Selected Q&A From Webinar: Using Web Stats to Improve Usability & Get Results

January 31, 2012 | by Chrissie Alquinta | Posted in Web Strategy

Thank you to everyone who listened in on my webinar last Thursday on using web statistics to gain insight and improve your website. It was a lot of fun and I got a ton of great questions! Unfortunately, time did not permit me to answer all of the questions, so I selected a few of the most common and answered them below.

Q: Define bounce rate

A: The bounce rate is the percentage of visitors who arrive to a website and leave the website before viewing any other pages on the site. Some analytics packages define the bounce rate as the percentage of visitors who enter your website and leave within the first 5 seconds and without viewing another page on the site.

Q: What are typical bounce rate standards?

A: This depends on the type of content. A general rule of thumb for a website is to have the average site bounce rate somewhere between 30% - 40%. However, if it’s a blog or other content-heavy site, bounce rates may be significantly higher and the site still may be successful because it is giving users all the information they need on a single page of content (like a blog entry, news item, event registration, etc…). If a page on your website has a high bounce rate, it is important to look at everything that is happening on that page to see if the bounce rate is indicative of non-sticky or bad content, bad usability or poor architecture, non-existent calls to action, or other content mistakes, or rather if users are in fact getting all the information they need from that single page of content. Often, some form of usability testing or user surveys can help answer that question.

Q: What are some of the more frequent mistakes made when reading/interpreting the data produced by Google Analytics or similar packages?

A:

• Taking the data at face value without looking at the website and not using other methods to gain insight. The data alone is just numbers. It doesn’t tell the whole story of what’s happening on the website nor why it’s happening.
• Looking at any one statistic and drawing conclusions from that. The number of visits to a website only tells you how many people came to your site. It doesn’t tell you whether these people were part of your site’s target audience, nor what they did on your site. For all you know, these could have all been bounces and no one is actually consuming your content!
• Comparing apples to oranges. For true insight you want to compare same to same. For instance, you wouldn’t want to compare sources of traffic and draw conclusions based on that because these overlap and are not the same. Referral traffic often includes mobile traffic and social media traffic and thus might appear “higher” than social media traffic. See what I mean?
• Drawing firm conclusions based on limited data. Seeing your bounce rate jump from one month to the next is not cause for alarm. Same with visitor flow and every other statistic. There is a natural ebb and flow on the web and website statistics reflect this. Before making changes to the website, observe a trend over a couple months, do some testing, and then make the change.
• Not defining goals. If you don’t have something to measure or to measure against, you can very easily just get lost in the data. Define your website’s goals, set key performance indicators, customize your GA set up, and measure, measure, measure!
• And last but not least, because I am a statistical geek at heart, please make sure that you make decisions based on statistically significant data. Just because your visits jumped 10% in the last year does not mean that the difference is statistically significant, i.e. that it means anything. Download Avinash Kausik’s statistical significance calculator and you will be in stats heaven.

Q: Can you help me understand how to read the "Visitors Flow" tab under audience? Is this useful?

A: The Visitors Flow tab shows the path users take from entering to leaving your site. I love the red bars that show where most people exit – this is super useful in determining where you might have some content problems or where issues might exist in a process on the site. What’s also great is that you can customize it to view the visitor flow by different segments of your site’s visitors. The example below shows visitor flow by traffic source. In this case, it is useful to see that most visitors from Twitter want to see the kinds of services the company offers, thus we know that this is potential source of clients and can tailor our tweets accordingly.

Q: Could we track the visitor traffic source (domain) for each content/page? Who is visiting which pages of our website? 

A: Absolutely! One way to look at it in the aggregate and for only the top few pages is to view the visitor flow chart and filter it by traffic source. Another way is to navigate directly the content page in which you’re interested and select Source under Secondary Dimension. You will get output like this that shows data for each traffic source to visit that particular piece of content.