Headline Statistcs aren't the answer

3 reasons “Hero” statistics aren’t necessarily so heroic

Why headline statistics don’t paint the whole picture when it comes to visitor engagement….

+ more

Whilst there are those who understand data and many more who are interested in meaningful data, there are far too many that are happy to grab the headline or “hero” stats and use these to proclaim success.

When it comes to Web Analytics, the arch enemy of such stats is Average Session Duration. On the face of it an innocuous – useful even – statistic. Surely knowing how long users are spending viewing your website is a good thing, right?

Well, yes and no. You should absolutely want to know how long visitors spend on your website, however, Average Session Duration isn’t the most accurate statistic to provide this insight.

Here’s why it won’t tell you anything about actual visitor engagement:

REASON 1. Google starts the clock as soon as a visitor loads a page of your website for the first time. That clock doesn’t stop until there has been a continuous 30 minute period with no activity. This means if your visitor goes off to a meeting, or to make a coffee, but returns within the 30 minute period to continue looking at your website, the clock is reset and all of the time that your visitor was away from their desk is counted as part of their session even though their browsing time could have been as little as a few minutes.

REASON 2. Similarly, if a visitor clicks a link to a third-party website, the time a user spends on that website before returning to your own (within 30 minutes) is still being recorded as if they were looking at your own content. This has a knock-on detrimental effect for the Time on Page statistic too.

REASON 3. Google can’t see into the future. When a visitor loads the final page they choose to view the session timer continues to run until the 30 minute period is over because no more events are received. Google has no way to know how long your visitor was reading that final page for. If your visitor views only two pages, Google effectively only records 50% of the actual session duration. And in a worse scenario, if they view just one page, the session duration will be zero, even if they spent five minutes reading the whole page.

So, what metrics should you be using?

Rather than relying on session duration you should be measuring visitor engagement. How far did a user scroll through your pages? How many pages were viewed for more than 8 seconds – the average adult attention span? How many assets were downloaded or videos viewed? Did the visitor complete any form fields? Did they return or was it a one-off visit?

Use a combination of pageview metrics and your own custom events to capture a multitude of user interactions. These will all combine to provide a much greater picture of your website’s usage than a mere figure displaying elapsed time.

You’ll need to decide the appropriate metrics to measure against your individual KPIs.

It will take time and effort, but that work will be rewarded with greater insight and hopefully better evidence of your website’s performance than by just sharing your Average Session Duration.

Author: Gavin Parkes

Gavin Parkes

Gavin is in charge of making sure we measure the success of all our campaigns. As a new dad, he's also a student of the art of nappy changing.

Mat_260614

Show some digital hospitality

This is about retailing – but frankly, it applies to anyone with an online, customer-facing presence. I do the lion’s share of my shopping online. And as I browse, I’m often surprised by how confusing and frustrating retailers’ web sites can be….

+ more