Measuring Landing Page Effectiveness via Google Analytics

The importance of landing pages and their roles in SEO & conversion rates is well documented. It is now common practice to design landing pages based on the source & medium of the referral. Clearly then, it is equally important to find a way to measure the effectiveness of individual landing pages. Simply knowing that they are important and that they have been built does not mean the job is finished. As with everything involving SEO, landing pages need to be reviewed, edited, & modified over time. Therefore, finding a quick and easy way to determine which of your landing pages are effective, and which are under performing, would seem to be a necessity. Bad landing pages are more akin to land mines in that visitors tend to disappear when they encounter them. Enter Google Anaylitics.

Readers of this blog know that we recommend Google Analytics for all our clients. This is not to say that Google Analytics is the right solution for every client. Larger clients running complex campaigns can benefit from a top tier analytics program such as Omniture or Core Metrics while such programs would be cost prohibitive for smaller clients. However, given that Google Analytics is free, simple to install, and fairly powerful for the vast majority of e-commerce sites, we typically install Google Analytics for every client in addition to any other analytic programs they require. Google Analytics can quickly answer 90% or more of typical analytic queries with minimal effort or training. To illustrate that point, let’s look at a nifty report in Google Analytics that will help us answer the landing page effectiveness question very quickly. The report we want to look at is the Top Landing Page report within the CONTENT section of Google Analytics. When you first bring up the report, you will be looking at the default view.

Click on Image for Full Size

The first thing we need to do is to change the view. Note the 4 VIEWS icons at the top right of the report. We want to change the VIEW setting so that we look at the tree like icon that is the icon at the right end of the list of icons (These views can be referred to as Grid, Pie, Bar, Tree views respectively). The tree icon allows us to see how individual landing pages perform against the site average.

Click on Image for Full Size

Now we just need to set the parameters we want from the 2 drop-down menus that are now available within the report header. Each drop-down menu has the same 3 choices: Entrances, Bounces, Bounce Rate %. For the purpose of this report, we want to measure bounce rate % for each landing page against the site average bounce rate, and we wish to view this information from most entered page to least entered. To do this, simply set the first drop-down menu to ENTRANCES, & set the second drop-down menu to BOUNCE RATE % (note how the second menu has “compared to site average” next to the menu). Bounce Rate % will tell us what percentage of visitors hit the landing page and then immediately leave (bounce off) the site. The resultant report should look something like the image below.

Click on Image for Full Size

Now, you should have an easy time picking out under performing landing pages. By using the top landing page report and the tree view, under performing landing pages will appear with red bars, while well performing landing pages will appear with green bars. The bars simply represent how much better or worse than site average each page performs when it comes to visitors bouncing off the landing page. Google Analytics also lets you jump straight to the offending page in question by clicking on the “double box” link icon just to the left of the url page link (if you click on the url page link itself, you will be taken to a summary Google Analytics report page for that landing page, NOT to the landing page itself). If we look at the same info using the “BAR” view, we can see the actual bounce rates for each page, but the TREE view works best for the identification stage.

Click on Image for Full Size

Before heading off to redesign all your under performing landing pages, it is important to realize the limitations of Google Analytics, or any analytic program for that matter. Analytics programs are great at pointing out non conforming data. In other words, the value and purpose of an analytics program is to allow the analyst the ability to spot inconsistencies in data. The data point(s) can be positive or negative indicators, but in either case, the data point in question appears to be out of place or outside the realm of typical expectations. What an analytics program cannot do is determine the cause of the obscure data point. This is the role of the analyst. It is important to realize this distinction when working with analytics because many people are under the impression that analytic programs can fix websites. They can’t, people fix websites.

Why spend a paragraph explaining a seemingly obvious point of fact? Because, the first thing you should do when analyzing your underperforming landing pages is to ignore analytics for a moment and ask yourself whether there are any simple marketing, technical, or other simple reason a page is bouncing. For example, it is not uncommon for a page to have a high bounce rate do to the fact that it is image or data heavy (the size of the page in Kb, or Mb(let’s hope not Mbs!)). Google Analytics will never be able to tell you that your landing page is 500k and take forever to load. This is something a human must deduce. Another example would be a search results page with no search results or a page that cannot be found. Again Google Anaylitics cannot tell you that this is the issue, but it can point out to you that something is amiss with that page.

Hopefully the Top Landing Page report in Google Analytics can help you fine tune your site’s landing pages. Just remember, while Google Analytics can find the land mines on your site, it is up to you to disarm them. Most times, a simple explanation or small fix will be enough to guarantee safe landings for all your future visitors.