Contingency Plans

As a manager, or anyone responsible for a team, you have to be worried about continuity. An unexpected departure will always slow a team down, but when it happens, it's up to the manager to respond quickly with a compensating adjustment, lest the team be hobbled for a lengthy period. Often the difference between a decisive response and a haphazard scramble can hinge on if the manager has a ready backup plan they can fall back on. Such plans are created with the hope that they'll never have to actually be used, but woe betide the manager caught unprepared by someone unexpectedly handing in their notice. If you're a manager of other managers, your contingency plans can start to get quite complex, but given the larger scope of your responsibilities, they're even more critical to have.

Moreover, voluntary departures are not the only reason you will want contingency plans at hand. Promotions or requested role changes can have similar effects, and while you'll be able to see involuntary departures coming, they will still require the same sort of planning around. Even more complex plans may deal with the effects of expanding teams (growth) or contracting them (layoffs). Regardless of what is driving the need for organizational changes, I use a similar analytical framework.

My own contingency plans are rarely fully specified; rather, they consider the gaps created by a departure and the shape of what a solution should look like. When I plan for resiliency in a complex software system, I consider each sub-system individually, think about what their likely failure modes are, and then specify how each related sub-system should react to that failure to minimize overall system impact. So it is with the complex system known as a "team", and the even more complex "organization".

With teams, first consider the impact of the departure of each particular member. Obviously, the departing person's current projects slow down, or even stop entirely. But do they also leave behind a knowledge gap? Do you lack redundancy with respect to any of their critical skills? It's especially important to identify the single points of failure you have within your organization; these will keep you up at night.

But my solutions are rarely as simple as "if Jeff leaves, I'll have Cindy take over his role". Instead, I consider everyone within the organization, and think about what degrees of flexibility they have. What other projects could they work on if needed? Are there any other roles they could fill? Could they step up into a larger role if necessary? How happy would they be with a change? Would they welcome it? Or have they moved recently, and need stability for a while? Are they on a project or a team that cannot afford to lose them right now? Once I've mapped out what degrees of freedom my organization actually has, then I can start to evaluate whether replacing Jeff with Cindy is the best possible response, or if something else might be preferable.

Whenever I'm considering organizational changes, I'm balancing several different goals in tension with one another. In particular, I want to:

Make sure my most important projects and initiatives are adequately staffed and remain on track
Put people in a position to succeed -- don't give someone responsibility they're not ready for
Use the opportunity to advance the careers of people who are ready for it
Minimize the disruption to everyone else

System adaptations don't happen in isolation. Every change will have second-order impacts, and possibly third-order or more. Thus, solving for a gap in your organization involves not just plugging a hole, but in minimizing the impacts on wherever you took the plug from.

Furthermore, you can't forget that people are not plugs. They have their own ambitions and attachments, and will need to feel valued and challenged and comfortable to perform their best. For everyone who reports to me, I want to have had a conversation beforehand with them, usually in vague terms (don't promise a promotion!), to get an idea of what sort of roles and projects they'd be interested in. Knowing this gives me a sense of which potential moves they not only could do, but would also satisfy and engage them. This is the sort of conversation that needs to be refreshed every so often, because people change, and their answers today are not necessarily what they were two years ago.

In the end, is what I have even a "plan"? Perhaps it's more like a plan-generating framework, one that can be activated when necessary. The framework looks something like this:

Understand the organization's current structure, and where everybody fits into it.
Understand all of the organization's priorities -- which initiatives must be supported fully, and which can we back-burner for now if necessary?
Understand all of the people in the organization -- how flexible is each person to move into any other role or take on additional responsibility? Are there people who are ready for a promotion, or who would benefit from a role or team change?

Having assembled all this knowledge into a coherent structure, you can then just specify a particular organizational gap, solve for the moves that fill it with the least amount of pain, and the result is your just-in-time contingency plan! Simple, right? Perhaps an example would be illustrative.

Let's say I'm responsible for three teams, and the manager of one of them has just handed in his notice. Here's what I'm considering:

If the team has a member who I think could be a good prospective manager, I'll ask them if they're willing to step up and take over as manager. They should be experienced and well-respected enough that the rest of the team will accept them as a manager without reluctance.
If another team has a prospective manager who might be ready to step up, I'll consider if the current manager on that other team would be willing to move over. Being a manager for the first time is challenging, and that challenge is much greater if you don't already know the people and the project. Two moves are more disruptive than one, but the shuffle may be worth it.
I may have an experienced manager on one of the other teams who is capable of taking over a second team. Such a person should be committed to a management career path, because they won't be getting much individual work done for quite a while, but for a strong manager who is ready, this is an important opportunity for further growth.
If I have multiple candidates to fill the manager role, I need to balance organizational disruption with who is most ready for the position. If candidates are roughly equivalent, I'll basically always prefer the one currently on the team needing a manager. However, if the candidate on another team is significantly stronger, they may become discouraged if they're passed over for a potential management role for someone not quite as advanced.
Since it's critical that team members work well together, with any move I'm always considering whether there's any potential pair of personalities that might be better left separated onto different teams.
Timing can also play a role. There may be a "right" move to make, but I can't make it until some particular project ships, or the person I'm planning to move otherwise becomes free to shift roles. I may have to be a caretaker manager of a team for a time.
In some cases, prospective team leads can be apprehensive about officially moving into management. If needed, I can offer a trial period of splitting the difference, asking them to take over as team lead while I retain direct reporting relationships, reevaluating in six months or so. I can always promote them to manager later, but if it turns out team leadership isn't for them, it's much easier to back out of the change if you don't have to go back and readjust titles and reporting relationships.
Default case: I'm stepping in to temporarily run the departing manager's team directly. Everyone on the team directly reports to me until I can hire a replacement. I hit up recruiting to open a req immediately. This may also be the only correct move if you have a team that needs an experienced manager, especially if there is specific technical expertise required to lead that team effectively.

Whew! Not only is that a lot to consider, but I may have still missed something. Even if not, it's rare that there's a contingency plan with no downsides. Often the "best" plan really is just the "least bad" solution to the problem.

Even once you determine your favored plan, executing it can be a trial of communication and compromise. There will probably be several key people who essentially have a "veto" over your preferred solution, whose blessing you must secure to move forward. If you run into insoluble resistance, you must go back, adjust your understanding of your organization's flexibility, and generate a new solution. I've had occasions where I had to run several iterations of solutions past different groups of stakeholders, and while working through layers of resistance can be frustrating, it's just so important to get this right. Given that we're dealing with people here, their careers and their work lives, getting to a solution where everyone is reasonably satisfied is critical -- unhappy employees tend to leave for other jobs, and when they do, you're right back to where you started, needing a contingency plan to handle an unexpected departure.