Graphs are a great way of giving clients and developers a quick overview of the usability of a design and where problems lie. They summarise large amounts of data into a quick message.
Task success rates and the severity of problems give a brilliant overview of the usability of a design. I recommend reporting these for all your usability evaluations.
These two bits of information - task success rates and the severity of problems - can be neatly combined into one graph. This is great - it packs a huge amount of your key findings into one page. This is how you do it: Within each bar you report whether the task was completed with no problems, with significant problems, or failed/abandoned. Here’s an example:
This type of graph quickly helps show where redesign efforts should be focussed. In the graph above you can see that 20% of users could not get through the registration process - and a further 30% had significant problems. Such a poor design will have a significant effect on the bottom line, and should be fixed as a matter of urgency.
When creating these types of bar graphs don’t use too many categories (colours). I’ve used three here: green, orange, and red. Any more, and it will be more difficult for readers see the big message.
If you’re using Nielsen’s 5-level severity rating system, you probably shouldn’t show all levels on this graph. Instead combine severity levels into one category: You could combine passes, cosmetic problems and minor problems into the green category; major problems could be represented by the orange category while catastrophic problems (task failures) could be shown in red.