There is no separate product management goal and no separate engineering goal — only a single product team goal that both sides plan, present, measure, and are held accountable for together: "there's nothing like a product management goal. And there's nothing like an engineering goal. There's only a product team's goal."
This reframes the PM/engineering relationship: "you're moving from a traditional subservient relationship to a collaborative one. And that's a big change." It's a sharper, goal-level version of the same misalignment problem named in Misconception: PM Owns the 'What', Engineering Just Executes the 'How', and it depends on the shared context described in High-Alignment/High-Autonomy Model and Three-Branch Responsibility Logic.
Require product managers and engineers to co-pitch initiatives — including tech debt — as one shared business case with one shared goal, rather than each side bringing its own agenda into the room.
A company example used to illustrate the principle in the negative: having 17 declared 'north stars' is functionally identical to having none, because a north star that doesn't differentiate priority isn't guiding anything — teams can't sail toward 17 different stars at once.
Apply: when auditing a transformation's stated goals, treat a long list of undifferentiated priorities as equivalent to the Vision Vacuum: No Vision, Not Bad Vision, Is the Real Failure Mode, not as more direction than a single clear goal would give.
Cagan extends the one-goal principle to compensation design: incentives should be structured at the team level — one or two shared OKRs plus equity or profit-sharing — so no team can "win" at another team's expense. He's explicit that product incentive design should never mimic sales-style individual competition, which pits people against each other instead of against the problem. See OKRs/KPIs as Business-Strategy Mechanism.
Cagan is explicit that OKRs under the product model belong to the team, not the individual — one or two shared objectives per team. He calls individual OKRs a "distraction," and cites Google itself as a company that fell into using them despite the model's own logic arguing against it: individual OKRs push a team back toward isolated output ownership instead of a single shared team-level outcome.
Из тем: The Product Operating Model