Organizational Dynamics

Everything in the previous chapter was true and also incomplete. Meridian’s redesign was defensible, its reassignments were real, and thirty people had somewhere genuine to land. None of that changes what the other seventy-five experienced: months of not knowing whether they’d be one of them.

That uncertainty is the actual subject of this chapter, and it deserves to be treated as a business risk, not just a management-sensitivity footnote. A reengineering effort that mishandles it doesn’t just make people unhappy. It can fail outright, or succeed at a cost far higher than the plan assumed.

Who Leaves First

The people who leave first are rarely the people you’d choose to lose. Once word gets out that a process, and the roles inside it, are under review, the employees with the most options start testing the market, whether or not their own job was ever actually at risk. They can afford to leave; the people who can’t find another job easily are the ones who stay and wait to find out their fate. A company can design the fairest, most defensible redesign in the world and still watch its best people walk out the door during the many months it takes to execute it, simply because uncertainty is itself a cost, and talented people are the ones best equipped to avoid paying it.

The Self-Preservation Problem

This is where the self-preservation problem from earlier in this book gets sharp. I said, back when this book was still asking what reengineering even is, that strategy conversations are full of people who are certain their own function is indispensable, for reasons that are sometimes true and sometimes simply self-preservation talking. Once the mapping work starts, that fear doesn’t sit quietly. People withhold detail, round their workload up, quietly make their own process look more essential and more complicated than it is, not out of dishonesty exactly, but because nobody volunteers evidence for their own elimination. It’s precisely why the mapping should be done without names attached in the first place; it’s also why, even de-personalized, the underlying data can’t be fully trusted, and a leader who forgets that will build a redesign on top of a flattering fiction.

Protecting Friends, Not Just Yourself

It isn’t only the employee whose job is on the line who has an incentive to look away. The manager standing over that employee often has the same problem, in a different shape. It happens in company after company: work that stopped being necessary quietly continues, not because anyone is fooled about it, but because the manager who could eliminate it is too uncomfortable calling it out. The person doing the work is a friend, a long-tenured colleague, someone they hired themselves. Organizations don’t just tolerate that instinct; they tend to reward it, because “protects their people” reads as loyalty, right up until it’s the reason a redesign quietly fails to touch the very work the diagnostic already identified as unnecessary.

This is a harder problem than the employee’s own self-preservation, because there’s no dishonesty to correct. The manager usually knows exactly what’s true. The failure is in the decision, not the information — which is why eliminating a specific role can’t be left entirely to the manager closest to the person in it. Someone needs the authority to make that call who isn’t also that person’s friend: a reengineering team, a process owner one level removed — the same separation of authority this book argued for earlier, before knowing exactly why it would matter this much. A manager’s discomfort with cutting a specific person is real information, worth listening to. It just can’t be allowed to be the final word.

The Operational Risk

This is the one boards underestimate most. The old process has to keep running until the new one is actually live, which can be a year or more. If the people who know how to run the old process, including its undocumented workarounds and quiet exceptions, leave early out of anxiety, the company can find itself failing customers on the process it hasn’t reengineered yet, for want of the very people it was planning to reassign or let go on its own schedule. A redesign that’s financially sound on paper can still come apart operationally, not because the ideal state was wrong, but because nobody thought to keep the old state adequately staffed until the switch actually happened.

None of this is a reason to avoid reengineering, or to pretend the human cost isn’t real, both of which are their own kind of failure. It’s a reason to treat the transition itself as something that needs deliberate protection, not just good intentions.

What Actually Helps

Specificity beats reassurance. An employee who knows precisely what happens to them, and when, tolerates bad news better than one left to imagine it. Retention agreements, real money, for the specific people whose knowledge the old process needs through cutover, are cheaper than the operational failure of losing them early. And when reassignment is real, as it was for the thirty people who moved into Meridian’s new Digital Operations team, say so plainly and early rather than treating it as a footnote to the headcount number. People can carry hard news. What erodes them is not knowing which kind of news is coming.