How a work area balances speed and cost.

Each work area shows the balance it strikes between speed, depth and cost.

Status and explanations. It runs Read-only.

See why the hotline answers at once and acquisitions take their time.

Open the hotline’s work area, and it shows its balance: answers come fast and stay to the point.

A quick answer on the hotline, a careful one for a purchase.

Each request follows the balance of the work area it comes from, so a caller on the hotline and a team weighing up a purchase each get what their work needs.

A caller asks whether roses can arrive by Friday, and gets a fast, short answer.

The same balance costs less as Duplex learns.

An area’s balance stays where your team set it. As Duplex learns the area’s work, more of it is answered without a model call, so the same balance costs less.

In its first month, the hotline runs fast and light, and much of its work still goes to your AI.

What the area shows is what it runs.

The balance an area shows is the one in force. People and agents see the same balance, looking it up changes nothing, and only your administrators can set it.

What it shows is the balance the hotline runs right now.

Follows your settings.

Each work area has its own settings for speed, depth, cost and confidence, and a model’s output can never change them.

Next: Your team tunes speed, depth and costSee it →

See it on your own work.

From your own assistant to every agent your company runs, a pilot puts Duplex on your own work, with the measures agreed before it begins.

Start a pilot →All 94 capabilities