Summer ’26 release
Summer ’26 turns audiences into living, measurable assets: reuse filters under governance, see who joined or left, automate safely, and measure what activation actually achieved.
Reading outcome results
Section titled “Reading outcome results”Changed. Outcome results read the same way wherever they appear. A rate always carries the population it was measured over, a checkpoint rail shows how far measurement has travelled, and one split bar shows who was contacted and who was held out. In CAB Outcomes, a three-column band separates what happened from who it happened to from what it is measured against, and checkpoint charts draw a bar per checkpoint with its own sample size and run count. Inside a run, measurement moves into a Measurement column of the table that already lists the run’s additions, alongside a Held out column that appears only where a control group exists. See Measure what happens after activation and Measure campaign impact with a control group.
Filter Library
Section titled “Filter Library”New. Save a proven filter once and reuse it across audiences as a fixed copy, a configured copy, a synced reference, or a locked policy administrators enforce. Entries can cover one object, a connected-object chain, or an aggregate qualification, and Cabby can find or author them. See Reuse and govern filters.
Audience canvas layout
Section titled “Audience canvas layout”Changed. The canvas lays out complex audiences in a predictable left-to-right reading order, keeps connection logic aligned, and adds a compact plain-language summary. Switching is a per-user choice and does not change any audience definition. See Work on the audience canvas.
Run history and membership changes
Section titled “Run history and membership changes”Changed. Run history separates “how large was this audience?” from “who actually changed?”, showing size trends at a glance while eligible runs record joiners and leavers against the previous baseline. See See audience size and membership change over time.
Delta processing
Section titled “Delta processing”New. An eligible scheduled action can process only the membership changes since its last successful activation instead of the whole audience every time. See Process only what changed between scheduled runs.
Audience Composition
Section titled “Audience Composition”New. Beyond a result count, see what the group looks like: prominent values, recency bands, missing data, differences from the organization baseline, and a short Cabby narrative. See Understand an audience with composition profiles.
Scheduled sequences
Section titled “Scheduled sequences”Changed. Combination and waterfall sequences can run on a recurring schedule with the same automation controls as a standalone audience, including delta processing, tracked outcomes, and change approval. See Schedule combination and waterfall sequences.
Sequence ledger
Section titled “Sequence ledger”Changed. The sequence record page presents a combination or waterfall as one table carrying each step’s setup and its result together, replacing the separate setup and results panes. It also tracks which results an edit invalidates and offers a partial or full re-run. See Build audience sequences.
Change approval
Section titled “Change approval”New. A scheduled run can alert an approver — or pause all of its scheduled actions — when total churn crosses the schedule’s thresholds, with the decision made in the app and org defaults an administrator can lock. See Review large membership changes before actions run.
Outcome measurement
Section titled “Outcome measurement”New. Define success once and CAB measures each listening activation at your chosen checkpoints, with results in the CAB Outcomes tab and alongside the run that produced them. See Measure what happened after activation.
Outcome storage
Section titled “Outcome storage”New. A settings page reports what outcome measurement is holding org-wide, draws the limit inside your org’s own data allocation, and switches measuring on or off. Manage lists every campaign or list by weight so you can stop tracking one or delete what it holds. See Control what outcome measurement stores.
Control groups and lift
Section titled “Control groups and lift”New. Withhold a stable share of genuinely new members and compare their conversion with the contacted group to estimate what the campaign actually caused. See Reserve a control group and measure campaign lift.
Outcome-aware Cabby
Section titled “Outcome-aware Cabby”Changed. Cabby can draft a conversion definition before measurement starts and — when an administrator opts in — use aggregate outcome history to rank ideas and explain measured performance. See Let Cabby help define and explain outcomes.
Export options
Section titled “Export options”Changed. The results-table download opens a configurable export rather than one fixed CSV: choose file type, encoding, and columns, and for Excel include the audience definition and run metadata. See Export results in the format your downstream work needs.
Notification cards
Section titled “Notification cards”Changed. Standing, actionable notifications appear as a consistent stack of cards whose severity, message, and call to action follow the current user, so administrators see setup controls where others see next steps. See Act on important setup and runtime conditions.
In-segmentation deduplication
Section titled “In-segmentation deduplication”New. Collapse duplicate records inside the run by match field, and exclude the audiences that must keep every copy. See Deduplicate audience results.
Custom Filters
Section titled “Custom Filters”New. Filter an audience with your own Apex when fields and related objects cannot express the rule. See Custom Filters.