In 1975, Frederick Brooks, who managed the development of the IBM OS/360 operating system, formulated an observation in his book "The Mythical Man-Month" that has since become one of the most quoted and simultaneously most ignored rules of management: adding manpower to a late software project makes it later.
Intuitively, the opposite seems obvious—if a project is falling behind, you just need to add people, and it will speed up. The manager's logic here is completely natural: more hands mean more work done per unit of time. The problem is that this logic only holds true for tasks that divide perfectly into independent parts. Real, complex work doesn't divide like that: it has internal dependencies that require coordination, and coordination does not grow linearly with the number of people, but roughly quadratically—because the number of possible connections between participants increases.
Brooks adds another factor that makes the effect worse rather than merely neutral: new people need time to understand the context of the project, and the ones who have to train them are the very people who are already under deadline pressure. That is, in the short term, adding people doesn't just fail to speed up the work—it temporarily slows it down even more, taking away resources from those who are already bottlenecks.
The law was formulated for software development, but the mechanics apply to any teamwork with a large number of internal dependencies—not just to code. A committee falling behind schedule; a product launch on the verge of failing; any complex cross-functional project—everywhere, the intuitive reaction of "let's add more people" risks having the opposite effect.
The practical takeaway is not that you should never strengthen teams—sometimes it is justified, especially in the early stages of a project, while dependencies and context are still minimal. The takeaway is that the decision to add people should be made deliberately, rather than as a knee-jerk reaction to falling behind schedule—and one must honestly calculate not only the gain in hands, but also the cost of coordination and onboarding that will have to be paid right now, when time is shortest.