Traffic But No Enquiries? Find the Leak in 20 Minutes (2026)

A site that gets traffic and no enquiries has one of four faults, and twenty minutes of deliberate looking is usually enough to work out which one. The visitors were never buyers. Nothing on the page earns enough trust to make contacting a stranger feel safe. The act of getting in touch is harder than it needs to be. Or the whole experience falls apart on a phone.

Those four are distinguishable, which matters more than it sounds, because the standard response to a quiet inbox is to buy more traffic. More traffic through a leaking site is just a more expensive leak. Working out where the leak actually is costs an afternoon and nothing else.

Before anything else, prove the enquiries aren't already happening

This is the least interesting possibility and it is worth eliminating first, because it is free to check and embarrassing to discover six months late.

Forms fail silently. The submit button fires, the visitor sees a thank-you message, and the notification email goes to a mailbox nobody has opened since the accountant set it up. Or it goes to spam, every time, because the form sends from a server the domain never authorised. Or the address it forwards to broke during a domain move and nobody noticed, because a broken forward doesn't announce itself. Or the phone number is baked into a header image, so on mobile it can't be tapped, only squinted at and retyped.

The test takes ten seconds. Pick up your phone, switch off wifi so you're on mobile data like a real visitor, go to your own site, and send yourself a genuine enquiry. Then ring the number on the site from a phone that isn't yours. If either one doesn't land where you expect, stop reading and fix that. You don't have a conversion problem, you have a plumbing problem.

Fault one: the traffic was never going to enquire

Traffic is not one thing. A thousand people arriving because they searched for what something costs are a completely different audience to a thousand people searching for someone in their suburb who can do the job this week, and a site converting the second group at 3% will convert the first group at nearly nothing. That is not a failure of the website. It's a mismatch between what the traffic wanted and what the page was built to do.

Google Search Console tells you which of the two you've got, and it is the single most useful free tool in this whole exercise. The Performance report shows clicks, impressions, click-through rate and average position, and lets you group the same data by queries, pages, countries and devices. Two minutes there answers the question. Sort by clicks, look at the queries, and look at which page they landed on.

If the queries are overwhelmingly informational — how much, what is, can you, versus — and the top landing page is a blog article rather than a service page, the site isn't broken. It's attracting people at the research stage and offering them nothing to do next. The fix is to build the commercial page that answers the buying-stage version of the same question, and to route the article's readers to it, rather than to redesign a homepage nobody is landing on.

The other version of this: check where the traffic comes from at all. A referral spike from a directory listing, a burst from a social post, or crawler noise inflating the numbers will all look like traffic in a dashboard and none of it behaves like a buyer.

Fault two: nothing on the page earns the enquiry

Assume the traffic is right. The next question is whether a stranger has enough reason to hand over their name and phone number to a business they found four minutes ago.

Reviews carry more of that weight than most business owners want to accept. BrightLocal's Local Consumer Review Survey, run in February 2026 across 1,002 US adults, found 97% read reviews for local businesses and 41% always do so when browsing, up from 29% the year before. It's US sampling, so treat the exact figures as directional rather than an Australian benchmark, but the direction is not in dispute: buyers verify before they enquire, and they verify off your site as well as on it.

The things that read as trust are unglamorous and mostly factual. Named people with real faces rather than a stock photo of a boardroom. A street address and a service area, so it's clear where you actually are. A review count with the score, near the point where someone decides. Licence, registration or membership numbers where the industry has them. And some indication of price. Silence on price is not neutral — it reads as either "expensive" or "we don't want to say", and a meaningful share of visitors will resolve that ambiguity by leaving rather than by asking.

Fault three: getting in touch is harder than it needs to be

This one is measurable and it's the fault most often responsible for a site that looks fine and performs badly. Every additional required field is a chance to lose someone who was already convinced. The contact form that asks for a company name, a budget range, a dropdown of how-did-you-hear-about-us and a message of at least fifty characters is a qualification exercise dressed up as an invitation, and it is doing its qualifying in the wrong direction. We've written the longer version of this argument in the three fields worth deleting from your contact form, including which forms genuinely justify more fields.

The other friction patterns worth checking on your own site: contact details that appear only on a contact page two clicks deep, a form as the sole channel with no phone or email alternative for people who won't use forms, validation errors that only appear after submit rather than as someone types, and a phone field that doesn't bring up the numeric keypad on a phone. None of these are visible to someone who already knows their way around the site. All of them are obvious to a first-time visitor with one thumb free.

Fault four: it falls apart on a phone

Speed and mobile layout are the same fault wearing different clothes, and Google publishes the thresholds. The Core Web Vitals targets, as documented on web.dev, are a Largest Contentful Paint within 2.5 seconds, an Interaction to Next Paint of 200 milliseconds or less, and a Cumulative Layout Shift of 0.1 or less, measured at the 75th percentile of page loads and segmented by mobile and desktop. That last part is the bit people skip. A site can pass comfortably on a desktop connection and fail badly on a mid-range Android over mobile data, which is the device a good share of enquiries would have come from.

Worth checking your actual split rather than assuming, though. StatCounter's platform share for Australia in July 2026 put desktop at 53.79%, mobile at 43.54% and tablet at 2.67% — so Australia is not the mobile-only market it's often described as, and the mix varies enormously by industry. Search Console's device tab will tell you which one your traffic is, and that determines which one to test first.

The twenty minutes

Here's the sequence, in order, with rough timings.

Minutes 0–5, Search Console. Open the Performance report, set the date range to the last three months, and read the top twenty queries and the top ten landing pages. Then switch to the device tab. You are looking for one thing: whether the people arriving are buyers, and on what.

Minutes 5–15, watch ten sessions. Install Microsoft Clarity, which is free with, in Microsoft's own words, "no limits on traffic", then wait a day or two for data and come back. Watch ten recordings of the page that gets the most traffic. Microsoft's own dashboard documentation describes what it surfaces: users clicking on things that aren't links, users scrolling up and down hunting for something they can't readily find, and JavaScript errors firing across sessions. Those three behaviours are the leak, rendered visible. Most business owners have never watched a single stranger use their website, and ten sessions changes more opinions than any report.

Minutes 15–20, be the customer. On your own phone, on mobile data, from the search result rather than a bookmark: find the business, decide it's credible, and complete an enquiry. Time it. If it takes more than about ninety seconds, or you had to zoom to read anything, or you knew where to click only because you built the site, you have your answer.

What the leak is actually worth

Some arithmetic, on invented but plausible inputs, because the abstract version never lands. A service business getting 1,000 sessions a month at a 0.8% enquiry rate collects eight enquiries. Move that to 2.5% — not a heroic number, closer to the middle of the range for a competent service-business site — and the same traffic produces twenty-five. If one in four enquiries becomes a job at an average value of $6,000 AUD, that's the difference between roughly $12,000 AUD and $37,500 AUD of work a month from traffic you were already paying for.

Those inputs are illustrative, not a promise. What's worth taking from them is the shape: the enquiry rate is a multiplier on everything upstream of it, which is why fixing it is almost always cheaper than buying more traffic. If you want a sense of what a realistic target looks like for your industry, we've collected the benchmarks in what counts as a good conversion rate for an Australian service business.

Where to start if you only do one thing

Watch the recordings. Search Console tells you who arrived, and the arithmetic tells you what it's worth, but neither shows you the moment someone gives up. Ten sessions on your busiest page will usually make the fault obvious enough that you stop needing to guess between the four.

For the half of this that's platform rather than behaviour, run the free audit — it checks Core Web Vitals and page speed, page weight, the certificate, your DNS and email records, the HTML structure and whether the site is crawlable, in under a minute. That won't tell you why a visitor hesitated at your contact form, but it will tell you within sixty seconds whether the page was slow enough that they never got that far.