Applying management science in succession planning

Otso Massala

Experienced managers know that building, retaining and leading your team well is an important prerequisite of reaching an organization’s objectives.

After hiring and firing employees myself, I find it appropriate to share what advice the business academics are giving. There is a whole discipline of human resource management dedicated for such thoughts. However this time, I took a look elsewhere for more accurate advice — namely to the quantitative discipline of management science.

We know that there are no easy answers. In management, we are always forced to make trade-offs to seek a balance. Real business management is more like balancing on a tight rope, and models should help you to understand the factors which need to be balanced, and how to find the right balance among them.

Some managers are concerned about succession, so let’s look into stochastic models suitable to guide decisions in succession planning. One major lever a manager can control, at least to a certain extent, is the amount of slag an organization maintains in terms of human capital.

Such slag can exist either in the form of employees waiting on a proverbial substitution bench, or employees with excess competence so that they can be promoted when a succession becomes desirable. Having slag comes with cost, but on other hand, not having it makes succession difficult.

Management science suggests a model called “Newsvendor,” and, when applied in this setting, it suggests the right amount of slag depends on the ratio between cost of unused slag and value of slag when succession is needed.

In this model it is essential to recognize that the right amount of slag is not equal to the expected amount of slag needed. In other words, it might be beneficial to build a big substitution bench, even if it is not all used, or it might be optimal to keep no substitute ready for succession, even if a substitute is likely needed.

How about another important decision: Searching for new employees? How long shall we keep looking? Searching is costly, at least in terms of work being insufficiently done while the search is ongoing.

Again we seek a balance: If we are hiring someone very soon, our search cost is low, but we might not get most competent employee. If we’re searching a long time, our search cost is high, but our likelihood of finding a great candidate increases.

Search theory suggests several insights for striking the right balance. The “secretary problem” model suggests a particular solution. Let some time pass so you get a good idea about the competency level of available candidates, and after that, hire the first candidate who is better than any of the candidates you have already evaluated.

This is quite interesting, as it suggests that you don’t jump into hiring anyone at the beginning of the process — no matter how competent they look — before you have built a substantial knowledge of available candidates on the market.

A rule of thumb for how much money is reasonable to spend on a search can be shown with mathematical precision. It should be about half of the standard deviation of a candidates’ value. In plain language, when candidates are very homogenous, it doesn’t pay off to make an extensive search. However, when candidates vary in competence, a high search cost is worth it.

Of course, the above ideas are hard to digest until they’re put into practice. However, I sincerely hope your interest is piqued. For more clarification, you can search online for “newsvendor model” and “secretary problem,” and we at Shippensburg University will happily guide you further.


Otso Massala is associate professor and director of Charles H. Diller Jr. Center for Entrepreneurial Leadership and Innovation at the John L. Grove College of Business of Shippensburg University in Shippensburg, Pa. Email him at