To ensure we are impact-driven, we measure our success by how well we understand and address customer needs, rather than just the volume of features we deliver.

We measure our success not just by the features we ship, but by the real-life impact those features have on our customers' experiences and outcomes.

To determine the right metrics, I focus on the value we aim to deliver to customers and how we can measure that from their perspective.

To avoid perverse incentives from metrics, we ensure that our tactics align with our overarching goals. For example, we track the number of support tickets to gauge customer satisfaction and adjust our strategies accordingly.

We focus on a single overarching metric to guide our product development, which helps us prioritize our efforts and make informed decisions. This approach allows the entire team to rally around a common goal, enhancing our effectiveness.

Metrics are essential in product development as they provide a numerical representation of the value we deliver to customers. It's important to measure not just internal metrics but also how well customers achieve their goals using our product.

We must be cautious about how we define metrics for AI systems to avoid unintended consequences, such as optimizing for the wrong outcomes.

In biotech or hard tech businesses, your primary metric might not be revenue or active users, especially if you're still validating whether your technology works. Instead, focus on technical milestones that demonstrate progress.

Don't change the definition of your metrics just to make the numbers look better; consistency is key to understanding your progress over time.