Adobe Audience Manager Business Practitioner Expert cheat sheet

Audience Manager (AAM) core objects: signals (key-value data) → traits (rules on signals) → segments (AND/OR trait logic) → destinations (activation targets).

Core ID types: Adobe Experience Cloud ID (ECID/MID), AAM UUID (AAM device ID), and declared IDs (CRM/customer IDs) for cross-device and offline matching.

DIL (Data Integration Library) on-page code fires the /event call; the AAM UUID is stored in the demdex cookie and synced with the ECID via the ID service.

Server-side forwarding lets Analytics send data to AAM through the Analytics destination, avoiding a separate DIL call on the page.

Profile merge rules define how identities are stitched: device-only, current authenticated + device, authenticated + device, and authenticated + device + profile link (legacy).

AAM region shards: data can be stored in multiple regional data centers; the demdex cookie and ID syncs are region-scoped.

Implementation order: deploy ID service → deploy DIL/event calls → validate signals in the Signal dashboard → build traits → build segments → configure destinations.

The /event call carries signals as c_ parameters and d_ parameters; signals are the raw data, traits are the rules, segments are the audiences.

Signal key-value pairs use c_ (customer-defined) or d_ (Adobe-defined) prefixes; the trait rule references the key without the prefix in the rule editor.

Trait rule operators: equals, not equals, contains, starts with, ends with, greater than, less than, exists, and regex match.

Trait types: rule-based, onboarded (file/CRM), algorithmic (look-alike), and folder traits (auto-generated from folder structure).

A trait qualifies a user when the incoming signal matches the rule; trait population counts update after data is processed, typically within hours.

Regex traits can match patterns in a single signal value; use them sparingly because regex evaluation has performance costs at scale.

Trait TTL (time-to-live) controls how long a trait stays on a profile; default is often 30 days, configurable per trait.

Signal-based traits fire from real-time /event calls; onboarded traits come from inbound files and are matched via declared IDs.

Trait folder organization affects UI navigation but not the rule logic; folder traits inherit rules from their parent folder path.

Segments combine traits with AND (all traits), OR (any trait), and NOT (exclude trait) logic; nesting is supported via segment rules.

Segment qualification is evaluated against the merged profile, so profile merge rule choice directly changes segment membership.

Real-time segments qualify users immediately on signal arrival; standard segments require batch processing and refresh on a schedule.

Segment population is the count of unique profiles that currently meet the trait logic, not the number of signal events.

Recency and frequency (R/F) rules let a segment require a trait to have fired N times within a time window.

Segment overlap reports show how much two segments' memberships intersect; use to avoid over-targeting the same users.

Segment TTL and trait TTL interact: a segment can only qualify as long as its constituent traits remain on the profile.

Excluding traits with NOT reduces reach; validate exclusion logic against the trait population before activating.

Inbound data sources: real-time (DIL/event calls), onboarded (files via FTP/S3), and server-to-server (S2S) API integrations.

Onboarded traits require declared IDs in the file (e.g., CRM ID, hashed email) that match identities already known to AAM via ID syncs.

Data source types: first-party (owned data), second-party (partner data), third-party (data provider); each has distinct governance implications.

Inbound file formats: CSV, TSV, and JSON; each row must contain a declared ID and the trait signal values to onboard.

The AAM S2S API allows real-time trait and segment updates from server-side systems without a browser.

ID syncs (pixel-based) map AAM UUIDs to partner IDs; without syncs, onboarded data cannot be matched to online profiles.

Adobe Experience Platform and AAM integration: AEP audiences can be shared to AAM and vice versa via the Experience Cloud audience sharing.

Data source governance: set data source purpose and data use labels so downstream destinations honor usage restrictions.

Destination types: cookie-based, URL-based, server-to-server (S2S), and Adobe-native (Analytics, Target, Campaign, Audience Lab).

Cookie destinations fire a pixel to the partner with the AAM UUID; the partner must ID-sync to receive the segment membership.

S2S destinations push segment membership via API batches on a schedule; latency depends on the destination configuration.

URL destinations redirect users to a partner URL with segment parameters; used for on-site personalization handoffs.

Destination mapping: each segment must be mapped to a destination before membership is sent; unmapped segments are not activated.

Audience Lab creates test segments and splits them across destinations to measure incremental lift.

Destination rate limits and batch schedules affect how quickly segment changes propagate; real-time destinations update per event.

Destination eligibility is constrained by data governance labels; a segment with restricted data uses may be blocked from a destination.

General reports: trait and segment population trends, signal reports, and overlap reports; all reflect processed data, not raw event counts.

The Signal dashboard shows incoming signal keys, values, and counts; use it to validate implementation before building traits.

Trait and segment population counts lag real-time events by the processing window (often hours); do not expect instant counts.

Overlap reports compare two or more segments/traits to find shared membership for deduplication and frequency management.

Audience Optimization reports (Audience Optimization for Advertisers) measure audience performance across campaigns and destinations.

Algorithmic traits and look-alike models use historical conversion data to score profiles; scores update on model refresh cycles.

Segment trend reports show population over time; sudden drops often indicate trait TTL expiry or a broken signal.

Data export from reports is limited by AAM's reporting retention window; long-term analysis requires exporting to an external system.

Data governance labels (data use labels) restrict how data can flow to destinations; destinations must be authorized for the label.

Data source purpose (e.g., advertising, personalization) must align with the destination's allowed uses; mismatches block activation.

Opt-out: users who opt out of AAM are flagged and excluded from trait/segment qualification and destination activation.

GDPR/CCPA: AAM supports data subject access and deletion requests via the Adobe privacy service and the AAM privacy API.

Declared IDs and hashed identifiers must follow the organization's consent policy; do not onboard PII without lawful basis.

Trait and segment TTLs act as data minimization controls; shorter TTLs reduce retention of behavioral data.

Third-party data onboarding requires contractual and governance review; data use labels should reflect the provider's terms.

Audit logs and data source access controls track who configured traits, segments, and destinations for compliance reviews.