Instinctive Software » Testing

TV Ratings with Nielsen

admin — Sun, 23 Jan 2011 21:51:03 +0000

Here is an article on Nielsen ratings which explains how they are starting to take in account viewership habits online for TV shows.
http://blastr.com/2011/01/the-truth-about-tv-ratings-online-viewing-and-sci-fi-shows.php

The new breed of statistics are very similar and much more readily available.
http://compete.com/
http://www.quantcast.com
http://www.alexa.com

Operating system market share

admin — Sat, 08 May 2010 19:47:33 +0000

When targeting the internet consumer knowing what they are using for Operating systems and browsers is necessary. This determines what browsers and operating systems you are going to support.

Additionally monitoring usage changes allows you to predict when you need to add or remove a supported browser or operating system.

This can be done with two tools, one google analyics or a report from net market share .
http://marketshare.hitslink.com/operating-system-market-share.aspx?qprid=8

Even though Windows 7 was released in Oct 2009, windows continues loosing market share over the past 6 months. This is the general consumers usage pattern.

For your web site, things might be completely different, a Mac OS site may naturally have mainly Mac users. Where as a game site for a PC based game would mainly have windows. These are obvious factors.

If you are targeting the general user and your profile does not match net market share’s report then finding out why may teach you more about your customers or highlight unknown browser compatibility issues.

Read cross browser testing for an example of the importance of cross browser testing.

Cross browser testing

admin — Wed, 22 Apr 2009 17:30:16 +0000

Cross browser testing is one of the critical tasks that needs to be done if you have a high volume of customers or are looking to grow your existing user base. One way to stop users dead in their tracks is to not cross browser test an upgrade.

While a site might work fine on your new computer with the latest browser. That same site might not work so well on your customers computer. For example, say you have 100,000 people visiting a shopping cart this week and you earn on average 1 dollar per person visiting. You sell lots of widgets and have a top notch customer service.

Now you want to make your site better, faster and easier to use so you upgrade your add to shopping cart button and then release your site with out cross browser testing. There is two things that may of not been taken into account, one users with Internet Explorer 6 can no longer add items to their cart. Two 15 % of your traffic uses IE 6. This results in a 10 % or more drop in sales on the week of the upgrade.

If you are lucky you have some kick ass internal reporting and pick this up with in hours of the upgrade. Though, if you don’t it might take weeks to find out about the problem.

With out it, every 1 in 1000 users who have the problem complain, but not accurately enough for customer service to figure it out. Maybe customer service takes two weeks to notify development that there is a problem. Then the software developers take another two weeks to figure out what it is and they must just say it is not worth while to fix.

If you are luckily you have some good internal reporting on sales and you see the drop in sales with in hours or a few days. Other wise might might be a month down the road before you see the noticeable drop in sales.

What this really adds up to is for the sake of having a developer do a days work 10 % of sales might be lost for any where from a few days to a month. The profits lost on a month of sales is far more then the cost it would to have the site cross browser tested.

Then there is the hard to measure costs:
How many people complained to their friends that the site did not work ?
How much future business did you lose due to that customer ?