Campaigns

A campaign has an outbound voice agent call a list of contacts automatically — following up leads, confirming appointments, or running a reminder round — instead of only answering calls that come in.

Campaign vs. a single test call Placing a manual test call (see Testing & publishing) dials one number so you can hear how the agent sounds. A campaign dials a whole list, with pacing, retries, and compliance controls built in — use a test call to check the agent, and a campaign to actually run it against real contacts.

Setting up a campaign

A campaign needs a published, outbound-mode agent and an owned caller-ID number to dial from (carriers reject calls with no caller ID). Upload a contact list — up to 1,000 rows, phone numbers only need to be recognisable, they're normalised automatically. Duplicate numbers, and numbers already in the campaign, are rejected before the campaign runs, so you'll see exactly what will actually be dialled before you commit.

Before launching for real, use dry run to validate the whole list — bad numbers, duplicates, and configuration problems are reported without placing a single call.

Pacing

• Concurrency — how many calls run at once, from 1 up to 50 (the tenant-wide ceiling). Higher concurrency finishes a list faster but needs enough available agents and lines to handle it.
• Calling hours — restrict the campaign to a daily time window, in your chosen timezone, so it never dials outside acceptable hours.
• Retries — 1 to 5 attempts per contact, with a configurable delay between attempts. Which outcomes count as worth retrying (no answer, busy, and so on) is set on the agent itself.

Compliance

Launching a real campaign (not a dry run) requires confirming two things:

• Consent — that you have a lawful basis to contact everyone on the list.
• Recording disclosure — if the agent records full calls, that your disclosure practice is in place.

Anyone who's opted out is automatically skipped — both when the campaign is created and again right before each call is placed, so an opt-out recorded mid-campaign still takes effect.

Budgets

Set optional daily and monthly spend caps on the agent. Once a hard cap is hit, the campaign stops dispatching new calls until the next period rather than continuing to spend.

Watching a campaign run

A running campaign reports live counts — queued, dialling, completed, failed, and blocked (skipped for opt-out or budget reasons) — plus a per-contact timeline showing the outcome, call summary, cost, and duration of each attempt.

Everyday tasks

• Follow up a list of leads automatically → publish an outbound agent, upload the list, and launch a campaign.
• Make sure you're not calling too early or late → set Calling hours in the right timezone.
• Avoid overwhelming your team or lines → keep Concurrency conservative and raise it once you've seen a batch run smoothly.
• Stop a campaign from overspending → set a daily or monthly budget on the agent before launching.
• Check the list is clean before committing → run a dry run first.