A recent report by Gartner (Six Degrees of Failure
or Success in Portal Projects, 24th Sept 2002) identified poor
usability as one of the three reasons for 'B2C portal catastrophic
failure scenarios'. The other two reasons were that the system
wouldn't accept transactions due to requirements failure or
technological anomaly and incorrect information being displayed.
A requirements failure or technology anomaly suggests that the
project was poorly scoped or that the scope changed. No doubt at
the time the project manager was translating the requirements of
the various stakeholders into a tangible set of deliverables and
something got missed along the way. Although costly to the
organisation and annoying to users many organisations would put
this down to the complexities of running a large integration
project and deal with it in subsequent development under change
control. This assumes they survive the fall out.
Displaying incorrect information, such as prices the example
given by Gartner creates financial loss and is essentially lack of
attention to detail in the implementation. Disappointing yes, but
both this and the previous reason for failure are mistakes. They
are aspects of the project scope that were considered within the
implementation and simply incorrectly executed.
So why poor usability? First let us consider what good or bad
usability is. Good usability can be described as a situation where
the interaction between user and interface achieve the goals of
both the user and the provider. For example a user wants to find
information quickly and the organisations wants to provide it
quickly so that the user either comes back again or doesn't use an
alternative channel. A simple goal, and perhaps that is where the
problems begin.
When the portal project is being scoped and the implementation
process defined it is unlikely that usability is considered
overtly. The reason for this is that it is a soft and largely
misunderstood discipline and one, which many believe is common
sense. Many vendors explain that their applications have been
subject to a level of user testing during development and clients
are only too willing to believe that this is all they need to do.
However, and almost by definition, portals are bespoke
implementations where users will interact differently dependent on
the content. As the content is unique between applications there
is a strong argument for adopting a user centred design approach
which is becoming popular in website development.
User centred design (UCD) is an end to end process that maps
onto the development cycle and insures that usability is
considered at every step. The big advantage of UCD is that it not
only informs about the usability aspects but also it provides
crucial data about the features and functionality. This helps to
avoid over-development and wasted investment during the project.
Gartner categorise the most common type of failure for portal
projects as 'Teflon' portals, due to their lack of stickiness.
They go on to specify the main reasons for a lack of stickiness as
lack of clear vision and mission for the portal, failure to
understand the user requirements, lack of usability in the design,
lack of commitment to the content, no attempt to track user
behaviour and inadequate information architecture. All of these
are contained within the UCD approach and whilst they don't
guarantee a good project, if a specialist organisation is
employed, that know what they are doing, the chances of failure
are certainly minimised.
Luckily Gartner also explained that the shelf life of a portal
project is about 6 to 9 months due to the pace of change in the
technology, so at least organisations that got it wrong this time
will have another chance to get it right... quite soon.
Return to
newsletter |
