Assuming you’ve taken the brave steps towards starting again improving a web app or website – what should you do to benchmark success? This post gives an overview on how to use Google Analytics if you are a design professional, and to use data as the basis for some of the decisions you make during that process.
Start with a Baseline
Assuming that your client has already been collecting metrics, and understands WHY the website needs redeveloped, you need to keep, record and understand that historical data to make sense of what needs attention, and how the overall problem needs solved. I’ve seen far too many projects fail twice, because no one thought it sensible to examine why it wasn’t successful first time round from a data perspective.
Web Analytics data is an obvious starting point. Ask for access to website logs before you start the project and inwardly digest. I’ll cover the sort of things to watch out for below.
Find the Popular Content
Knowing what content is proving popular with an existing audience is a great starting point when making design decisions. There are a number of ways to work this out using G.A.
The top content report in Google Analytics can at a glance let you know which pages on a site are receiving the most traffic, or how long a visitor spends on a particular page. BUT – this doesn’t necessarily mean that a page is popular or useful to visitors.
Instead, look deeper at your top content report via metrics such as the content which drives repeat visits. This may reflect that content has made it’s way to people’s bookmarks. Selecting Returning Visitors from ‘Advanced Segments’ then select the ‘Top Content’ section in G.A.
Returning Visitors may give you an overall feel for successful content, but to really know the nitty gritty of why those visitors are coming back, pivot tables are a great way to break down the data. Select the pivot icon:
Then break it down by medium…
To get a report pivoted like this:
This will allow you to see the returning visitors that search engines send, and the corresponding content. Or.. see the returning visitors that arrive via direct traffic. This lets you cut to the content that people return to because it is of a high quality, and doesn’t need high search rankings to be drive traffic.
If Google Analytics isn’t available to you, or you for whatever reason can’t get access to data – the social web can uncover some hidden gems about the information people are finding useful.
You can then begin the process of highlighting the popular content in your design process.
Analyse Existing Goals and Conversions.
The overall goals of a website should be core and central to the design process. Forget fancy graphics, these are the issues that really matter for your customers and the success of a project. Find out what those goals are, and how they are currently being measured.
This could be a simple URL that is visited when a form is submitted, it may be a sale, or simply someone reaching a contact page with a telephone number. Here are a few ways to more deeply analyse what is happening with a particular URL:
Through what path was a page found?
Knowing how visitors managed to reach important URL’s such as conversion pages is extremely insightful data for designers. For example, if you are selling Goods – do visitors read technical data on a separate page before making a purchase? Or would integration of this data on product pages increase conversions?
Select Top content first and foremost, then perform a search for the landing page or URL in question. For example, lets assume that you want to find your contact page, and the information relating to it.
Clicking on the dropdown highlighted in green will allow you to see a breakdown of how people reached that page from sources, campaigns and give detailed information on visitor types. This will let you see for example, if people came directly to the contact page from Google, through direct traffic, and whether they were repeat or new visitors.
Navigation Summary will identify the previous pages through which this one was reached, giving an aggregate summary on what other pages were viewed. Lets take a look at that:
The information on the left of the navigation summary lets see the pages visited prior to that page. About 23% of clicks came from the home page through to this contact page.
On the right hand side of the navigation page:
We can see that nearly ten percent went on to convert, but a good chunk of people changed their minds too, and returned to the home page again. This kind of breakdown lets us see both the pages that make a difference to conversions, and whether the final landing page is performing or not.
What pages on the site are driving traffic?
One of my favourite reports in GA is the content drilldown report. For WordPress, it’s one of the reasons I tend to keep the year and month in my URL’s. Here’s why:
(Select Content drilldown, then click on the performance icon:
You can breakdown, month by month, year by year, the amount of traffic certain pieces of content are providing you. Not only that, but if you are selling stuff, and have the data tied up to Google, you can see what categories of product are driving most sales, broken down by month.
For a designer, you can gauge what content is making a difference to the business bottom line, and take this into account early into the design process.
Find the under performing pages
You are of course going to want to know not only the pages which are doing good, but the ones which are failing as well.
Poor performance can be deemed to be any number of factors, depending on the nature of your site, and what it is designed to do. It’s all about context.
Does a high bounce rate on a page mean that a visitors found what they were looking for quickly, and subsequently left, or does it mean they came they puked they left?
1) High traffic, High Bounce Rate, Low conversion
2) Low repeat traffic
3) Low time spent, High Bounce rate
4) High time spent, Low conversion rate.
We are going to use an advanced segment to show things. Firstly, we want to segment only users that Bounce. This will then let us use other reports to gain insight.
Select Advanced Segments, then use this:
Next, we want to find out what pages are generating the most traffic, and are bouncing. Again, select the top Content Report. This is going to tell us what pages are generating traffic, but are also bouncing.
We can see from this report that this is the page that generates the most bounces, and correspondingly the most traffic. From this, we can then examine the all important WHY!
Clicking on that particular result will allow you to see a number of additional metrics, for example, what language did those visitors have set?
Well, we can see that there is a fair chunk of spanish traffic. That would be an area to investigate. Can we for example translate the page into Spanish and lower the bounce rate? Another thing to analyse would be the keyword used to find the page. If visitors are using long tail phrases, but the content on the page isn’t satisfying their needs, you may want to raise the bar in terms of quality.
Top Landing Pages can also be found by looking at Content > Top Landing Pages. This data will be useful at identifying the popular content, with the highest bounce rate, and therefore in need of care and attention.
Design, Implement, Iterate and Test
Another great feature in Google Analytics for designers is the annotations feature. I love this because it lets you make notes about your changes, and over time come back and review them.
If you rely on data to help you make decisions about your website, and the design process this is invaluable. You can pinpoint the exact moment you changed something, and compare conversion rates or other metrics the next day to see the impact of your change.
As you move forward with your fresh design, you can begin the testing process right within Google Analytics itself, and iterate and evolve the design over time to bring a truly superior solution not only in terms of looks, but in terms of performance as well.
You might also enjoy..
Filed in: Web Design
About the Author (Author Profile)Paul is a regular 28 year old web bloke / programmer with a penchant for online marketing. This blog is a personal outlet, with an eclectic mix of articles.
Sites That Link to this Post
- 7 awesome realtime design feedback tools for instant results. | July 23, 2011