Quora Voice Commands: Hands-Free Asking, Listening and Acting on Answers

You are a product manager at Quora, the question-and-answer platform where people ask questions, write answers and read what others have written. Design a feature that lets people use Quora through voice commands. Explain which users and situations you would build it for, what people should be able to do by voice, what the first release would include and how you would tell whether it is working. ```hint Find the moments voice wins Reading and writing on Quora assume eyes on a screen and hands on a keyboard. Think about when those are unavailable or tiring for a Quora user, and which of asking, answering, reading and reacting is worth doing by voice in those moments. ``` ### Clarifying Questions - Is this for Quora's mobile apps only, or also the website, smart speakers and cars? - Does "voice command" mean controlling the app (search, navigation, upvoting, following), dictating questions and answers, hearing content read aloud, or some combination? - What should the feature achieve for Quora: more reading time, more questions and answers written, or better accessibility? - Which languages and markets must the first version support? ### What a Strong Answer Covers - A goal tied to Quora's loop of asking, answering and reading, and a reasoned choice of users and usage moments for voice - The pain points of the chosen users when they try to use Quora without a free screen or keyboard - A prioritized first release with a concrete command set, and the trade-offs behind it, such as push-to-talk versus a wake word and which content works when spoken - Success metrics specific to voice use, with guardrails for content quality, ad revenue and privacy - Risks such as misheard niche terms, answers that depend on images, code or formulas, and garbled or duplicate questions asked by voice ### Follow-up Questions - Quora relies heavily on advertising. How would ads work in a listening mode, and what would you do if voice sessions earn less per session? - Questions asked by voice turn out to be duplicates or poorly phrased more often than typed ones. What would you change? - Would you bring the feature to smart speakers, and how would someone there ask a question and learn when it has been answered?