All terms

Consumer Behaviour

Voice search

Also known as: spoken search

Voice search is the act of speaking a query to a device or assistant and receiving a spoken or on screen answer. It covers smart speakers, phone assistants, in car systems, TV remotes and voice input inside search and shopping apps. Spoken queries tend to be longer and more conversational than typed ones, and often return a single answer rather than a list.

What it is

Voice search combines speech recognition, query understanding and answer selection. The system transcribes what was said, interprets intent, then picks a response, which may be read aloud from a single source, shown as a results page, or handled by a device action such as calling a shop. Assistants built on large language models also handle follow up questions in the same conversation.

Why it matters

When the answer is spoken there is usually one result, so ranking second often means no visibility at all. Voice queries skew towards local, factual, transactional and hands free tasks, which makes them commercially useful for retail, hospitality, services and support. The conversational phrasing also shows how people actually ask questions, which is useful for content and AI answer optimisation more broadly.

How it works

Practitioners write clear question and answer sections that can be lifted as a spoken response, keep business listings and opening hours accurate, use structured data for products, recipes, FAQs and local details, and keep pages fast and accessible. They test real spoken phrasing on the devices their audience uses and check which source the assistant credits. Content is written to answer the question in the first sentence, then support it.

When it applies

It applies to local, on the move, in kitchen, in car and accessibility contexts, and to any category where people ask short factual or task based questions. It matters most where a single spoken answer can win or lose the whole query.

Examples

  • A driver asks a phone assistant for the nearest garage open now and is read one business name, address and phone number.
  • A cook asks a smart speaker how long to boil new potatoes and hears a step read from a single recipe page.
  • A shopper says 'reorder my usual dog food' into a retailer app and the basket is filled from purchase history.

How it is measured

  • Impressions and clicks for long, question shaped queries in search console data
  • Calls, direction requests and clicks to website from local business listings
  • Share of tested spoken prompts where the brand is the cited source
  • Position and presence in featured snippets or answer boxes for target questions

Related terms in Consumer Behaviour

Primary research · August 2026

How ChatGPT Shortlists Software Brands

An audit across 10 categories and 60 buying questions. I recorded what ChatGPT reads, throws away and links to when a buyer asks it which software to buy, and what that decides.

60
Questions asked
10
Software markets
2,680
Results read
367
Links shown
Free35 pages · PDF · 536 KBDiscovery Digest every Friday

Free download

Get the full report

35 pages · PDF · 536 KB. Enter your details and it downloads straight away.

How ChatGPT Shortlists Software Brands downloads straight away. No spam, unsubscribe anytime.