Skip to main content
Power Dialer gives each rep a fixed number of lines. Predictive pacing takes that decision away from the rep and gives it to the engine, which watches how often people are actually answering and adjusts. The goal is simple: a rep should never sit idle, and never be handed a call they cannot take.

​
⚖️ The two modes

Power Dialer

The rep picks their lines. Predictable, fully in their control, and the right choice for a small floor or a high-touch list.

Predictive Campaign

The engine owns the pace. Reps never choose lines. Best when several reps work one shared list together.
Which mode runs is a campaign setting chosen by an admin — never a rep’s browser setting. Pacing carries compliance consequences, so it is not a personal preference.

​
🔒 Turning it on takes two deliberate steps

1

Turn the engine on

It always lands in observing mode first. The engine records what it would have done against live traffic without moving a single call.
2

Read what it would have done

Check the pacing explanation on the campaign. Every decision states the constraint that bound it in plain language.
3

Leave observing mode

A second, separate action. Only now does the engine route real calls.
There is no single switch that goes straight to routing live. That is deliberate: nobody should be able to tick one box and start spraying calls.

​
🧱 The limits it will not cross

Predictive routing needs 2 reps minimum, and 3 before the fastest setting is allowed.With one rep and multiple lines, most simultaneous answers have nowhere to go. Below the floor the campaign quietly falls back to Power Dialer — reps keep working at full speed, they just stop being paced predictively.
Regulators cap abandoned calls at 3% of live human answers, measured per campaign over a rolling 30 days. The engine stops launching new calls well before that.No setting, no target and no “be more aggressive” instruction can push it past the ceiling. A human can make the dialer slower. Nothing can make it riskier.
The engine never dials more than 2 leads per free rep, whatever the maths suggests.Published guidance is consistent that true predictive pacing needs a much larger floor before the statistics behave. On a small team a higher ceiling buys nothing real and costs abandoned calls.
After any adjustment the engine waits before adjusting again.There is a delay of 20 to 30 seconds between launching a call and learning whether it connected. Without a pause, the engine reacts to the effect of its own last move and oscillates — a burst of calls, everyone buried, a stop, then everyone idle. Reps feel that as feast or famine.
Speeding up requires sustained evidence. Slowing down happens immediately on the first sign of risk.Over-dialing costs a regulated abandoned call. Under-dialing costs a few idle seconds. The penalties are wildly different, so the response is too.

​
🎯 What it actually steers

The engine targets the wait between conversations — the gap where a rep is free and nothing has landed.
It cannot target the time between connections. That includes talk time, which belongs to your rep and your prospect. If a rep talks for three minutes, no dialer on earth can hand them another conversation 20 seconds later.
A voicemail greeting is never treated as an answer. If machines counted, the engine would read them as people picking up and dial faster into a list that is not answering.

​
📞 When two people answer at once

A person who picked up the phone is never hung up on in silence. The worst outcome available is a short hold with a message, and every one of those is recorded as an abandoned call so the engine slows down rather than repeating it.

​
➡️ Where to go next

How calls actually work

What each outcome means before you tune anything.

Dialer Performance

Watch the effect of your pacing in the numbers.

Phone numbers

Why dialing harder needs more numbers.

Power Dialer

The manual mode and when to prefer it.