Consulente Web Marketing e Consulenza Motori di Ricerca

Web Marketing & Search Engine Consultant

Search Engine Optimization and Bounce Rates for Web Sites

In the never ending adventures the search engine landscape has to offer, today was a good day for a particular Client.

It’s good when things work as you would want them to. They gave me a call and it was a pleasure to hear how happy they were.

We had a long conversation, many considerations on how things are progressing this season; I went on to highlight the bounce rate that has always been a problem for this Client.

But there’s nothing worse that a Client in love with his web site – you know they sit in front of the screen and admire it  … that’s more or less what I am dealing with here. 

The bounce rate is very high here and it has further increased in the past few days with the new top level (generic keyword) positioning achieved  recently. Suffice it to say that we have reached 77% – just too much for me to take sitting down. 

Prior to this new top position, the bounce rate was in the order of 67% – way too high from my perspective, considering it is keyword focused and generates search engine traffic on specific keywords tightly related to their line of business.

After 3 years of lengthy discussions, today I think I finally convinced them we need to look at design which in my view provides a poor user experience.

I wrote out a matrix of questions, asking them to identify at least 10 people who are not acquainted to their web site, potentially interested in their services, have them surf the web and take a quick survey.

Here are the points I raised (sorry, can’t publish the questions … you should be able to extrapolate your own list of questions based on the following considerations)

  • Is it Useful ?
    [what was the users experience, was it a positive one, if not why. You need to understand if what you are saying is relevant in the eye of the reader]

    • Were any problems experienced ?
      [Did the pages download quickly or was the wait too long ? Were there any problems identified during the navigation ? You need to understand if the experience is viewed as reliable or not.
      Was it difficult to navigate ?
      [in other words you need to understand if your web site is functional or not. It’s a good idea to ask the users to perform typical tasks like a request for further information as a first step]
  • Is it Useable ?
    [Make sure it is accessible, convenient to use and familiar to the user]
  • Is it Desirable ?
    [People listen to their heart – emotions are the protagonists. Remember that desire is the ability to go beyond logic. People tend to buy what they want not what they need … You’ll want to engage your users with a call to action after empowering them with an empathetic, relevant and timely interface]

A real challenge isn’t it ?

What about your web site – what is your bounce rate ?

Have you ever performed a check on your website interface ?

A great Podcast on Humanizing the Digital Experience is available on the Forrester Research web site.

8 Replies

  1. We’re currently redesigning our site and I’m probably rather like your client – I love our new design (its not live yet) and do gaze at it adoringly. However, we do also understand that its primary purpose is to help us sell our services to more customers.

    A bounce rate of 77% certainly sounds very high and whilst I realise this is a very open question, what sort of rate would you consider acceptable?

    Thanks

  2. Hello Julian,
    thank you for stopping by and leaving your comment rich in details that gives me the opportunity to elaborate a bit on the subject. I took a look at you design and I too like it very much at first sight. Very elegant loads quickly, it gives me a good feeling.

    I ran a quick analysis for you – here are the results:

    At first sight I would say there hasn’t been much on site optimisation here – the look and feel of your titles is a bid “stiff” not keyword rich, not user friendly in the sense to reach out to the user and make a call for action. I may be wrong

    I took a look at your sitemap and I’d change it around a bit: keep in mind that you shouldn’t have more than 100 active links on a page if you want to be sure that the search engines will crawl them. Your best bet is to make a 1st level sitemap containing links to your main categories, hence linking out to pages with links to category pages always keeping in mind the 100 l ink threshold you don’t want to exceed.

    Overall the site looks good on Google, supplemental results show up well after the 95th page when issuing a site: command. This is very important, but you want to keep you eyes on what’s happening, your titles are not very strong and personalised.

    congratulations on your deep link ratio, well over 84% – that’s great news for your website – it means many links are point to inner pages of your website.

    Now having said all that, let’s get back to the bounce rate – levels of 40% are more than acceptable in most cases maybe even lower but I wouldn’t push too far in saying how low it should be.

    IMO 50% is a good starting point from which you might want to start your on site work and attempt to identify just what is it that’s driving people to leave your site.

    It could be there are landing on your website with the wrong keywords, and you’ll need to study the dynamics of user interaction to understand this.

    Software like clicktracks is good, or Google analytics can help you in your performance studies. There is no such thing as an ideal bounce rate as fare as i know – In the web analytics filed there are still debating on the definition of bounce rate, so focus on your fundamentals adn give me a call when you decide to enter the Italian market for a quote on translating and optimising your site 😉

    (Sorry couldn’t resist the temptation …)

  3. The trouble with talking about bounce rates is that every website is different and deals with different variables that can affect the rate. It’s not even correct to say that generally a lower bounce rate is better. It’s better, “all other things being equal”, but since when are all other things equal? If a website’s bounce rate is very low, it might just mean it doesn’t rank for anything and the only traffic it gets is from those persons who are specifically looking for that site. In the case of my firm’s website, I have a three letter domain (mwi.com) and that means there are a lot of other companies out there with the same acronym for their name, and so I get a lot of traffic from people looking for those other companies, who of course bounce right away once they come to my site and realize it’s not what they’re looking for.

    To effectively use data about the bounce rate you must only be doing comparisons about one site’s data compared against itself over time, and every variable that can affect the bounce rate must be taken into consideration.

  4. I agree and I’d go just a bit further to say that every website has it’s own history and users.

  5. SEO

    i totally agree with Sante J. Achille

  6. Can you derive value in this sense: Lots of bad traffic (10,000 visitors) vs. good traffic (2,500 ). Maybe the bounce rate is pretty high, but if the voilume is great enough, that is not bad, unless you are ignoring targeting good words as well (low bounce rate). I have an overall bounce rate on a couple sites as low as 23%.

    I also agree with “Utah SEO” (mwi) as this is clearly taught in Google’s Analytics video series on Youtube. It’s all relative.

    Personally, I would find it more useful to look at conversion rates too next to bounce rates to see the correlation.

  7. Dictionary.com is one of the most popular websites in the world. It ‘probabaly’ has bounce rates of 90% for obvious reasons; see the correct spelling, hit the back button.

    Do not get caught up in the bounce rate figure too much, take much more notice of your intended goals.

    Regards

  8. theGREATunknown

    Bounce rate is a very weird metric which usually includes different meanings across different platforms. The world wide web analytics association will say one distinction while google will say another. The issue is by using ajax and alike, its getting harder to gauge BR with traditional tagging.