The firm is among the largest of the Big 8 accounting firms, with one of the biggest management consulting practices in terms of revenues and personnel. Founded in 1913, the firm has offices in major cities around the world, with headquarters in a large US city in the midwest. The research focused on the management consulting practice (MCP) division of the firm, whose primary business is the design and building of computer-based information systems. The MCP's New York City office consisted of three branches in Manhattan, Connecticut, and one in New Jersey, employing a total of some 750 consultants. MCP's main consulting practice is to custom-build application software for its clients by sending in project teams who remain and work on the client site for months or even years to produce a computerised information system. Building software for clients is a highly complex, knowledge-intensive activity that is fraught with uncertainty. Over its history, MCP evolved two innovations to manage its internal knowledge and to cope with the task-related uncertainty -- a standardized system development methodology, and a suite of computer-aided software engineering (CASE) tools. Orlikowski explains how MCP's standardized methodology ('Modus') came to be:
When the MCP division first started developing information systems for clients some thirty years ago, the only written "knowledge" of systems development in the Firm was extracted post hoc from the documentation generated for each project. These so-called "client binders" served as the Firm's information expertise about the systems development production process during the initial years of the consulting practice. As the practice grew, some attempt was made to systematize this varied and highly idiosyncratic knowledge. During meetings partners would review the project documentation, trying to extract general procedures, and identify the common factors that made some projects successful, others mediocre, and still others failures. Over time these generalized "rules of thumb" became more extensive and more sophisticated as the MCP division gained more experience. Eventually the informal guidelines about how to run a successful systems development project and what factors constitute good systems practice, evolved into the formal, standardized methodology that "Modus" is today.Thus, MCP's system development methodology grew out of the daily activities of consultants working on projects. By analyzing and reflecting on this practical know-how, MCP partners were able to generalize and formalize their experiences into a methodology which specified the sequence of tasks to be performed at each stage of the system development life cycle, and defined the standards for documentation, control, scheduling, and project estimation. The institutionalization of a standard methodology was also in line with the Firm's "one-firm" philosophy which required all partners to follow a common approach in the ways they dealt with clients' problems and communicated about them. From its earliest days, the Firm had espoused the policy of speaking with one professional voice, and abiding and upholding the official viewpoint of the Firm.
(Orlikowski 1988, p. 166-167)
The formalization of the Modus methodology made possible the next major innovation in MCP's consulting practice -- the introduction of a standard set of computer-aided software engineering (CASE) tools, which MCP called "productivity tools," to support and implement the methodology. This integrated tool environment included software to capture ongoing documentation of the new system into a data dictionary; project estimating aids; the project control system; screen and report design aids; data and program design aids; installation tools; and prototyping facilities. The tools "implemented the standard software engineering design philosophy and project management method articulated in "Modus." In fact, "the tools were deliberately based on the methodology as it was recognized that production technology logic had to be compatible with that of the production process, else inconsistency and discontinuity would disrupt the systems development process." (Orlikowski 1988, p. 183) The use of the methodology and the CASE toolset was mutually reinforcing. Since the tools were based on Modus, their use ensured compliance to the methodology. At the same time, Modus was also constantly being updated to better reflect the tool environment. This reciprocal interdependence characterized the co-development of the tool and methodology. Furthermore, the use of computerized tools enhanced the aura of professionalism in the consultants' normal work activities:
Tools render an image of a room of consultants all seated in front of their personal workstations, all bent over their keyboards, flashing through complicated-looking screens, performing sophisticated cut and paste procedures, and all done to the accompaniment of the reassuring whir of the disk drives, the steady tapping of keys, and the regular sigh of the laser printer emitting its professional-looking documents. It certainly looks industrious.As a result of employing the tool-methodology, MCP reported savings of 30% up to 50% in code generation, and an elimination of between 50% and 70% of the systems installation phase. The use of tools "dramatically" increased MCP's profitability, and allowed it to reap the benefits of operating economies of scale. Competitive position has been improved by enabling the firm to bring the price of its services down, to lower its bids on contracts, to go after larger projects, and to increase the income contribution of each partner.
(Orlikowski 1988, p. 403)
Besides productivity and profitability gains, there were other important and somewhat surprising benefits derived from the use of the CASE tools and methodology. As a professional services firm, MCP is expected to provide customized solutions to each of its clients. Indeed, each client will have its own data processing environment that made customization mandatory. Although this might appear incompatible with MCP's standardized production process, the software utilities in the CASE toolset were in fact relatively easy to modify so that they could work well with a client's hardware and software configurations. Each client company therefore was provided with tools that were customized to the project and technical characteristics of the site. At the same time, since the underlying process logic may not change that much from project to project, MCP is able to reuse significant portions of their development outputs:
With the deployment of productivity tools it is able to adapt a set of system designs and documentation developed for one project for use in selling a similar system to another client. By being able to customize the visible features of the design to the potential client's needs while leaving the essential logic of the systems design intact, the Firm can exploit the power of the tools in saving time by not having to design another system or generate new documentation. It can use the logic of the existing system customize the labels, change the screen and report headings, and change client references in the documentation, and have a new comprehensive systems proposal to present to a potential client. And if the client accepts the proposal and the project gets underway, many of the tools, shells, macros can be directly transferred to the new project site, hence avoiding reinvention of the wheel.The standardization process thus extends beyond tools and methodology to the "industry standard solutions" that MCP is able to offer to its clients who receive tested solutions that have been optimized for their local computing environments. (The success of its integrated CASE tool environment prompted MCP to sell the toolset as a generalized productivity tools package to the clients themselves and to other data processing companies.)
(Orlikowski 1988, p. 352)
[Based on Orlikowski, Wanda J. 1988. Information Technology in Post-Industrial Organizations. PhD, New York University.]