Michael Schrage
Harvard, 1999
ISBN: 0875848141
Buy it from Amazon.com
Buy it from Amazon.co.uk
In this book written for business executives, MIT research associate Schrage tells the story of companies that have had the good sense to realise that playful exploration is the best way to learn. The thesis of the book is simple: a lot of cheap rough prototypes are much more effective than a few beautifully-finished ones.
You can take this further: the very worst sort of development approach for innovative products is the traditional "big bang" or "waterfall" life-cycle, where (to mix metaphors hideously) you stake everything on one roll of the dice. The big bang is of course an ironic name: you find out what you think the requirements are (once); you specify the whole complex system; you design, test, and manufacture or commission it, and then, Pow!, you launch it on an unsuspecting market or customer. And, of course, it's the wrong product: the requirements were not correctly understood, or the analysts and designers did not perfectly trace back to the requirements, or of course the world has gone round the sun a few times and the requirements have moved on. Basically, you have built just one, much too slow, much too costly prototype, and the results are predictably dreadful. (Actually, not even the most dyed-in-the-wool systems engineers do not actually advocate the waterfall approach, but we don't want to spoil a good story.)
Instead, Schrage argues, you should use plenty of models to communicate what you thing might be a good idea (that's pushing ideas at people), and at the same time to let them communicate to you what they actually want (that's the market pulling products from you). Gosh wow, I can hear you saying: he's going to tell us that businesses ought to listen to their customers and give them what they want in a minute. This story is indeed as old as the mountains, yet somehow it seems to need to be told and retold.
Schrage's message is the business equivalent of the "agile" school of thought pioneered by Kent Beck, championed by Alistair Cockburn, and now becoming an accepted practice in the software industry, along with the use of brief stories in place of requirements, eg as argued by Mike Cohn.
The book is not without its defects. It is somewhat sloppy, verbose, and repetitive. You can hear its thoughts coming along pages before they arrive. It risks over-generalising from its examples. Is it true, for instance, that what is good for businesses big enough for Wall Street and the London Stock Exchange is also right for small and medium sized businesses? Maybe they can't afford to play around. And maybe the big boys don't spend much time in the sandpit, either, despite a few well-publicised successes.
However, Schrage is certainly right that prototyping and modelling "with spreadsheets" and other means is valuable, and if well managed can enormously increase the chances of developing good products. The book is well worth a look for the examples it gives, and if you need to persuade a recalcitrant manager that you really ought to have time and resources for prototyping and exploration of ideas with clients, this book might just do the trick.
(c) Ian Alexander, 2008