Latency

Latency is the wait time between a buyer typing a question and the assistant starting to answer.

It's measured in milliseconds.

The shorter it is, the snappier the chat feels.

Most of it comes from the model thinking, fetching data, or calling a tool before it can reply.

Why it matters for the ChatGPT funnel

Buyers in ChatGPT are impatient.

If the assistant stalls while reaching out to your app, the moment of intent cools.

A slow tool call is a dropped lead.

Speed is part of getting picked: assistants favor sources that respond fast and cleanly over ones that hang.

So latency isn't just an engineering metric here.

It's the difference between booking the meeting inside the chat and watching the buyer scroll past you.

How drio fits

You don't have to think about this.

drio is built to answer the assistant fast, so your app shows up the instant the buyer is ready.

The booking happens in-chat, while intent is still hot, not after a lag that loses the room.

Speed is handled for you.