Check the tablet row in the device report, see which pages tablet visitors use and whether they complete forms, then test those pages on a real tablet held both ways up, with the keyboard open.

The middle row

Universal Analytics sorts visits into three device categories: desktop, mobile and tablet. Most owners compare the first two and skip the third.

Tablet visitors are easy to overlook: there are usually fewer of them, and they fit neither picture. They have a computer's large screen and a phone's touchscreen, and often visit at different times from both.

On some sites they are a small, steady group who read more pages than anyone else and almost never send a form. That pattern usually points to a problem with the site rather than with the visitors.

The device report shows whether it is happening on yours.

Finding the tablet figures

In Analytics, go to Audience, then Mobile, then Overview. The table splits sessions, pages per session, bounce rate and goal completions by device category.

Audience, Mobile, Devices lists the actual models, phones and tablets together. On most sites the tablets are mainly iPads.

The built-in Tablet Traffic segment narrows any other report to tablet visits, so you can see which pages tablet visitors read and at what hour they arrive.

What tends to go wrong on a tablet

When tablet figures look odd, the cause is usually one of a handful of design habits.

The third is the one that stops enquiries outright. The visitor can read everything, but cannot see what they are typing or reach the send button.

Two orientations, two layouts

A common tablet held upright is 768 pixels wide as far as a website is concerned; turned sideways it is 1024. Many responsive designs switch layouts at exactly those widths.

So the same visitor can see the phone layout held one way and the desktop layout held the other, with a different menu, different image sizes and the form in a different place.

Both layouts need checking on the tablet itself. A desktop layout that works with a mouse can be awkward with a finger, even at the same width.

Forms with the keyboard open

With the tablet held sideways, the on-screen keyboard covers roughly half the screen. A fixed header across the top takes another slice, and the field being typed in can end up in a thin strip between them.

Small type in text fields causes a second problem. Safari on iPads can zoom in on a field whose text is under 16 pixels, leaving the visitor to pan around for the next field or the button.

Neither shows up on a computer, and neither shows up on a phone held upright. Only typing into the form on a tablet, both ways up, reveals them.

A worked example

An antiques dealer with a shop in an old bank building posts new arrivals on its website every week, each piece with several photographs and an enquiry form underneath.

The device report showed tablets as a small but steady share of visits, mostly in the evenings and mostly on the new arrivals pages. Tablet visitors viewed more pages per visit than anyone else, and sent almost no enquiries.

Held sideways, an iPad got the desktop layout, with the enquiry form in a narrow column beside the photographs. Once the keyboard opened, it and the fixed header left too little room to see the message field.

Held upright, the site switched to a layout whose menu opened on hover, so the first tap opened it and the second closed it again.

The developer moved the form below the photographs, stopped the header staying fixed on short screens, set the form text at 16 pixels and made the menu open and close on a tap.

The pieces, the photographs and the prices did not change. The form now works whichever way the tablet is held.

Menus and small targets

A finger has no hover. Menus that appear when a mouse passes over them behave unpredictably on touch: some open on the first tap and follow the link on the second, others open and close at once.

Make the main menu work on a tap, and give links and buttons enough space that a finger hits the one it meant.

Reading the numbers sensibly

Tablet figures are often small, and small numbers swing. Compare several months at once rather than reacting to one.

Look at the pages tablet visitors use most, not the site as a whole. Evening browsing through a catalogue or a gallery is a different job from a quick look for a phone number.

And compare goal completions per session with desktop rather than raw totals. A tablet group that reads a lot and completes almost nothing is the signal worth acting on.

Testing on a real tablet

A browser's tablet preview on a computer is a start, but it cannot show the keyboard, the zoom or how a finger meets a hover menu.

Borrow a tablet. Open the pages from the report, turn the tablet both ways, tap through the menu, and fill in every form to the end, including the send button.

Ten minutes of that finds more than a month of looking at charts.

The counter-case

Some sites get very few tablet visits, and for them this is a small check rather than a project. A business selling to trades on job sites will see far more phones than tablets.

A site built recently on a well-tested responsive theme may already behave well on tablets. The report will show that too.

But where tablet visitors read a lot and enquire rarely, test the forms on a tablet before concluding anything about the visitors.

The tablet check

  1. Open the device category report in Analytics.
  2. Note tablet sessions and goal completions.
  3. Apply the Tablet Traffic segment to top pages.
  4. Borrow a tablet for half an hour.
  5. Turn it both ways on each key page.
  6. Fill in every form with the keyboard open.
  7. Tap through the menu without a mouse.

Step six is the one that matters most. Reading a page on a tablet rarely reveals anything wrong; typing into one, with the keyboard covering half the screen, is where enquiries get lost.

Putting the tablet figures in front of an owner who will never open Analytics is covered in Data Studio reports for an owner who hates analytics.


Frequently asked questions

Where are tablet visits in Google Analytics?

Under Audience, Mobile, Overview, which splits sessions by device category: desktop, mobile and tablet. Audience, Mobile, Devices lists the actual models visitors used.

Why do tablet visitors behave differently?

They have a large screen like a computer but a touchscreen like a phone, and they often browse at different times, such as evenings, reading more pages per visit.

Why does my site change when the tablet turns?

A common tablet is 768 pixels wide upright and 1024 sideways, and many designs switch layouts at those widths. The same visitor can see two different sites.

What goes wrong with forms on tablets?

Held sideways, the keyboard covers about half the screen, fixed headers take more, and small text fields make the screen zoom. The send button can end up out of reach.

Is a preview in my browser enough?

It is a start, but it cannot show the on-screen keyboard, the zoom on small fields or how a finger meets a hover menu. Borrow a real tablet.

What tablet figures should worry me?

A tablet group that reads many pages but completes almost no goals. That pattern usually points to a form or menu problem on tablets, not to uninterested visitors.

West Coast Media Solutions Inc. provides web design, web development, hosting, digital marketing, and business consulting to organisations across Canada, drawing on more than twenty-five years in the field.

Tablet visitors who never send the form?

Tell us which pages they read. We will test them on a tablet both ways up and report what breaks.

Start a Conversation