# Time to first word

> Time to first word is the delay between a prospect clicking to start a demo and hearing the AI agent's first spoken word. It includes loading the call page, connecting audio, starting the agent and producing the first sentence of speech. A long time to first word makes people think the call is broken; a short one makes the demo feel instant.

Source: https://talkdemo.com/glossary/time-to-first-word · Updated 2026-11-19

## Key takeaways

- Measure from the click, not from when audio connects.
- Report the median and a high percentile, not just the best case.
- Warm pools of ready agents and browsers shorten it.

## What adds to time to first word

1. The call page loads and asks for the microphone.
2. The browser connects to the real-time audio room.
3. An agent process joins and, in live product mode, a cloud browser starts.
4. The agent's greeting is generated and the first audio plays.

## Why the first seconds matter

People decide quickly whether something works. Silence after a click reads as a failure, and some will close the tab. Starting the greeting before every system is fully ready, and filling short gaps naturally, keeps them on the call.

## TalkDemo's target

The PRD sets targets of a median time to first word under three seconds and under five seconds at the 95th percentile. These are design targets for a product in private pilots, not measured results. The agent and browser start in parallel to help meet them.

## Time to first word vs reply latency

- Time to first word happens once, at the start of a call, and includes setup.
- Reply latency happens after every question and covers listening, thinking and speaking.
- Both shape how human the call feels, and both are worth reporting as percentiles rather than averages.

## Frequently asked questions

### Is time to first word the same as latency?

It is one kind. Reply latency, the gap after each question, is measured separately during the call.

### Does the microphone prompt count?

Yes, if you measure from the click. That is why typing works as a fallback when someone declines the microphone.

### How is it measured in production?

By logging the click time and the first audio frame time for each session, then reporting percentiles.

## Related pages

- [Barge-in](https://talkdemo.com/glossary/barge-in)
- [Speed to lead](https://talkdemo.com/glossary/speed-to-lead)
- [Engaged demo](https://talkdemo.com/glossary/engaged-demo)
- [How it works](https://talkdemo.com/how-it-works)
