Task Based Digital Development

A person's behavior on the Web is highly goal-driven. People have something they want to accomplish, whether it's making a purchase, finding a recipe, finding a colleague or learning how to do something new. This model attempts to use the concept of user tasks to break the web design (or application design) up into a series of steps that have specific functions that contribute to efficient task completion.


The overall task of the user is identified and specified. This is often more complex them one may think. Useful techniques techniques are: surveys, interviews, focus groups, user observations. For an in-depth summary of analysis options read Step Two Designs excellent article.


The goal of this step is to simply listen to what users are actually doing on the website. This is where the reality of the web site use becomes evident. A key here is to establish a measurable baseline to task completion. This will allow any improvements or changes to be evaluated using metrics, not opinions. Useful tools here are a careful review of web statistics, eye tracking and formal usability testing.


This is where everything comes together and solutions need to be developed. Task processes need to be closely evaluated and modified. Content needs to be evaluated, edited or removed (gasp!). An overall redesign of the specific area can also be called for. The scope of changes will depend on the specific area and how the new design will influence the rest of the website or application. The best improvements will often be the most subtle for the user.


Here the various changes consider should be tested. This may seem obvious, but it is surprising how many times things are just incorporated because it 'seemed right'. Basic users testing  is one of the most simple methods for achieving this. In more advanced situations, eye-tracking and full usability testing can be useful. Testing should be designed to as actionable as possible, so clear conclusions can be drawn and fed back into the design process. Comparison to the baseline statistics conducted in the 'Listen' stage should be also done.  This will provide a quantifiable indicator if the design is improving task efficiency.


Based on the feedback from testing, adjustments can be made to the design (or not). In the ideal world, the task efficiency shall be improved, but in the real world, things may not be so straight forward. This is why the process is a circular one, and another iteration may be appropriate.

None of these steps are rocket science, but together can provide a framework and structure to the web and application development process that has the task and user in mind.

This is my method, what is yours ?