A recent article in the Harvard Business Review authored by Anne Mulcahy, the former CEO of Xerox (succeeded by Ursula Burns as CEO in 2009), makes a case for not pitting succession CEO candidates against each other and creating a dysfunctional horserace where critical executive talent may run for the exits if not given the golden ring. She points to one of the most public succession processes in history – the three way race to succeed the retiring Jack Welch at GE in 2000 among Jeffrey Immelt, Jim McNerney, and Robert Nardelli. Immelt got the GE top spot and both McNerney and Nardelli, each major contributors at GE at that time, left the company, presumably diluting the depth of GE's executive talent pool. McNerney become CEO for a short four years at 3M and then departed to take the top spot at Boeing where he remains. Nardelli was hired as CEO at Home Depot and then, after a rocky and unsuccessful stint (other than his egregious $210 million severance package), was hired as CEO by an already struggling Chrysler and proceeded to steer it into bankruptcy. He now works in private equity.

In contrast, Mulcahy started a dialogue with her Board about who might succeed her at the time she stepped into her job in 2001. Four candidates, including Burns, were identified and given a series of development assignments and visibility with the Board. She had regular conversations with each of the candidates and was realistic along the way about their status and, by implication, their chances of succeeding her at Xerox. A key element of her approach was to discuss other positions for them in the company if they didn't get the role of CEO such that, unlike GE, they felt good about their future potential and the contributions they could make to Xerox beyond CEO succession. Fortunately, Xerox had the time to develop her successor over time. This process played out over seven years, though Burns performed exceptionally well as she was given wider responsibilities and responded positively to constructive style and behavioral feedback she received from Mulcahy, ultimately emerging as the Board lead succession candidate in 2008. Notably, two of the other three CEO candidates are still with Xerox in key positions and the third has retired.

Mulcahy has some thoughtful points to make about CEO succession:

• Succession discussions and planning need to start a lot earlier than is generally comfortable for the sitting CEO.
• Succession should play out over 3-5 years at a minimum.
• Candidates for succession should be given critical development responsibilities and visibility with the Board.
• Clear guidelines need to be set with regard to timelines and expectations for the Board, the CEO, and the succession candidates.
• If a clear frontrunner candidate for succession emerges, that candidate should be groomed by the CEO, making sure that one person is absolutely the best fit for the job.
• Limiting CEO tenure to no more than a decade is sensible and prevents, among other things, sclerosis, entitlement, and the "killing off" of successors at the top.

I think you would agree that CEO succession is both tricky, critical, and complicated. Approaching it in the same thoughtful way other key business initiatives are pursued, regardless of whether a company or organization is public or private, is absolutely essential for the continued health and success of the enterprise.