It is a good idea when looking at a website to figure out any issues with the site. There is many ways to do this and many different pitfalls. One way to spot issues is by looking at the bounce rates. Depending on your analytical package, bounce rate can be viewed per page, device, browser, traffic source, language, amongst others.
Here we look at an example
The bounce rate is an indicator of how well your website does in enticing visitors to continue through to different pages on your site, though there are differences in how bounce rate is calculated depending on the analytical package you are using, so it is worthwhile checking how they do it. There is also a paradox when it comes to casino affiliate sites as we want the traffic to leave to one of our recommended partners as well as have good user metrics. Its all about balance.
One of the sites I have done work on is 3cardpoker.com, an informative website that has hints, tips and offers for where to, how to and why to play 3 card poker online (or indeed offline). It is a niche site and does not get that much traffic compared to the average industry site but we still wanted to improve the metrics and any issues with the site. Here I am going to go through one issue that was found with the site which was leading to an increased bounce rate.
Identify the overall site bounce rate
Using the information available from the sites analytics, in this case clicky.com, the sites bounce rate was identified for the previous 3 months, it was between 22-23%, not a bad bounce rate but there was certainly room for improvement.
Now that the overall bounce rate had been identified it was time to delve deeper into the analytical information. A time consuming process but well worth it as it can identify some things that otherwise would not have been spotted. Also with many analytical packages reports can be setup to make the task easy on future occasions.
After looking at the overall site bounce rate, each page was looked at individually (a report can be set up to do this on many analytical platforms otherwise it is a matter of checking each page individually to see if there was any major differences between pages bounce rates. Recording the result of each page on my spreadsheet (some analytical packages will produce a report for this or one can be setup) I identified two pages that had a considerable amount of traffic (in comparison to the rest of the site) and a high bounce rate.
One of the pages was about playing 3 Card Poker in Las Vegas and the other Atlantic City. The Las Vegas page was the second most visited page with a high entrance rate and high bounce rate.
Having already put the site through Google Page Speed Analysis amongst other applications, to identify any issues with speed (many users especially on a mobile will not wait for a site to render if it is taking too long) it was time to delve deeper into the analytics available.
It was possible to check many different metrics to the site. The devices that visitors viewed the site on, the browser used, the location, the languages, time spent on pages, buttons clicked, videos played etc.. All of these things can be used to identify an onsite problem. After checking through the bounce rates for numerous filters, the different traffic sources for the two pages were downloaded. On Clicky.com the Traffic Sources listed are Searches, Direct, Links, Advertising and Media Searches.
Applying the traffic sources filter it was clear that many of the searches were coming from media searches with monthly nearly a1000 visitors having a bounce rate of 64.1%. The media that was causing this was the images on the page. With the image ranking high in Google Images for the search term “Las Vegas”, it was clear that the users clicking on this were looking for information regarding Las Vegas and not about the niche specifics of Three Card Poker in Las Vegas.
By informing the Googlebot in the robot.txt file to disallow the image we prevented this traffic from arriving at the site. As we did not have the information available it was doing us nor them any favours and slightly sullying the analytical information available. Better to have quality visitors to a site who want to be there as opposed to people who are looking for other information and will distort the analytics.
# Google Image
User-agent: Googlebot-Image
Disallow:/IMAGE SOURCE LOCATION
Allow: /*
Though only a couple of percentage points were shaved off of the bounce rate every little bit helps lead to better analytical data and improved metrics (which has a realtionship with improved rankings). It should also be noted that this might reduce the chances of getting an image backlink accrediation.