Choosing a North-Star Metric and Guardrails for Curated Account Lists on X

On X (formerly Twitter), Lists let a person put together a custom timeline that contains posts only from accounts they choose to add. Adding an account to a List does not require following it, so the person's main home timeline stays free of clutter. How would you measure whether Lists is a success? ```hint Follow a List through its life A person creates a List, adds accounts to it, and then reads it often, rarely or never. Ask which of these steps shows that the List is actually useful to them. ``` ### Clarifying Questions - What should Lists achieve for the company: deeper engagement from heavy users, a better experience for people who do not want to follow many accounts, or discovery of accounts? - Do Lists a person subscribes to, made by someone else, count the same as Lists they built themselves? - Are we assessing Lists as a mature feature, or measuring a recent change to it? - Should Lists be judged only on its own usage, or also on its effect on the home timeline and on following? ### What a Strong Answer Covers - The purpose of Lists and the user needs it serves, tied to company goals - A user journey from discovering Lists through creating and curating one to reading it regularly, with a metric at each step - One north-star metric that reflects the value people get, with the reasoning behind the choice - Guardrails for effects on the home timeline, on following, on the accounts that get listed, and on abuse - How to segment the data and separate the effect of Lists from the fact that heavy users are the ones who adopt it ### Follow-up Questions - After a redesign, List creation jumps but time spent reading Lists stays flat. How do you interpret that? - Heavy List users follow fewer accounts than similar users who do not use Lists. Is that a problem for the platform? - How would you show that Lists makes people more likely to stay, rather than simply being used by people who were going to stay anyway?