Choosing a Primary Metric, Funnel Metrics and Guardrails to Judge a Product Launch

A product has just launched. Which metrics would you use to decide whether the launch was a success, and how would you choose and organize them? ```hint Start from what the launch was for The right metrics depend on what the launch was meant to change and over what window; a brand-new product and a feature added to an existing one call for different success signals. ``` ### Clarifying Questions - Is this a brand-new product, or a feature launched inside an existing product with an established user base? - What was the launch meant to achieve: new users, deeper engagement from existing users, revenue, or a strategic position? - Over what window should success be judged: launch week, the first quarter, or longer? - Were success targets agreed before launch, and was the rollout staged or released to everyone at once? ### What a Strong Answer Covers - Tying the metrics to the launch's goal and target users before naming any metric - One primary success metric plus a small set of funnel metrics from reach through retention and revenue - Guardrails for product health, cannibalization and harm to the rest of the product - Phasing the read-out over time against targets set in advance - Separating the launch's real effect from novelty, seasonality and launch marketing ### Follow-up Questions - A month after launch, adoption beat the target but retention among adopters is well below plan. Is the launch a success, and what do you do next? - The launch went to every user at once with no holdout. How would you estimate its effect? - How would your metrics change for a launch in an enterprise product sold to companies rather than to individuals?