You are here: Home > Resources > Accessibility Resources > Automated Accessability Testing
As technology develops, websites are becoming increasingly filled with high specification graphics and multimedia applications.
In an attempt to save time in the testing phase, many web designers have begun turn to automated software such as Web Xact, Bobby and SiteMorse to make a website accessible. These robotic tools are run through a website to see if there are problems in the site or if there are any accessibility hotspots
The most famous of these automated tools is 'Bobby', an automated validation tool that analyses web pages and gives warnings about how pages appear in different browsers.
'Bobby', assesses a website using intelligent algorithms to identify problems in a site. Sounds to good to be true? Sadly it is.
Whilst 'Bobby' may be able to identify some problems with your site, it by no means will make your site accessible.
There are 61 checkpoints within the WAI- of these only 13 can be clearly tested using 'Bobby' (and even they may need to be checked manually). Of the remaining checkpoints, 27 of those can only be tested for the presence of a solution. In other words, 'Bobby' will be unable to tell if you are correct. Of the remaining checkpoints, 'Bobby' cannot assess, measure or suggest solutions.
Recently, we have noticed a vast increase in the number of sites carrying 'Bobby' approved banners.
This is because many designers believe that if their website passes an inspection by an automated tool such as 'Bobby', they believe the site is accessible and stick a 'passed graphic' on to their website. However, whilst 'Bobby' and other software tools may be good at identifying lines of code, this doesn't mean that the site is anywhere near accessible.
The problem is that automated testing software cannot identify the majority of issues in regards to accessibility. In particular, some problems will be overstated, whilst others are understated (for example- an automated tool will detect if an image has been assigned an alt tag but is not able to tell if the alt tag adequately describes the content of the image).
Automated software cannot check for appropriate meaning. For example you may have an image on your website of an apple and the alt tag reads 'This is a camel'- 'Bobby' will not notice!
Remember whilst 'Bobby' may be able to identify some accessibility issues or syntax code problems, 'Bobby' does not assess the navigation and the content of your website or how the content is presented.