When a product improvement is meant to benefit another department (e.g., reduce support tickets), don't just ship it and hope for the best. Create a joint, aligned goal with that department's leader. This ensures they are accountable for accruing the benefit (e.g., reallocating saved capacity) and solidifies your impact.

Before committing to an outcome, teams should ask: "If we achieved this number via methods I'd be embarrassed to see in a news headline, is it a worthy goal?" This simple thought experiment acts as a powerful, practical guardrail against unethical tactics.

Instead of cascading goals directly from a vision, use "Strategic Themes." These are broad, directional choices (e.g., "Leverage critical partnerships") that act as guardrails, or "lanes on the interstate," guiding how teams set their specific, measurable objectives.

Cascading OKRs through multiple layers (company to department to team to individual) often results in "OKR theater" where the connection to business impact is lost. Instead, an individual product manager's goals should be no more than one link away from a core business objective that leadership cares about.

For goal-setting to be effective, limit company-wide goals to three. Designate one goal as the ultimate tie-breaker in resource conflicts. Ensure goals are simple enough for an intern to understand. Crucially, your strategy must involve painful trade-offs ('strategy should hurt'), otherwise you haven't truly prioritized.

Simply stating a goal, like "increase sales by 15%," is insufficient for autonomous teams. Leaders must also articulate the "anti-vision"—the negative outcomes to avoid, such as eroding customer experience. This rich context provides clearer guardrails and a more nuanced understanding of the mission.

A strong AI goal is a structured directive, not a vague wish. It must include six components: a desired outcome, a verification method, constraints, boundaries (tools/files), an iteration policy (how to decide next steps), and a stop condition. This mirrors the rigor of setting measurable business objectives.

Bottom-up goal setting often leads to conservative, achievable targets. Instead, leaders should set an ambitious top-down goal with a resource constraint ('achieve X with Y people'). This forces teams to rethink their approach, not just incrementally improve.

Don't build a feature roadmap and then write OKRs to justify it. Instead, start with the outcome you want to achieve (e.g., "move metric X to Y"). This frames all features as experiments designed to hit that goal, empowering teams to kill features that don't deliver value.