Skip to content
Campaign Audience BuilderCampaign Audience BuilderCAB Documentation

Let Cabby help define and explain outcomes

Cabby can help define success before an activation and explain measured results afterwards. These are separate workflows: drafting a rule needs the configured Agent, while explaining measured outcomes also needs the administrator’s outcome-data opt-in.

Ask Cabby to draft a conversion rule

Section titled “Ask Cabby to draft a conversion rule”
  1. In CAB Outcomes, create or edit a conversion-event outcome and open its conversion-rule step.
  2. Select Cabby. In Describe the outcome you want…, describe the success record, qualifying conditions, relevant date, and conversion window.
  3. Review the draft in the editor. Use Undo or Redo to review changes, and inspect the relationship and filters before accepting them.
  4. Complete the name, tracked object, listening assignments, and other required fields, then select Save outcome.

For example, to measure new recurring donations, ask: “Track accounts that gain an active recurring Gift Commitment within 90 days after activation. Use its effective start date as the conversion date.” Review whether the generated rule matches your org’s actual objects and fields; this example is not a schema-independent preset.

Cabby can work on the conversion object, relationship, date, window, and supported filters. Complete outcome metadata and assignments in the editor yourself. The reading schedule follows the outcome’s timing; asking for more checkpoints does not make additional conversions eligible.

Ask why performance looks the way it does

Section titled “Ask why performance looks the way it does”

Open an outcome’s record and its Measurements tab. When the feature is enabled, the Cabby card sits beside the metrics. Select Ask why to request an explanation. A current cached explanation can appear on opening the record; opening the page does not always generate a new response.

Read the explanation’s timestamp and compare it with the displayed checkpoint and population. Choose Regenerate with the current measured data after new readings arrive. CAB explains when there is not enough measured data instead of supplying a result from an empty history.

Other supported entry points include:

  • An activation in Run History: Let Cabby explain this send’s performance.
  • A control-group comparison: Let Cabby explain the lift.
  • An audience’s Outcomes view: its Cabby card explains the result being shown.

For Monthly giving started, the day-7 comparison is 18.1% activated versus 8.3% held back. For Lapsed donor reactivated, it is 9.9% versus 15.5%. An explanation must retain that direction, the checkpoint, and any limits on the evidence. It should not describe the waiting volunteer outcome as a failed campaign or turn a positive difference into a significance claim.

Keep interpretation tied to the evidence

Section titled “Keep interpretation tied to the evidence”

Ordinary rates and audience rankings describe observed results. A held-back comparison supports a different question about campaign impact; an organization average cannot substitute for it.

The optional outcome-data feed shares aggregate outcome statistics rather than member or converter lists. The visible numbers and Salesforce evidence remain the source for checking an explanation. Information that a user types into a prompt or filter is a separate input: do not assume the aggregate-feed restriction removes it from the conversation.

Use Act on it for the explicit workflow that turns a result into suppression, follow-up, or a similar-audience starting point. See Use the result in your next audience.