Why your website gets visits but no enquiries

A site with visits and no enquiries has one of two problems, and they need opposite fixes: the wrong people are arriving, or the right people arrive and leave. Conversion rate optimisation only helps with the second. So before you touch a headline or a button, split your numbers three ways and find which group is far weaker than the rest: by device, by language and by source.

How do you tell the wrong visitors from the ones you lose?

Compare your own groups against each other, because the biggest gap between two large groups points at the cause.

Ignore published average conversion rates. They blend industries, devices and traffic types, so they can't tell you whether yours is low. Your own groups, compared with each other, can.

Are you counting every enquiry, or only the form?

Count calls and WhatsApp taps as enquiries, because a site can look dead when it isn't if only form submissions are counted. DataReportal, citing GWI's Q2 2026 survey, reported that 84.1% of UAE internet users aged 16 and over said they'd used WhatsApp in the past month, ahead of YouTube, Instagram, TikTok and Facebook. It's a self-reported survey of internet users, not a count of customers, but it explains why so much intent never touches a form.

Add an event for every tap on the call button and the WhatsApp button, and the post on conversion tracking for calls, WhatsApp and forms shows how. Then ask whoever answers the phone and WhatsApp to log where each enquiry came from for a fortnight. A paper tally checks the analytics, and conversion tracking only counts what you've told it to count.

Does the first screen repeat what the visitor clicked?

Mismatch is the cheapest cause to check, and on paid traffic it costs the most because every click is paid for. Open the page from your own ad or search result on a phone and read the ad headline, then the page headline, with nothing else in view. If a stranger couldn't tell they belong together, rewrite the page headline to repeat the words the person searched.

For a service page, the first screen should carry the service, the area you cover, a price signal if you can honour one, and a single action: call, WhatsApp or book. A landing page built for one search does that job better than a home page built for everyone.

How fast does the page load on a phone, on mobile data?

Test on a phone on mobile data, because a laptop on office Wi-Fi is the kindest test you can run. Google's web.dev treats 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 as good, measured at the 75th percentile of page loads. PageSpeed Insights scores a page against those numbers, and the Core Web Vitals report in Search Console shows how real visitors fare once the site has enough traffic.

Look first at what loads on the first screen: a rotating banner that fetches several large images to show one, a font that arrives late, and chat or tag scripts that load before the content. Remove the banner first. Core Web Vitals for UAE sites gives the fix order.

Would a stranger believe you on the first screen?

Trust is whatever a stranger can check without calling you: the legal company name and trade licence number, a street address with a map pin and a landmark, named people with real photographs, reviews with dates, and a line saying what happens after they enquire.

A licence number and issuing authority in the footer is cheap proof that the business exists. Whether you must show it depends on your licence, so ask your licensing authority, and the pages a UAE business site needs covers where it goes. If your sector is regulated, such as healthcare or property, the authority sets what you may claim, so check before adding badges or promises.

Is the form asking for more than a first call needs?

Ask only for what you need to call back: a name, a mobile number and one line about what the visitor wants. Company size, budget, document uploads and a long message box can wait for the call, and each extra field gives a visitor another reason to stop.

Then check the UAE details. The phone field should accept 050 and +971 50 formats, the form should say when and how you'll reply, and the thank-you page should say what happens next, because a form sent on a Thursday evening or over a weekend raises the question of who answers and when.

Reply speed matters too. A 2011 Harvard Business Review study of US companies, quoted by SmartCompany, found that firms which tried to contact web leads within an hour were nearly seven times as likely to qualify them, meaning to have a real conversation with a decision maker, as firms that tried even an hour later. It's American, it's old and it shows an association rather than proof, but it's the best published case for a fast answer. Speed to lead is an operations problem that the website only exposes.

Where does WhatsApp sit on the page?

Give WhatsApp a button as visible as the call button, with a pre-filled message that names the page or service so whoever answers knows what the person came for. Write the link with the number in international format, with no plus sign, zero, brackets or dashes, and test it on an iPhone, an Android phone and a desktop.

A button nobody answers does harm, so decide who replies and in which hours, and say so beside the button. The lead follow-up process post sets out how to run that without a rota that depends on one person.

Does the Arabic page convert as well as the English one?

Compare the two by language in your analytics: if Arabic converts worse on similar traffic, the Arabic pages have a fault the English ones don't. Look at who wrote the Arabic, whether the layout was mirrored, whether the phone number reads in the right order and whether the form accepts an Arabic name. The right-to-left guide has a test pack for exactly this.

What do you change first, and how do you know it worked?

Fix the largest problem in the weakest large group, change one thing at a time, and compare four whole weeks before with four whole weeks after. Whole weeks matter because enquiries move through the week and UAE working patterns differ by company: The National reported in 2021 that the federal government works a half day on Fridays with a Saturday and Sunday weekend, while private companies set their own week.

Work out what a fix is worth before you pay for it. Say a site gets 1,000 visits a month, 10 become enquiries, you close 3 in 10, and each sale earns AED 6,000 in gross profit, all made up. On those made-up figures lifting enquiries from 10 to 12 would add AED 3,600 a month, so a fix costing AED 18,000 would take five months to pay for itself, which is why a site with low traffic needs cheap fixes or big ones.

With only a handful of enquiries a month you can't read an A/B test, because the result would be noise. Judge each fix by the weekly count of enquiries instead.

If your numbers can't be split into groups at all, work through the fixes in this order: counting, matching the page to the click, speed, proof, the form, WhatsApp and then Arabic. Each step costs less than a redesign, and each removes a problem that would hide the next.

If your analytics can't produce the comparisons in the table, get an outside read of the numbers before you pay for a redesign. Our Growth Audit Sprint is AED 18,000 excluding VAT.

Do this today: open your site on your phone, on mobile data, and try to enquire as a stranger would. Time it, and write down every point where you hesitated. That list is your first set of fixes.