Past three priorities, the word stops meaning anything
A priority is a decision about what loses. A list of ten top goals is a list with the deciding left out.
If you have more than three priorities, then you don’t have any.
A small software company sets its goals for the quarter. Sales wants a new pricing page, the largest customer wants single sign-on, the founders want a mobile app, the lead engineer wants to pay down technical debt, and someone has promised a partner an integration. All five go on the list as top priorities. Three months later each has moved a little and none is finished, because every Monday the team split its attention five ways.
Jim Collins, the author of Good to Great, has made this point in talks and interviews for years, and the version above is the one most often repeated. It works less as arithmetic than as a definition. A priority tells people what to do when two good things compete for the same afternoon. If everything is a priority, that question is left to whoever happens to be in the room, and the answer changes every day.
Collins pairs the idea with another habit he has recommended: keeping a “stop doing” list next to the to-do list. Adding a goal without naming what gets dropped is how lists reach five and then ten. The discipline is in the subtraction, not the ranking.
The number itself deserves some skepticism. Three is a rule of thumb, not a research result. For a two-person startup one priority is often right, and three would already be too many. For a company of several thousand people, each division can hold its own three, and the company as a whole will have more than that without losing focus, as long as each team knows which of them is its own.
There is also a category the rule tends to hide. Payroll has to run, taxes have to be filed, servers have to stay up, customer data has to stay safe. These are not priorities in the sense Collins means. They are constraints: things that must not break while the priorities get done. Teams that list them alongside real goals inflate the count, and teams that leave them off entirely get surprised when one fails. Keeping them on a separate line makes the real list shorter and more honest.
So the useful test is simple. Ask anyone on the team what matters most this quarter. If the answers differ, or if every answer is a list, the company has goals but no priorities, and the next hard trade-off will be settled by accident.