Which action is most appropriate to investigate unexpected low usage after a release?

• A. Turn off the features that are never used without user input
• B. Increase the price to reduce usage
• C. Wait for more data from analytics before taking action
• D. Run experiments to understand customer needs

Answer: (D)

When usage drops after a release, the best approach is to learn why by testing hypotheses about how customers use the product. Running experiments lets you validate assumptions about customer needs, behavior, and the product’s value proposition with real data, quickly revealing which changes will actually boost usage.

Experiments provide fast, targeted feedback. You might test different onboarding steps, how features are discovered in the UI, or messaging that explains value. By comparing metrics such as active usage, feature adoption, and activation rate across controlled variations, you can pinpoint what improvements move usage in the right direction without guessing.

Other options rely on assumptions or delaying action. Turning off features without understanding why they’re underused can reduce value for some users and misses the chance to learn from the data. Waiting for more analytics slows down learning and keeps you in a state of inertia. Raising the price to reduce usage addresses a symptom, not the underlying needs or barriers, and can alienate users.

So, the action that best fits the goal of understanding customer needs and driving informed product decisions is to run experiments to uncover why usage is unexpectedly low.