|
|
ABSTRACT
The Claris Interface Design Group is an organization built from the ground
up to help Claris Corporation define new levels of usability for its Macintosh
and Windows products. This overview offers a retrospective of lessons learned
through the creation of the organization I created.
KEYWORDS: Organizations, usability testing, human factors, visual
design.
SETTING
Claris Corporation is a wholly-owned subsidiary of Apple Computer and is
the leading publisher of Macintosh software. Some of the notable Claris
products include FileMaker(a data base product), ClarisWorks(an integrated
package), and Amazing Animation(a kid's cartoon maker).
Claris is firmly focused on the issue of interface design and usability.
Usability isn't a key market factor, it is THE key market factor in Claris'
mind. This translates into a management structure that understands and supports
interface design from the top down.
I was hired by Claris to help it revive its once strong human interface
effort. The group had shrunk from a team of five to a team of one employee
and one contractor. Claris believed strongly in usability and was willing
to invest in it. My job was to figure out how to proceed.
GETTING TO THE BOTTOM OF THINGS
I approached the problem of building this organization in the same way I
would in designing a product: I went out and talked to the customers. In
this particular case, the customers were managers and organizations within
the company. I set up conversations with engineering and marketing managers
throughout the company and I discovered the following issues:
· The group was loved by some and hated by others. In several
cases, there were personal relationships that kept the communications going
between engineering and the designers. Where personal relationships were
missing, the interaction was a disaster.
· Some product managers did not want to use the human interface
resources because they felt the designers were not dedicated to their efforts.
Specifically, the group was set up as a central pool of people that assigned
designers to individual issues as they popped up. The product teams wanted
the designers to work with them throughout the entire life cycle.
· People wanted this organization to succeed.
In addition to these findings, I observed that the usability testing conducted
previously, was very limited. Field studies were non-existent.
Given all this, it was clear that there were challenges on all fronts.
MAKING THINGS BETTER
Talking to various people around the company was enormously important to
me because it gave me a look at was wrong, but more importantly, what was
right. One of my paramount concerns was to make sure I didn't break what
was already working.
One of the positive attributes of the previous group was the way it was
set up: The group was a centralized body within the Product Engineering
organization with its own budget and was given the ability to request headcount.
This allowed the group to function autonomously for the greater good of
all the products. This also allowed the group to pool resources together
to develop "future-focused" ideas independent of any one product.
ORGANIZING
To begin with, I moved away from the "central pool of talent"
approach and assigned at least one designer per product. This immediately
brought up the question of whether the designers should be distributed throughout
the organization and therefore remove the need for a centralized organization.
To me, the centralized organization had some clear advantages over a distributed
model and deserved preservation. There was central accountability as well
as the ability to pool resources. For Claris, this would mean greater efficiency
and more creativity applied to each of the problems. For the designers,
a centralized group offered not only the possibility for career growth,
it offered a sanctuary from the difficulties often encountered by designers.
I created a centralized organization from which designers were assigned
long term projects. They took everyday direction from the product teams
but reported into my organization. This allowed members of my team to be
there when the "hallway decisions" were being made and still
have the benefit of working with a larger body of designers that they could
draw upon for inspiration or designs. In this fashion, the organization
took on a hub and spoke nature: the spokes were interaction designers working
directly with each product group and the hub was management, usability testing,
and visual design.
In order to improve the groups maligned image, I created an identity for
the group which included a logo, typeface selection for documents, etc.
Usability testing and field studies were made part of the design process.
WHAT WORKED THINGS LEARNED
· The hub and spoke model worked well as long as the right conditions
were in place. It was important that the product groups wanted the designer
around. In addition, the interface designers needed to be mature enough
to put the needs of the product ahead of their own ego.
· The central group created an environment where designers could
spread their wings and fly. Going forward with very high ideals and lots
of imagination, the group was able to produce designs that went beyond what
could be expected from 10 people.
· I set out to recruit designers from a variety of backgrounds.
My aim was to create discussions and viewpoints that would not have come
about if everyone was a computer scientist. The backgrounds included:
· Software Engineering
· Psychology
· Industrial Design
· Fine Arts
The diversity of people created innovations, ideas, and usability. It was
also made work fun.
· Usability testing turned out to be an enormous success. Where
nothing had existed, usability testing became a standard part of the product
cycle. As the group matured, product groups took testing for granted: that
was a big compliment.
· Field studies were conducted by the group. They wound up influencing
design decision, marketing directions, and creating whole new products.
WHAT DIDN'T WORK
· Although having a group logo and other trappings of an identity
seemed like a good idea, it created the impression that we didn't belong
to either engineering nor marketing. This tended to isolate the group.
· Usability testing was an extremely visible portion of the group.
It was a success but in some ways it became too great a success. Some people
in the company came to believe that all the group did was usability testing.
· In spite of all the attempts at educating people, many marketing
managers never quite understood what an interaction designer was. Thus,
it was difficult to have them support headcount requests.
· Because the group was centralized, it was important to constantly
create growth opportunities for the designers. One way I did that was to
sometimes accept exciting work that we weren't quite staffed for. This created
crunches that were usually taken care of in the next round of hiring. My
intentions were good but they had some negative side effects. The first
was the perception of empire building. As the group grew to deal with these
additional demands, some began to interpret my actions as being politically
motivated. The truth was, no ambition was involved. The second problem was
that the group wound up being spread too thin. When the resources stopped
coming, the group was caught short and was not able to deliver services
with the same level of quality.
FINAL THOUGHTS
To create a successful HI organization you need the following:
- Highly trained and motivated designers
- Management that supports your actions
- Product teams that want your contribution
- Money
Proceeding without one or more of these will make life difficult at best.