CHI 97 Electronic Publications: Organizational Overviews
HCI at Trilogy: Bringing the Design Stance to a Startup
J. Epstein, E. Loh & J. Marks
Trilogy Development Group
6034 W. Courtyard Drive, Austin, TX 78730 USA
+1 512 794 5900
http://www.trilogy.com
jeremy_epstein@trilogy.com
J. Lilly
Advanced Technology Group
Apple Computer
1 Infinite Loop, Cupertino, CA 95014 USA
lilly@cs.stanford.edu
ABSTRACT
A successful startup in the arena of enterprise software, Trilogy Development Group began experimenting with HCI as a means for improving user reactions to their products. Two years have passed since the first experiments; in that time an entire HCI group was created and has subsequently become a respected and critical component of Trilogy's development process, as well as taking some responsibility for providing a vision for Trilogy's future. This paper chronicles our experiences in bringing the "design stance" to Trilogy.
Keywords
Organizations, HCI, user interface, design, interaction design, enterprise software, startups
© 1997 Copyright on this material is held by the authors.
THE SETTING
Trilogy Development Group (http://www.trilogy.com), in Austin, TX, is a 250-person, privately-owned software company specializing in enterprise-wide deployments of sales and marketing software, "Trilogy is out to change the way people buy and sell." It is an extremely young company: the CEO is 28 years old, just slightly older than the average employee. In many ways Trilogy is the prototypical software company: a group of people who care deeply about the technology and are willing to do whatever it takes to get it done.
The HCI group at Trilogy investigates three general classes of problems at Trilogy: Software for Configuration, Software combining the sales pitch with accuracy of the engineering congifurator, and maintenance tools.
- Software for configuration. The core of Trilogy's business revolves around enabling people with little knowledge of how a complicated product is engineered to sell and deliver complicated products to customers without making mistakes.
- Software combining the sales pitch with accuracy of the engineering congifurator. "Interactive media" is the catchall term for tightly scripted presentations. "Applications" might be defined as the flexible, maintainable tools that create quotes, engineer products, or track orders. HCI is trying to incorporate the visual dynamics of interactive media into applications for configuration that can be maintained easily.
- Maintenance tools. An extensive series of constraints, object structures, media archives, pricing, and order status information must be maintained by Trilogy consultants and clients. HCI tries to make the behavior and structures of trilogy's proprietary maintenance applications understandable.
Today, Trilogy's HCI group totals seven: four from a background of graphic design, and three from computer science/interaction design. We provide advice, design, research, and prototypes that support the rest of Trilogy. The group usually develops prototypes to a high degree of fidelity, with the burden of implementation and maintenance resting with other organizations within Trilogy.
THE FIRST HIRE
The beginnings of HCI at Trilogy can be traced to late 1994, when it became apparent that our clients' problems and our software solutions were becoming increasingly complex. At the time, the perception was that the appearance of Trilogy software could be improved (especially for the company's sales needs), so a single graphic designer was hired to improve the presentation of existing projects. From the company's perspective, the operative word in the job description was graphic--it was believed that a superficial retouching of some key screens would be adequate. However, after this individual rotated through several different projects, it became apparent that problems ran deeper than just visual changes; Trilogy required the rethinking of its entire design philosophy. In this way, graphic design planted the seeds for the establishment of a complete HCI group. More importantly, it planted the seeds of a design methodology--what others have termed the "design stance."
STARTING THE GROUP
An individual was chosen to manage and assemble the group. He embraced the deign methodology that was in place before the group was officially assembled. (1) HCI must be managed as a distinct group, as opposed to several HCI staff members distributed in various parts of the company, (2) The group must maintain autonomy in hiring more staff, and (3) The HCI group should report directly to the head of the company and not to a VP of development or engineering. HCI was given the freedom to define its own role within the company. This decision was supported at the highest levels of management.
BUILDING THE TEAM
Due to timing issues, the first hires after the two group founders were three summer interns. This turned out to be extremely fortunate; while the two full-time members of the group focused on the existing interaction problems, the interns were able to spend their time thinking outside the boundaries of day to day work. Their creative work formed much of the basis for our group's work over the following year, and this methodology of having interns as our own research department has been adopted for all interns since. Following the hire of the interns, two new people joined the HCI staff in August of 1995: a computer scientist and a graphic designer
GROWING PAINS
August 1995 also brought our first drink from the proverbial firehose. The four members of the team began to assist four multi-national clients, in addition to working on products in our development organization. Each person spent the next few months learning and understanding his client's problems, trying to come up with creative solutions. It was a time of adjustment and refinement of the application of HCI, but by the end of 1995, we had each developed good intuition for our individual clients and industries. Armed with domain knowledge and credibility our clients were excited about the expertise Trilogy HCI brought to the table. Experiences with these clients gave HCI the insight which would later drive the development of future products. The fact that we started with consulting was no accident--while we had an initial credibility problem with our developers (we suspect this to be a common phenomenon), our consultants were the people at Trilogy who constantly dealt with users and customers and were acutely aware of the need for a better understanding of their problems.
DEFINING OUR PROCESS
From the beginning our process and team methodology was predominantly client focused. In order to maintain a consistent face at a client site, one member of the HCI group "owned" each project, and was responsible, from an HCI perspective, for doing whatever it took to get the account finished. This included securing time from other members on the team, when their particular skills were necessary.
THE REORGANIZATION
Reorganizations are traumatic for any group, and ours was no exception. During the summer of 1996 Trilogy moved from a functional organization into its current industry alignment, with five separate cross-functional departments focused on particular industries, such as the automotive or computer industry. At the time of the shift, HCI had only five members. There were strong pressures to divide the group and assign individual members of HCI to each Industry. After some discussion with management, the decision was made to keep HCI intact as the only technical functional group (other than training). Keeping the group a single entity has enabled us to leverage solutions across different industries, and provide a consistently high level of quality across the company.
YEAR THREE AND BEYOND--What's Next?
Looking ahead to year three, we expect to continue to evangelize and teach HCI and build the critical mass required to be a company focused on providing great design. As we grow in number to a reasonable size (probably 15 by next year), we've begun to plan how to build mini-groups to support each industry. While HCI groups of one member don't make a lot of sense, we expect HCI groups of three to function well, given the right makeup (for example, one graphic designer, one interaction designer, and a utility player who can act as a graphic designer, interaction designer, and evangelist).
CONCLUSION
To sum up, we've come to understand a few things that have been obstacles to our success, as well as some of the ideas that have helped us overcome them.
Major Obstacles
- Establishing credibility. Because graphic design was so effective as a leverage point for us at Trilogy, we've had to continually establish credibility, in ways both technical and business-oriented. Hiring talented computer scientists with HCI backgrounds helped that image.
- Dealing with change. Internet startups are at the extremely quick end of the change spectrum. Because of daily changes in the market and the company, we've had to stay particularly nimble and responsive.
- Designing under severe time constraints. Again, the startup mentality is that everyday you don't release a product is a day that you lose potential revenue. We've had to respond by finding ways to be effective within tight development cycles.
Success Factors
- Pick your fights. It's impossible to fix everything at once. We started by picking small, specific problems and executing effectively, then leveraging the resultant credibility to work on bigger projects.
- Work together. No matter what else happens, and how else a group works, the people is what make or break the group. We tried to be careful and smart about who we hired; we were always conscious of the chemistry of the team, and that sensitivity has enabled us to build a real team.
- Think big. We've believed all along that we would change the way that Trilogy is "changing the way people buy and sell."
ACKNOWLEDGMENTS
We thank a number of people, but most importantly, the other members of our group, as at the end of the day they're why this is such an outstanding place: Lynn Pausic, Perry Arnold, and Amanda McCoy. And the letters J, Q and the number 3.
CHI 97 Electronic Publications: Organizational Overviews
