4 Common assumptions not to make when using Google Analytics

On the face of it, the standard reports in Google Analytics (GA) are simple and intuitive.

But, as an old boss of mine was fond of saying: “To assume is to make an ass of yourself.”

So, in a bid to help prevent people inadvertently making asses of themselves by misinterpreting their website data; here are 4 common assumptions to avoid making.

1. That total users represents that number of individual people that visited the site.

Most people look at this figure and assume it means a unique visitor, an individual who has come to look at the website.

The Google Analytics help text even supports this interpretation. It tells you: “The Users metric shows how many users viewed or interacted with your content within a specific date range.”

It would be easy to dedicate an entire blog to explaining the term user and how Google Analytics calculates it in different situations, but the key point to note is that:

A “user” is a unique combination of browser and device, not a person.

So a visitor who first viewed you website on their office computer and then checked it out again using their iPad on the way home from work would be recorded as two users.

2. That Average Session Duration is a true representation of how long people spend on the site.

Average session duration is an important statistic because it measures people’s direct engagement with a website. But use it to identify trends; on its own it is probably the most misleading statistic in Google Analytics.

The problem is that whilst it is easy to determine when a visit starts, because the website has to download that first page. GA does not know when a visit ends.

Knowing that a second page has been downloaded is straightforward and so is a third or fourth. But GA has no way of knowing when the visitor moves onto the next website, and therefore does not know when the visit has ended.

This means that the session duration is actually the length of time between loading the first page and loading the last page. It does not take into account the time spent on the last page. Bear in mind that the last page may well be what the visitor was looking for and spends most of their time on it.

To make matters worse the statistic is horrendously distorted by the bounce visits.

A bounce is when a single page is visited. It is not uncommon for over half the visits to a website to be bounces. A bounce visit has no second page and therefore session duration for these is recorded as 0 seconds.

But because average time on site is calculated as total length of time of all visits divided by the total number of visits (including the bounced ones) the average time becomes significantly understated.

3. That direct visits are always reported as direct visits.

If you look at the All Traffic / Channels report or the All Traffic / Source Medium report in the AChannel Groupingscquisition section of the GA reports it appears to show how visitors arrived at the site for each session.

Looking at this snippet from the Channels report you might think that 427 visitors came direct to the site, whilst 740 used a search engine, 82 were referred from other websites and 57 from links in emails.

However, Google Analytics is a bit more sophisticated than this.

It will categorise the session according to how the particular visit was initiated for all channels except when the visitor comes direct. For direct visitors, it will look back to see if they have visited before using a non-direct method and put the session into the same channel as that. Only if the visitor has not visited before using any other channel will the visit be shown as direct.

Hence a visitor, who first visits using a search engine and subsequently keeps returning by directly typing in the URL, will continually be categorised under the Organic Search channel.

4. That the Funnel Visualisation report shows the actual path followed by all visitors.

This is a great report. It provides a simple visualisation of the path you expect visitors to take through a website as they progress towards a specified goal. This can tell you a lot about how effectively each page is working in attracting visitors and guiding them towards the purchase, sign-up, download or whatever other conversion you are trying to achieve.

But people are prone to taking it at face value and assuming it is showing the exact path followed by those visitors to get Funnel reportto the goal.

It doesn’t.

For a start it does not matter what order the visitor visits the pages in, they are always shown as visited in the order defined in the goal set-up. Furthermore, the report will automatically include as visited any steps that were missed out by the visitor.

If you need to see an accurate picture of how visitors navigate around the website the Goal Flow report is the place to go.

Make sure you understand it, before you report it.

Whether you consider these faults or features, they don’t stop Google Analytics being a powerful tool for helping you understand what is happening on your website. But, like all analytics tools, you need to make sure you properly understand what you are looking at.