Make one to talk about: On excellence

NO ISSUE is more important to corporations today than innovation. And the key to that innovation just might be burning your planning manual . . . and getting started.

That's what innovation guru Michael Schrage said in a brilliant recent article in Design Management Journal called The Culture(s) of Prototyping.

'Effective prototyping,' he declared, 'may be the most valuable 'core competence' that an innovative organisation can hope to have.'

Mr Schrage observed that there were two types of organisations, 'spec-driven' and 'prototype-driven'. The former think a lot, write a lot of stuff down and eventually do something (usually elaborate). The latter do it, then think about it (once they have something concrete to think about).

Prototype mavens include 3M, perhaps the top big-firm innovator, and Sony, where design executive Nobuyuki Idei claims the average time from product concept to a rough working prototype is a shockingly brief one week.

'Prototypes are a way of life' in the most innovative firms, and 'an iterative culture defines the organisation, according to David Kelley of IDEO, the premier industrial designer. In such outfits, prototypes become 'the essential medium for information (transmission), interaction, integration and collaboration,' Mr Schrage adds.

At its roots, the cultural gulf between the quick prototypers and the rest is profound. Dan Droz of Carnegie Mellon University told Mr Schrage: 'The idea that you can 'play' your way to a new product is anathema to managers educated to believe that predictability and control are essential to new product development.'

I agree with Mr Schrage's conclusion that strong prototyping cultures produce strong products, but despite his artful prose, the use he makes is anecdotal. Enter Benham Tabrizi and Kathleen Eisenhardt of Stanford University's Department of Industrial Engineering and Engineering Management. They recently focused their research on the speed of product development, examining 72 projects from 36 companies in Asia, Europe and the US. They unearthed two primary and fundamentally different approaches to hastening product development. In the first, the 'compression strategy', they said the key to fast pace 'is squeezing together a rationalised product development process. The underlying assumption is that since product development is complex, it is best to plan ahead to streamline the process and then compress the remaining steps together.'

Alternatively, for the 'experiential strategy', moving faster 'simply by accelerating an existing (albeit streamlined) process . . . is insufficient. Rather, the underlying assumption is that product development is an uncertain path through foggy and shifting markets and technologies. Thus the key . . . is rapidly building intuition and flexible options.'

The authors offer and then test 10 hypotheses. Six underpin the compression strategy. The first is: 'More time spent in planning is associated with faster product development time.' The other five predict that speed will flow from: greater supplier involvement; more designers using computer-aided design tools; overlapping steps (eg concurrent engineering and production); multifunctional teams and rewarding teams for meeting schedules.

The seventh through 10th hypotheses assess the experiential strategy. Take No 7: 'More design iterations are associated with faster development time.' The remainder predict that development time will be cut by performing more intermediate tests, decreasing the 'time between' milestones, and relying on a high-ranking leader to focus the team's effort.

Working with corporate product developers, the researchers came up with precise quantitative measures for each hypothesis. Their findings: the calculative compression strategy was trounced by the 'just do it' approach.

More specifically, planning actually slowed the overall process; more use of computer-aided design also gummed up the works. Overlapping steps, greater supplier involvement and rewards for meeting schedules made little difference one way or the other. Among the variables in the compression strategy, only the use of crossfunctional teams speeded things up significantly.

Of the experiential variables, more iteration, more tests and more frequent milestones all shortened product development time significantly. Even a strong leader was a matter of indifference.

One must be wary of measuring too precisely such an inherently messy phenomenon. Nonetheless, Tabrizi and Eisenhardt's pioneering work provides a careful and systematic test of the intuitively plausible - and monumentally important - ideas presented by the likes of Michael Schrage.

TPG Communications