September 2, 2026

Voice-first

The phrase mobile-first was coined during a time when most everything was first and foremost a web app — mobile apps were an afterthought, bolted on if built at all. Everything was done in a browser: books were purchased through a browser, videos were watched through a browser. Later, as smartphones became ubiquitous, there was a realization that crafting the experience specifically for a phone opened up possibilities the web never had. Touch was a better experience than a mouse. Phone cameras could take photos or record video. GPS enabled location-based services. Payment apps could charge you directly. None of that was “mobile web” — it was only possible once you designed for the phone itself.

Voice-first, similarly, should be about making voice a first-class citizen, not a bolt-on. The technology has evolved to the point where this is now possible: speech recognition, language understanding, and speech synthesis are all mature. Many frontier models now ship voice interfaces alongside the traditional text/display one, so it’s no longer just Siri and Alexa in this space. And the same thing happens here that happened with mobile: new capabilities show up once voice is the design target, not an add-on. You can act on something while your hands and eyes are busy with something else — driving, cooking, walking. There’s no app to unlock and find. Presence doesn’t require a screen at all.