Most of My Visitors Are on Phones. What Should I Actually Change?

There is a panicked version of this question, usually asked after someone's nephew has looked at the website on his phone and pulled a face. And there is a calmer version, which is the one worth answering: most of my visitors are on phones, so where do I spend the two hours I have this month?

The honest answer is that four things account for most of the difference on a small business site, and they are not the things web designers get excited about. Whether your phone number dials when someone taps it. How many boxes your enquiry form asks people to fill in. Whether your text is readable without pinching. How heavy your images are. Roughly in that order. Everything else — animations, app-style menus, clever scroll effects — is real, but smaller, and usually costs more to change.

"Most of my visitors are on phones" is usually true, but it is worth knowing how true, and where. Analytics tools record device type — simply whether a visitor arrived on a phone, a tablet or a desktop computer — and you can read it as a percentage of visits. That takes about thirty seconds to check and it stops you rebuilding something that was already fine.

The site-wide average hides the interesting part, though. The split changes page by page and source by source. A bakery might see the vast majority of visits to the opening hours page come from phones, while the wholesale enquiry page skews to desktop because caterers look at it from an office. A solicitor might find that people arriving from a Google search are overwhelmingly on phones, while people arriving from a directory listing are on laptops. Same site, two different jobs to do.

So look at the split on the three or four pages that actually matter — contact, prices, booking, whatever your equivalent is — rather than the whole site. If you want the detail of how the phone-versus-desktop breakdown is worked out and where it can be fuzzy, the device type documentation covers it. Then go and fix the pages where the phone share is highest and the stakes are highest.

This is first because it is the cheapest change on the list and the one most likely to have been missed. On a lot of small business sites the phone number is typed out as plain text. On a laptop that is fine — people read it and pick up the handset. On a phone, tapping it does nothing at all. The visitor has to memorise eleven digits, close the browser, open the keypad and type them in. Some do. Some don't.

The fix is to turn the number into a link that dials. Any web designer will know how to do this in under a minute; on Squarespace, Wix, WordPress and Shopify there is usually a button in the editor for it. Ask for it by name: "make the phone number a tap-to-call link".

While you are there, ask where the number sits. On a plumber's or solicitor's site the number should be visible without scrolling, at the top of every page, on a phone. Not in the footer. Not only on the contact page. If your work is urgent by nature — a leak, a locked-out tenant, a legal deadline — a sticky bar pinned to the bottom of the screen with "Call" and "WhatsApp" is worth more than anything else on the page.

Then measure it. A tap on a phone number can be recorded as a goal, so you can see how many people did it rather than guessing. The conversion documentation explains how to set that up as a countable action, which matters for the next section too.

Every box on a form has a cost, and on a phone that cost roughly doubles. Typing on glass is slower, the keyboard covers half the screen, and dropdown menus are fiddly. A form that feels reasonable on a desktop — name, email, phone, company, postcode, how did you hear about us, service required, preferred date, message — is a genuine chore on a phone held one-handed on a bus.

Go through your enquiry form field by field and ask a blunt question of each one: if I did not have this, could I still ring the person back and find out? For most small businesses the answer is yes for nearly everything. A freelance designer needs a name, a way to reply, and a rough idea of the job. That's three fields. "How did you hear about us" is a question your analytics already answers better than the customer can.

In general experience across small business sites, shortening a form does more than rewriting the button text or changing its colour, and it is a change you can make yourself in an afternoon. I can't tell you the size of the lift on your site — nobody honestly can without testing it — but the direction is consistent enough that it's worth doing before anything cleverer.

One more thing: check that your form uses the right keyboard. A phone number field should bring up the number pad, an email field should bring up the keyboard with the @ sign. If a customer has to hunt through symbol menus to type their own email address, that is a fixable irritation.

Websites are usually designed on large monitors by people with good eyesight, and body text gets set smaller than it should be. Text size is measured in pixels — the tiny dots that make up a screen. As a rule of thumb, body text on a phone wants to be at least 16 pixels, and 17 or 18 is kinder. If you have to squint or pinch to read your own prices, so does everyone else.

Two related things matter as much as the number. First, contrast: light grey text on a white background looks elegant in a design mockup and is very hard to read outdoors in daylight, which is where a fair amount of phone browsing happens. Darker text on plain white is less fashionable and more useful. Second, tap targets. Buttons and links need to be big enough for a thumb, with space around them. Two links stacked a few pixels apart will get mis-tapped, and a mis-tap usually ends in a back button.

Test it the honest way: take your own phone outside, in daylight, and try to find your opening hours in under ten seconds. That test finds more problems than most audits.

A photo taken on a modern phone can be several megabytes — a megabyte being a rough measure of file size, where a page that loads comfortably over a patchy mobile signal generally wants to weigh under a couple of them in total. Upload eight of those photos to a gallery page and you have a page that takes an uncomfortable number of seconds to appear on a train, on a phone, on three bars of signal. People leave before they see anything.

The fixes are unglamorous and permanent. Resize images to the width they are actually displayed at rather than uploading the original. Save them in a modern format — WebP is the common one, and most website builders will convert for you if you ask. Turn on lazy loading, which means images further down the page only load when someone scrolls to them. And be ruthless about the big background video on the homepage, which is almost always the single heaviest thing on a small business site and almost never the reason anyone got in touch.

Speed is worth tracking over time rather than checking once, because it drifts — a new plugin here, an untouched photo there. Page speed tracking records how long your pages take to load so you notice the drift. If you would rather not go hunting page by page, run your site through the report card and it will tell you which pages are heaviest and what is making them that way, in plain language.

A few things come up constantly in mobile advice and are worth less than the four above. Hiding your navigation behind a hamburger menu — the three-line icon — is fine for secondary pages, but do not bury "Book" or "Call" or "Prices" in there. If it is the main thing you want people to do, it belongs on the screen as a button.

Chat widgets are a mixed blessing on phones. They eat screen space, they often cover the very button you want tapped, and unless someone is genuinely there to answer, they set an expectation you can't meet. If you run one, look at it on a small phone and check what it is sitting on top of.

Carousels — those rotating banners on homepages — perform poorly almost everywhere and worse on phones, because people scroll past before the second slide appears. And passing Google's mobile-friendly check does not mean your site is good on a phone. It means the text scales and nothing overflows sideways. It is a floor, not a standard.

Compare like with like. The number to watch is not total visits but the share of phone visitors who do the thing you care about — call, book, submit the form — before and after the change, ideally next to the same figure for desktop visitors as a control. If phone enquiries improve and desktop stays flat, that is a reasonable sign the change did something.

Be careful with bounce rate, the proportion of people who leave without going any further. It reads worse on phones almost everywhere, partly for real reasons and partly because a phone visitor who reads your opening hours and rings you has technically bounced while also becoming a customer. The bounce rate documentation sets out where it is useful and where it misleads.

And give it time. A small business site might get a few dozen phone visitors to the contact page in a week. That is not enough to tell a real improvement from a quiet Tuesday. Look at four weeks against the previous four, note what you changed and when, and resist reading meaning into a single day.

If you do nothing else, do these in this sequence. Each one is finishable, and each one is checkable afterwards.