Trust, Once Lost, Is Not So Easily Regained

I sound like a Victorian paterfamilias warning his daughters to guard their reputations, don't I?

I've been thinking about issues of trust a lot lately. What prompted this particular post is a book review in the current (14 Sept) issue of Business Week magazine. BW writer Burt Helm reviews Women Want More, by two Boston Consulting Group partners (Michael J. Silverstein and Kate Sayre) with writer John Butman. The opening paragraph of the review includes the following:

"...[The] work unintentionally raises a pair of important questions: How rigorous does research for business books need to be? And does it matter if the authors have much that's original to say?"

The first question is what caught my eye -- I'd probably have skipped the review otherwise (as a woman, I'm tired of being "part" of titles like Women Want More). I was a market researcher in a previous life, and have had many conversations about what makes for solid research and what doesn't. For example: what's a valid sample size?

Academic research in the social sciences is often built on what corporate market researchers consider to be woefully inadequate sample sizes. A friend, who wrote a book in 2002 on women who earn more than their husbands (Breadwinner Wives and the Men They Marry) , interviewed some 60 couples before writing it. I remember her saying that her academic friends were all impressed that she had such a large sample, while her market research friends all complained that her sample size was too small.

In this case, the BCG partners say that they had surveyed more than 12,000 women. Which sounds great, except that questions raised by Business Week "have prompted an amended version of the book that could be in stores, according to publisher HarperCollins, within days of the first printing."

For example, some material in the book appeared to have been taken from Wikipedia, while other information came from an academic paper published nine years ago (without proper attribution). The authors claim that most of the "questions" raised by the magazine represent honest errors which will be addressed in the revised book. If they've done that, can we trust what they say they've done in the way of quantitative research? (For what the authors say about their survey methodology, click here.)

Helm's review (co-written with John Cady), closes by saying that "...[These] kinds of glitches, on top of the unimaginative analysis that run through much of Women Want More, add up to a disappointing performance by Silverstein and Sayre, two senior executives at a respected consulting firm."

If I were one of their fellow partners, I would be concerned about the effect that such a review (and such a book) could have on the overall reputation of the firm. BCG is not just a well-known firm; it's highly respected, and its opinions are valued and trusted. For which the firm can charge handsomely.

If I were a BCG client, I think I'd be wondering whether I really got what I paid for. Do I still trust them as I did before? And what, exactly, can they do to earn that trust again?