Compare phone and desktop visitors separately in Analytics before judging either. Phone visitors often look, leave and call, so a low mobile conversion rate frequently means calls are not being counted, not that the mobile site fails.

Two kinds of visit in one average

Open Analytics and the site has one conversion rate, one bounce rate and one average visit length. Each is an average of two very different kinds of visit.

Somebody at a desk reads three pages, compares prices and fills in the quote form. Somebody on a phone, in a car park or a kitchen, checks one thing and taps the number.

Averaged together, the desk visitor makes the site look persuasive and the phone visitor makes it look like a failure. Neither impression is right.

Analytics can separate them in a couple of clicks. Most owners never ask it to.

Where to find the split

Universal Analytics records the kind of device behind every visit in one of three categories: desktop, mobile and tablet.

The Mobile Overview report, under Audience, lists them side by side, with visits, bounce rate and conversions for each. In almost any other report, Device Category can be added as a secondary dimension.

The built-in segments do the same for the whole site. Applying Mobile Traffic and Tablet and Desktop Traffic together puts both audiences on every graph.

What usually differs

The last distorts everything else. If the site counts forms and not calls, the main way phone visitors get in touch is invisible, and their conversion rate is understated by design.

A bounce that worked

Analytics calls a visit a bounce when it views one page and leaves. On a desktop, that often means the visitor gave up.

On a phone, the same pattern often means the page did its job. The visitor wanted the number, the hours or the address, found it and acted, by calling or driving over.

Analytics cannot see a phone call or a drive across town. Unless something records the tap that started it, the most successful visit looks exactly like the least successful one.

Counting the taps

The fix is to record the actions phone visitors take in place of forms.

A tap on a phone number written as a link, so that tapping it starts a call, can be recorded as an event through Google Tag Manager, and the event made into a goal.

The same works for taps on a directions link or an email address. Each is a visitor deciding to contact you or come over.

A tap is not quite a call, since some people hang up before it connects. It is still far better than nothing, and it sits in the same reports as the forms.

Calls the website never sees

Some phone visitors dial by hand, reading the number off the screen or saving it for later. No tap is recorded.

Call tracking closes most of that gap. The site shows a tracking number that forwards to your line, and the tracking service reports which visits led to calls.

It needs care, because the number on the site then differs from the one on your listing and your vans. Many setups show tracking numbers only to visitors from particular sources.

A worked example

A tire shop with three locations had a site where customers could look up tires by size and ask for a quote, plus a page for each location with its hours, address and phone number.

The owner noticed that the conversion rate on phones was a small fraction of the desktop rate, and asked for a quote on redesigning the mobile site.

Split by device, the picture changed. Phone visitors mostly landed on the location pages from searches for tire shops nearby, stayed briefly, and left.

Those were the pages people called from, and nothing counted it. Only the quote form was a goal, and phone visitors were not using the form.

Taps on each location's phone number and directions link were set up as goals through Tag Manager. Within weeks, the location pages showed as the site's main source of contact from phones.

The redesign was narrowed to the one weakness the split had exposed: the tire size search, which was awkward with a thumb.

What desktop visitors reveal

The split also shows how desktop visitors behave once the phone visits stop diluting the figures.

A desktop conversion rate lower than expected, on pages people read at length, points at the form, the price information or the end of the page. The combined average had been hiding those problems.

Tablets usually sit somewhere between the two, and most small business sites have too few tablet visits to read much into.

One person, two devices

Many people research on a phone and act later on a computer, or the other way round. Analytics counts them as two separate visitors, because each device carries its own cookie.

So the phone visit that started the decision gets no credit when the form arrives from an office computer the next morning.

There is no simple fix for a small site. Knowing it happens is enough to stop anybody concluding that the phone visits were wasted.

The counter-case

Sometimes a low mobile conversion rate is exactly what it looks like.

A quote form that is painful with a thumb, a phone number shown as an image, or a page that crawls on mobile data will all lower the figure honestly. Counting taps does not excuse a broken mobile page.

A business whose customers never phone, such as a store taking orders online, should judge phones by orders like everything else.

But before redesigning anything because of a mobile conversion rate, make sure the site counts the ways phone visitors actually get in touch.

Before judging the mobile site

  1. Open the Mobile Overview under Audience.
  2. Compare conversion rates by device category.
  3. Add device to the landing pages report.
  4. Record phone number taps as goals.
  5. Record directions and email taps too.
  6. Fill in each form on a phone yourself.
  7. Compare again after a month.

Step four changes the picture most. Until taps on the number count as goals, the visitors most likely to call look like the visitors least interested in the business.

Giving whoever sets up those goals the right level of access is covered in Google Analytics access for your web company.


Frequently asked questions

Why do phone visitors convert less?

Often they do not. Phone visitors tend to call or drive over rather than fill in forms, and Analytics sees neither unless the taps that start them are recorded.

Where does Analytics compare mobile and desktop?

In the Mobile Overview report under Audience, by adding Device Category as a secondary dimension in other reports, or by applying the built-in mobile and desktop segments.

Is a high mobile bounce rate bad?

Not necessarily. A phone visitor who finds the number on one page and calls has bounced as far as Analytics knows, though the page did its job.

How do I count calls from the website?

Record taps on phone number links as events through Google Tag Manager and make them goals. Call tracking numbers also catch people who dial by hand.

Why are some visitors counted twice?

Analytics identifies visitors by a cookie on each device, so a person who researches on a phone and enquires from a computer appears as two separate visitors.

When is the mobile site really the problem?

When forms are awkward with a thumb, the phone number is an image, or pages load slowly on mobile data. Counting taps does not excuse any of those.

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.

Mobile conversion rate that looks too low?

Let us check what your Analytics counts. We will set up the goals phone visitors actually reach.

Start a Conversation