The Seven Principles of User Experience
This is an extract from Delivering Data Analytics.
Create an effective report or dashboard for the end user’s experience of data
In Chapter 1, user experience was defined as the perceptions and responses a person has when interacting with a system, in this case a report or dashboard. When a user is interacting with data, the experience must be understood, designed for and then measured.
In the field of user experience, there are generally seven factors that contribute to the experience. These were popularized by Peter Morville and have particular application to the end user’s experience of data:
Why would anyone adopt a dashboard that is not useful to them? If it does not have some benefit to using it, then it is improbable that it would be become part of their workday. Indeed, they may have a spreadsheet that is more useful to them and a new dashboard would have to be significantly more valuable to overcome the inertia of switching technologies.
Whether or not something is useful is subject to interpretation and perspective, which must be considered. What a BI developer deems useful may not be so for the end user.
A usable dashboard is one that enables the end user to effectively and efficiently achieve the objectives the dashboard was designed for. The dashboard should become a tool for action and progressing work aligned to value. It should only have what is needed to fulfill this goal and no excess.
The temptation can be to have an array of bells and whistles like having as many filters as there are dimensions available in the source data. While it may seem effective, it may not be efficient, and hence will not be as usable. This is not to say it is all or nothing; there is certainly a gradient to usability, and it can improve through iterations as more user feedback is gathered.
Findable or discoverable means it must not only be easy to locate the dashboard in question, but also it must be possible to answer the questions within the dashboard. It sounds obvious, but if the dashboard cannot be found then it will not be used and, hence, not adopted. If it is not adopted, then people will not be acting on the information in there and the whole purpose of the effort fails.
Many professionals miss this point. It comes downs to a question of ownership and who is responsible for making a dashboard discoverable. While that owner is somewhat more obvious for the content within the dashboard, it is less so for the dashboard at large.
Credibility directly relates to trust, especially trust in the data being presented. As the adage goes, it takes time to build trust, but it can be lost in seconds and can be doubly hard to regain. When presented with a new experience, credibility with the end user must be a priority.
Consider the situation of a person using their spreadsheets. It may not be the best experience, but they trust it; it is credible to them. It is the credibility benchmark for their acceptance. Therefore, any new experience will be compared to it. If that new experience is not credible then they are likely to revert to their previous experience.
However, credibility is not just about trust in the data; it is also about the dashboard being fit for the intended purpose. The experience is credible for that specific end user. Should the user think the dashboard creator is trying to manipulate or deceive them, then the relationship will be damaged.
There are many tools and means to deliver insight. A spreadsheet is a very popular interface, but contrast it to a beautifully designed, interactive, multi-device dashboard. If it did everything the spreadsheet accomplished but much more, it would have greater desirability.
Take the example of a regular plastic shopping bag and a Hermes handbag. They are both useful, usable and findable but the Hermes is definitely more desirable. The plastic bag has a certain practical appeal, and it depends on usage; one probably would not carry shopping in a Hermes handbag. Nonetheless, the plastic bag is not desirable.
Desire is drawn from a composite of brand, identity, appearance and emotional design. People tend to brag about items that are desirable. The more people talk about it, the more desirable it will become to a wider group of people.
Accessibility in analytics and BI is both greatly misunderstood and underutilized. Accessibility relates to enabling a positive experience for the whole population through the spectrum of disabilities including impaired vision, hearing loss, motion impairment and those with cognition challenges.
The challenge is that designing for accessibility is not at the forefront of the mind of the designer for several reasons, but two of the leading ones are a lack of awareness and the incorrect perception that there is not a critical enough mass of people with disabilities to make it worthwhile designing for. Whatever the justification, it is unacceptable. According to the CDC, one in four people in the United States have some form of disability. While not all of these have a disability that would affect a dashboard interaction, it is certainly a very large percentage that, if one wanted to make the argument, justifies making an accessible experience.
Perhaps a surprising outcome from designing for accessibility results in a more thoughtful and intuitive experience for all users, not just those with disabilities. In the modern era it is critical to design for accessibility. So much so that not doing so can cause risk litigation. Indeed, there are legal practices that make a living from suing public-facing companies for not making their customer interactions accessible. Some companies mandate adherence to accessible standards to ensure awareness and proper implementation.
Dashboards without value are legion. They are the norm in the enterprise. Low-value dashboards are easily created, little thought is needed to create them and they are most easily produced in siloes. So, it is of little surprise that value is the key driver for the factors of a positive user experience.
Value can be interpreted on two levels: value for the end user, and value for the business. Ideally, both are true, but they are not always so and perhaps not always necessary. A dashboard can be valuable to the business but have little value to the end user. Conversely, a dashboard can have significant value to the end user but lack value to the business.
Without value in the equation, adoption will suffer. All the other aspects of a great user experience can be met and there may be fleeting adoption but, without value, the dashboard will fade into the background. To some, value can be the time-saving of automating four hours of manual data preparation into a dashboard that does the work automatically. To others, it might be being able to use data to identify customers with the highest potential of being cross-sold a new product. There are many paths to value. Regardless of which is chosen, value is the driving force.