One of my favourite ”remedy” for bad stinky pages with high bounce rate from different traffic sources such as display campaigns, SEM, SEO etc. is to analyse the Top Landing Page report. In this blog post I will describe a quick win fix analysis using ”top landing pages”. At the bottom of this blog post I have also added a custom report with inspiration of a blog post from Avinash which I found useful.
I will use SiteCatalyst in this blog post (but you can do the same things in Google Analytics). This report in SiteCatalyst will tell you which page was the most popular (in terms of traffic), sales, bounce rate, orders and so on.
In the example below I have created a scenario report with mockup data. The report contains different pages with metrics page views, orders, bounce rate and weighted bounce rate. A quick look will tell you which pages belong to shopping, sections, campaigns and products.
So what does the report tell you?
Firstly, we need to establish what a ”healthy” bounce rate and Weighted Bounce Rate is. Bounce rate = I came I saw and I Puked (using the expression from Avinash). If I would generalise, I would say that a bounce rate should be around ”30-50%” (the lower the better off course). If over 50% then you need to fix the stinky page. If over 70% then dial 911 and bring your buckets because there is a fire!
We can quickly establish by looking at bounce rate in the report that:
- line 1 and 8 is healthy
- line 6 needs attention
- line 2,3,4,5,6 and 7 needs serious attention
However, how relevant is this bounce rate compared to ”all traffic” to your pages. Remember this post is about ”quick win and fix”. The weighted bounce rate gives weight to pages that are more viewed than others and therefore pushing “stinky” pages to the top. Otherwise you will find irrelevant pages with 100% bounce rate at the top with low page views (see image below).
If we look at the report again we can see that the pages that will be part of our ”quick win” fix are the ”Shopping process step 2” (19% weighted bounce rate) and ”Customer Support” (10% weighted bounce rate).
Why do pages have high bounce rate?
1) Incorrect traffic to your page
- problems with campaigns e.g. sending people to a page about “hotdogs” for the keyword dog (I know bad example but you get it..)
2) The page is not good constructed
- If you visit the site the problem can be straight forward. A deep analysis can tell you how visitors came to your page (e.g. which keywords did they use) and which traffic contribute (referrers) to the high bounce rate
For the customer support page we can now investigate different strategies to lower bounce rate e.g.
- Investigate how the customer navigate to the page (use path reports)
- How are they using the internal search (top subjects) and is the information sufficient?
- Usability of page (Optimize the page with Test & Target, Visual Web Optimizer, Google Web Optimizer etc.)
- Use Click Maps, to investigate the click pattern on the page.
- Are visitor finding the support information they need or are they just looking for the customer support number.
One good example is a company who use a different customer support number on their homepage and then upload the data (customer support calls) into SiteCatalyst. Doing this will enable the company to measure their customer support performance on their homepage compared to their offline customer support. Their goal variable is, the less calls to their offline support the better (and cheaper).
In terms of the ”shopping process step 2” page, we look at more or less the same strategies above but we can also look at referrers (just click on the page name in the report and navigate to Referrers. See image below
In this scenario, analysing my campaigns I realise that:
- I send traffic from Google to incorrect landing page. I need to fix this because my visitors’ expects content that I am not delivering.
- Affiliate 1, seems to be sending lot of traffic but 0 sales. I will contact this affiliate and investigate how the traffic is sent to my homepage. Very important is to investigate how the affiliate is compensated. If my company pay him CPC then I must make sure to close the campaign immediately because this guy could be a crook.
Entries: how often do visitors land on this page
Bounces: how does the page perform
Bounce Rate / Weighted Bounce Rate: I can quickly see which page needs attention
Visits: how many visits did the page have (remember 30 min cookie)
Page views: for those who feel this is relevant
Time Spent: which content was consumed
RVP (Revenue per visit): how much did a page influence revenue during a visit
Orders: because we want the RVP to be relevant, see image.
I hope you find this blog post useful, and please do comment or add your own experience.