Blackbeaks Blog….All things Analytics

I was recently asked a question by one of our subscribers that deserved a good answer. Ben Reed, working for a UK company in the health sector, asked;

I have been researching my companys bounce rate and found your page. I thought it was very useful for my research. However, one thing got me. In a different article you suggest that the reason for a bounce could be that visitors leave to visit your article authors or that your site is found for irrelevant searches.

Couldn’t it just be that the page the visitor landed on perfectly explained the answer to their query? ie the keywords were relevant and the content was useful? Take my user experience as an example, I wanted to know what an average bounce rate for a website was. I typed in Average Bounce Rate into google, your page came in the top 5, I bounced 2 sites before I visited yours. Their pages did not satisfy my query, but your page was useful.

Your page solved my query yet I will be included in your stats as a bounce, despite the content being useful, and the keywords being relevant. So it is more an issue of conversion as I did not register to become a subscriber despite having a good user experience (I have done since though! - BlackBeak… we like it! Good man!).

Do you agree that a bounce could merely mean that the visitor found the information that they were looking for at their first attempt?

Thanks Ben, as I said, this was a very good question. My answer follows;

Since I wrote that article (and I am guilty of not writing a follow up) I have since added stronger engagement metrics to our overall strategy.

I’ll try to quickly explain.
I am in the B2B industry therefore single visitor behavior is interesting because the most engaged visitors could be leads we might try to identify.

We now look at what people do on our site over time. I want to know how many people have used a keyword I consider to be a content match. Bounce rates as a keyword is one of my brand match keywords. So had you typed that you would’ve gone into a segment. Subscribing is another action you would have been segmented on. Reading the blog is another segment. Referring a friend is another segment and coming back 3 or more times is another action I am using to filter out the noise.

Anyone we can see who falls into 3 of those segments is interesting to us and since people have often given us their name it’s often possible to make an offer.

Bounce rates used to concern me greatly as our rates were so high. Then I figured out what you figured out. Blogs are the same, they tend to be very high single access pages, despite the amount of links. Think about a blog and it’s quite obvious, it’s one long read (scrolling down) and then you’re finished. No need to click anywhere. This results in a very high bounce rate. So now I worry far less about bounce rates on the Chronicles and largely it’s because people are finding what they want.

What I needed to do was find more insightful KPI’s than bounce rate alone.

So there you have it. Basically the question is about how we measured engagement. While I think for many websites the bounce rate is a superb metric for finding potential problems, alone it often isn’t good enough.