Introduction
Traditionally, internal systems—such as financial systems or employee
data systems—have had few, dedicated end users who are experts in
the domain. Whenever that is the case, there is a lot more tolerance for
interfaces that are not self-explanatory and the user experience plays
a relatively small role in the overall success or failure of the system.
However, now that enterprise applications are moving into “self-service”
for a wide variety of essential data entry functions and approval processes,
the interface becomes a more critical ingredient.
User-centered design processes are not a guarantee of success. However,
not using them is a virtual guarantee of failure, and user-centered planning
and design processes are not an optional luxury for internal systems groups
anymore. However, the degree to which various research and design activities
are carried out in a project ought to depend on the challenges at hand.
If we recognize that not all of our projects are going to get equal time
and attention in the research and design phases, then we need a tool to
decide which projects require more attention. The score sheet below was
developed specifically to assess the degree of risk associated with failures
to involve the end user in internal systems deployment projects.
Please note that the scoring does not take into account the demands and
vicissitudes of the market place that apply to products and services developed
for external customers. To use this risk assessment tool for such projects,
the scoring would have to include such factors as competitive advantage
and time to market.
UXD Risk Factors
The need for a user-centered design process for internal applications
increases with the following risk factors:
- the number of end users
- the degree of difference from existing processes
- the importance of users’ ability to use the system independently
- the complexity of the workflow
- the frequency with which they will interact with the system
- the degree to which end users are domain experts
Scoring
Some of these factors have more influence than others, so they are weighted
in the scoring. A Score Sheet Template
(PDF) is provided by way of an example.
| Factor |
Scoring |
| Number of users |
0 = small end user group (< 15)
3 = medium end user group (15-50)
8 = large end user group (> 50) |
| Difference from existing processes |
0 = no difference
3 = some differences
8 = very different |
| Importance of independent task completion |
0 = impact of task completion failure is minimal
2 = failure to complete tasks will result in significant frustration
on the part of the end user
5 = users are likely to abandon the system if they are not successful
right away |
| Complexity of workflow |
0 = very simple (workflow is linear and has few steps)
3 = some complexity (workflow has many steps, which can be completed
in different ways)
8 = very complex (there are many steps, many flow configurations,
and many points of entry into the flow) |
| Frequency of use |
0 = use regularly (at least once a week, often more than that)
2 = use a few times a month at most
5 = use a few times a year |
| Users’ domain expertise |
0 = all users are domain experts
1 = users are generally familiar with the domain
3 = users have no domain expertise
|
Interpreting the Score
Low-Risk Projects (< 10)
When the overall score is less than 10, it is safe to assume that the
UXD risk index is low. That is, projects may fail but the likelihood that
it will be because of a poor user experience is relatively low. Consequently,
the investment in various user-centered design practices will have a relatively
small return. Benchmarking is in order, even if no other research involving
users directly is completed, to confirm that the project team’s
judgments about the challenges were indeed correct.
Medium Risk Projects (10 - 20)
When the overall score is between 10 and 20, there are significant risks
to the project from a failure to understand the users’ context.
It is highly advisable to do some form of upfront user research as a foundation
for design work and to include a prototype validation in the design cycle
to confirm that users were properly understood before the project goes
into development.
High-Risk Projects (> 20 or a High Score on Factor 2)
High-risk projects are either projects that have a high overall score
(> 20) or that score an 8 on Factor 2 (Difference from existing processes)
even if the overall score is low. Such projects call for extensive user
research upfront, preferably using early prototyping, in addition to at
least one but possibly more design validations. It is highly advisable
to use personas and scenarios throughout the design process.
More Information
Score Sheet Template (PDF)
Thoughts?
marijke@interfacility.com
|
