by Rydal
In the web analytics world, what do we mean by “bounce rate”?
Wikipedia: It essentially represents the percentage of initial visitors to a site who "bounce" away to a different site, rather than continue on to other pages within the same site.
The formula used to calculate bounce rate is: Bounce Rate = Total Number of Visits Viewing One Page / Total Number of Visits
Just think about a ball, a ball bounces when it collides against a plane surface, the same can be applied to your visitors, a visitor bounce once he/she leaves your site after the very first page view within that session. Apparently, if you have a one page site, your bounce rate will be pretty much 100% or very close, the bounce rate of your site shows you how active your visitors and how effective are your entry pages and local links that can be used by your visitors to find other relevant content that they could be interested in.
So in essence, if your session timeout is set to 30 minutes which is sort of the industry standard, then a single visitor is considered bounced if he/she lands on your entry page and remains inactive for over 30 minutes or simply navigates to another page on a different site before the session times out.
How effective is it? Well its hard to determine because it completely depends on your site content and what you really want your visitors to be doing, people that create one page landing websites want their visitors to read, scroll and buy if they have anything to sell, therefore, bounce rate means nothing to them. In the other hand, if you are an ecommerce site, you would want your visitors to browse, add to cart and buy the items in cart – in this case bounce rate is of extreme relevance!
Bounce rate means nothing in particular until you apply context!
e28f28b5-6fe3-4e52-a447-9c17f509fed0|1|1.0
Tag:
E-mail |
Kick it! |
DZone it! |
del.icio.us
by Rydal
I don’t know which planet I was when this happened but apparently Google has their own URL shortener service at http://goo.gl/
69fc56e1-e0bd-4241-b774-d63387da4243|0|.0
Tag:
E-mail |
Kick it! |
DZone it! |
del.icio.us
by Rydal
I finally finished the link log earlier on in the day today or should I say yesterday, I’ve included a screenshot below to show you what it will look like, the green arrow indicates an incoming link and the red indicates an outgoing link. I’ll push the changes to public beta soon.
b659ce23-e932-4e79-a3e1-d301caceb5f8|0|.0
Tag:
E-mail |
Kick it! |
DZone it! |
del.icio.us
by Rydal
If you have been to any slow website at all, you should know that speed plays an essential role when it comes to retaining your visitors and reducing your bounce rate. I’m a perfect example, I go to Google just to search for something that I already know what its going to be or what the URL is but I’ll rather click than type the long URL if I don’t have it bookmarked. Its incredible easy to type in the little search box on the top right of my browser and click on the link from within the Google search results page.
I’ve been watching myself and I keep doing it over and over again, I asked myself one day “Why do I do this?” and the answer is simple – its faster! Secondly, it seems your search engine ranking is now affected by your load times because who wants to seat around and wait for 5 minutes for your site to load and slow the rest of the world down!
Therefore, its safe to say that fast page load times increases our visitor satisfaction. So how do you track page load time within FoxMetrics, pretty simple as everything else is.
<script type="text/javascript">
var fxmPageStartTime = new Date();
</script>
Place the above script somewhere within the head <head> and </head> of your page or pages that you would like to track and that's it. Don’t forget that your regular page view should be placed at the very bottom of the page right above the closing body tag </body>. FoxMetrics will automatically track and store page load times for every page view.
What do you do if you find out that one or more of your pages are slow; well for starter, I’ll suggest you determine what is slow by using one of the performance tools that are at your disposal. I personally like the free Google tool called “Page Speed” that is a Add-onto Firebug for the Firefox browser.
You can also perform some external tests by using the free tools available at http://internetsupervision.com and http://tools.pingdom.com/
ada77448-5135-4ab4-ab1d-e7ebcab0ebd8|0|.0
Tag:
E-mail |
Kick it! |
DZone it! |
del.icio.us
by Rydal
Yes, I do have a secret well it won’t be after this post – as you are all aware, FoxMetrics has 3+ strong points that makes it stand out of the crowd.
- Real time
- All your desired analytics gadget within a single location
- Simplicity
#1 is a pain, hence the reason why even the big dogs like Google, CoreMetrics and etc can’t afford to show reports in real time and its simply because there is a lot of data to be collected and lots of data crunching. At this point in time our fetchers run at 1 minute intervals and they could take up to 5 minutes to complete their tasks, therefore, there could be a 5 min lag on your dashboard update which considering the facts I do think is bearable. However, I see a possibility of exploiting other measures and improving on those times. My thoughts are very valid, I assume that 5 minutes is ok now because FoxMetrics is not at its full potential and not all of its widgets are being used, so if you think about it at full capacity our maximum 5 minutes processing time could go up to as long as an hour or more.
So what I’m I going to do about it? Well after some research I found out that most of the bottle neck falls within the network and database layer. If I move some hardware around meaning keep each piece close to each other as possible and also move the dashboard from an RDBMS to a memory database or an OODBMS, there is a great chance that I could eliminate a great amount of lag and improve on performance substantially.
I’ve set out to write my in-memory database that will be used to store reporting information for the dashboard, the RDBMS will strictly be used to store data after the fact, sort of a backup which in essence; I could build the in-memory database from the RDBMS if one of the nodes decides to go for a walk to la-la land. I’m really happy with this new plan and I’ve already started writing code for it – development on the dashboard continues.
b82d5552-088a-412f-9c80-65fd46daf526|0|.0
Tag:
E-mail |
Kick it! |
DZone it! |
del.icio.us
