how to mark each enquiry as qualified or not. The guide to connecting search data to sales quality then shows how to tie those marks back to the pages and searches that produced them.

Service fit also shapes what you say on the page. If you only travel thirty minutes from base, say so near the top. If you don't do small jobs, say that too. Clear limits turn away some visitors, and that's the point: the ones who stay are more likely to be work you want.

Step five: check you can take and answer the work

Some quiet months have nothing to do with the website. If the phone rings while you're up a ladder and nobody calls back until evening, the customer may have booked someone else by then. If you're booked out for weeks, people who need help this week go elsewhere even after a good first call. In both cases the site did its job. The lead was lost after it arrived.

These checks take an afternoon and need nothing more than your phone, your inbox and your calendar.

Capacity cuts the other way too. If you can't take more work, more traffic isn't the goal at all. Better-fit enquiries are. That might mean steering pages toward bigger jobs, the suburbs closest to you, or the services that pay well, and saying no to the rest before the phone rings. Setting that goal properly is covered in choosing SEO goals for a service business.

• Missed calls: check your phone's call log against the hours your website and listings say you're open.
• Form replies: note how long the oldest unanswered enquiry has waited, and name the person whose job it is to answer.
• Voicemail: ring your own number and listen. A full mailbox or an out-of-date message loses calls quietly.
• Quotes and bookings: compare enquiries with booked jobs. If enquiries held steady but bookings fell, the gap is in quoting or follow-up, not search.
• Hours on Google: keep your Business Profile hours current. Google lets you set special hours for holidays without changing your regular hours, and to mark the business temporarily closed if you'll be closed for 7 or more days in a row.
• Season: compare like months. A slow winter for an air conditioning installer isn't a website problem, and the guide to local SEO seasonality explains how to compare fairly.

Worked example: a Buderim electrician with a busy site and a quiet phone

The business below is made up to show the method. The suburb is real; the business, its pages and its reports are not, and nothing here is data.

An electrician based in Buderim notices that Search Console shows far more clicks this spring than last, while the jobs booked from the website look flat. The owner's first instinct is to redesign the homepage. Instead, they work through the checks in order.

Which traffic grew? With the date comparison on, the Pages tab shows most of the new clicks went to an old article about a safety switch that keeps tripping. The service pages barely moved. The Non-branded filter confirms the growth is new reach, not people looking up the business by name, and the Countries tab shows it's almost all Australian.

What did those visitors want? The queries are 'why does my safety switch keep tripping' and close variants. These are people trying to understand a fault, and some can sort it out by unplugging a faulty appliance. The article answers them well, but it never says when it's time to call an electrician, and it never links to the safety switch service page.

Is everything counted? In DebugView, the form sends its event. But tapping the phone number in the new mobile header does nothing in GA4, because it's plain text rather than a tap-to-call link. Mobile visitors have to copy the number by hand. Some ring anyway, and those calls are invisible in the reports. The owner fixes the link and marks the call tap as a key event, knowing it only counts from today.

Is the traffic for work they sell? Yes. Safety switch faults are jobs they want, and Buderim sits well inside their travel range. Can they answer it? The call log shows a run of missed calls in the middle of the day, while the owner is on site, and the voicemail box is full.

The plan that comes out of it is small and specific. Add a short section to the article on the signs a tripping switch needs an electrician, and link it to the service page. Fix the header phone link. Clear the voicemail and set a routine for returning midday calls by a set time. Then compare the next few months with the same months last year, using enquiries and booked jobs as the measure, not visits. No redesign needed yet.

If you'd rather run these checks with someone, Smart Local's local SEO work starts from the enquiries you want, not the visits. You can request a strategy review to talk through what your reports show. Nobody can promise more leads from a review, but you'll leave knowing which cause you're dealing with.