The binder of procedures tells every engineer exactly what management thinks of them.
A software team is falling behind. The manager introduces a standard methodology, adds a weekly status review, tightens the quality gates, and ranks the engineers against each other at review time. Six months later the team is slower, two of the best people have left, and the postmortem blames the technology.
Tom DeMarco and Timothy Lister’s Peopleware argues that the postmortem has it backwards. Most problems of work are sociological, not technological, but most managers spend their time on convoluted technical puzzles as if they were going to do the work rather than manage it, because the human problems are harder and offer no clean solutions. The manager’s function is not to make people work but to make it possible for them to work. Each of the manager’s moves made the work less possible.
Threatened self-esteem
A novice manager believes work can be done without people’s emotions getting involved. Anyone with experience has learned the opposite. The major arouser of emotions at work is threatened self-esteem, and when a person’s sense of competence is at risk, the emotional response overrides everything else. Unreachable deadlines, merit rankings and enforced uniformity are precisely the threats most managers create.
Uniformity is the subtle one. The need for it is a sign of insecurity on the part of management: “professional” comes to mean looking, acting and thinking like everyone else. But people need a sense of uniqueness to be at peace with themselves, and they need that peace before a team can form.
Methodology as anti-thinking
The standard methodology was meant to make the team reliable. DeMarco and Lister see it differently:
A Methodology is an attempt to centralize thinking, destroying self-healing capacity. Implicitly it says that people aren’t smart enough to do the thinking.
Nothing is more demotivating than the knowledge that management thinks you incompetent. The predictable result is malicious compliance: if people did exactly what the Methodology said, the process would grind to a halt. The OSS sabotage manual tells saboteurs to insist on following every procedure to the letter. Peopleware shows that well-meaning managers produce the same behavior with no saboteur required.
The jelled team
What the manager actually needed was a jelled team, one where the whole becomes greater than the sum of its parts. A team’s purpose is goal alignment, and alignment cannot be commanded. It grows out of shared small successes, mutual trust and the safety to fail.
Read that way, the manager’s four moves were a list of what kills jelling. Merit reviews and rankings, the things managers spend the most time on, foster competition inside the team and are among the most teamicidal. Quality pressure, which forces a product below the builders’ own standards, leaves the team unable to look each other in the eye. And the weekly coordination backfires, because when you over-coordinate for your people, they under-coordinate their own efforts. A good coach understands that their job is not to coordinate but to help people self-coordinate.
What builds a team is frequent closure: a good manager divides a project into twenty closed iterations when only two are necessary for upper management, because each small, demonstrable win builds trust, and trust lets the team coordinate itself.
Protecting people from the organization
In the book’s view, management is not setting goals and holding people accountable, or even removing blockers: management is about creating the conditions under which people can do their best work — which means protecting them from the organization itself. The marketplace, clients and upper management will never make the case for quality that a team needs in order to jell. So the manager has to fight for quality against the institution they represent, tolerate uniqueness in a system that rewards uniformity, and create safety in an environment that generates threat.
Protection is not the same as leaving everyone alone. The book’s advice is to single out the few key spirits with the proper mix of perspective and maturity and turn them loose, a judgment no methodology can make. The reason to turn them loose is the book’s shortest insight: people perform better when they are trying something new. The standard methodology tried to make the work familiar and repeatable, and an organization that cannot tolerate novelty will eventually keep only the people who do not need it. Which of the manager’s engineers will those be?
References:
- Tom DeMarco and Timothy Lister, Peopleware: Productive Projects and Teams