Across B2B customers, we keep seeing the same pattern: teams invest heavily in data ingestion and integrations, but when they try to build audiences, they hit friction. Segments don’t return the expected profiles; opportunities don’t link to the right people, and account conditions appear to “silent fail” because the attributes live in the wrong place in the data model.
A practical approach to validating B2B audiences
In Adobe Real-Time CDP B2B Edition, most of the confidence should come from being able to answer simple questions with data:
-
“How many people are associated with target accounts?”
-
“Who are the decision makers on deals in stage ‘4. Propose’?”
-
“Who engaged with last week’s webinar and also belong to high-value accounts?”
These questions feel straightforward, but if audience sizes do not match the numbers in the source system, that discrepancy can reveal implementation issues.
Why this happens
B2B use cases are inherently multi-entity. Answering any of the questions above requires traversing relationships across people, accounts, opportunities, campaigns, and buying groups simultaneously. But the primary tools teams use to activate that data (Segment Builder, journeys, destinations) still require a single root entity, almost always the person profile.
Without a shared, explicit understanding of which entity should own each attribute and which relationship path Segment Builder is actually following at query time, every team ends up developing its own conventions. The result is fragile audiences, inconsistent counts, and persistent gaps between what the business requested and what the platform actually returns.
What this framework addresses
This framework provides a repeatable, structured approach to B2B data modeling and audience construction in Real-Time CDP B2B Edition. Specifically, it:
-
Establishes the appropriate entity for attributes across people, accounts, opportunities, and related entities.
-
Maps the relationship traversal paths Segment Builder follows so teams can predict and not guess what a segment will return
-
Aligns business questions directly to technical implementation patterns, eliminating the translation gaps that cause "mystery" audience discrepancies
-
Creates shared conventions that scale across teams and use cases, replacing ad hoc approaches with a consistent, auditable methodology
The goal is straightforward: when a marketer asks "everyone at Tier 1 accounts with open pipeline," the platform should return exactly that and the team building the segment should know exactly how and why.
The root cause is almost never “bad data” in the abstract. It’s misaligned B2B modeling and traversal:
-
Attributes that should sit on the Account are instead stored on the Person.
-
Relationship types (primary contact, owner, sponsor) aren’t modeled consistently, making role-based audiences brittle.
-
Opportunity and campaign membership are captured, but the connections between Person → Account → Opportunity → Program aren’t set up the way Segment Builder expects.
When that happens, marketers and operations teams end up:
-
Manually exporting lists from Customer Relationship Management (CRM) instead of trusting Real-Time CDP
-
Duplicating logic in multiple systems: Customer Data Platform, Marketing Automation Platform, Customer Relationship Management system, Business Intelligence analytics tools.
-
Wasting time debugging “missing” profiles that are actually just on the wrong entity or path
This guide exists to reverse that pattern and perform health checks as well.
We’re defining a practical B2B audience framework for Real-Time CDP that:
-
Aligns data modeling with how Segment Builder traverses relationships (Person ↔ Account ↔ Opportunity ↔ Programs/Lists).
-
Shows the canonical paths for the most common B2B questions (people by account attributes, people by role, people by opportunity conditions, behavioral + B2B combinations).
-
Clarifies when to use Segment Builder, when to lean on Buying Groups or Adobe Journey Optimizer (AJO) B2B Journeys, and when to rely on real-time streaming segmentation.
How Segment Builder follows the B2B data model
In Segment Builder, the folder hierarchy mirrors the B2B data model. The path starts at Profile, branches into B2B, moves through Person Key and Account-Person Relationships, and then reaches Account. Each folder takes you one step closer along the relationship path between a person and the account they belong to.
Finding the right people at the right accounts
When you're building B2B audiences in Segment Builder, one of the most common needs is pulling a list of person profiles based on something you know about their account or their role within it. Here are two patterns worth knowing:
1. People at accounts with specified account attributes
Say you want to reach everyone connected to enterprise-tier accounts in a certain region, or accounts flagged as "at risk" in your customer success data. This pattern lets you filter on account-level attributes - region, industry, tier, segment, customer status and then address every person tied to those qualifying accounts.
Example use case: "Show me everyone connected to accounts in EMEA that are Tier 1 and currently active customers."
-
Segment Builder path: Profile > B2B > Person Key > Account-Person Relationships > Account
-
Canonical XDM (Experience Data Model) traversal: Profile → Account-Person Relation via personKey.sourceKey; Relation.accountKey.sourceKey → Account.accountKey.sourceKey
2. People with a specific account relationship
Sometimes it's not the account attributes you care about, but a person's role relative to that account. This pattern targets people by their relationship type: primary contact, account owner, executive sponsor, advocate, and so on.
Example use case: "Find every primary contact across our top 100 accounts so we can send a renewal update."
-
Segment Builder path: Profile > B2B > Person Key > Account-Person Relationships
-
Canonical XDM traversal: Profile.b2b.personKey.sourceKey ← Account-Person Relation.personKey.sourceKey
Finding people by opportunity and account
Beyond account-level attributes and relationships, many B2B use cases hinge on opportunities and how people and accounts connect to them.
1. People who belong to an opportunity
Sometimes you need everyone directly tied to a specific deal, filtered by their role in it.
Example use case: Find everyone whose opportunity role is "Decision Maker."
-
Segment Builder path: Profile > B2B > Person Key > Opportunities > Opportunity-Person Relation > Opportunity
-
Canonical XDM traversal: Profile.b2b.personKey.sourceKey ← Opportunity-Person Relation.personKey.sourceKey; Relation.opportunityKey.sourceKey → Opportunity.opportunityKey.sourceKey
2. People tied to opportunities with a specific role and opportunity condition
This pattern layers a role filter on top of an opportunity condition, combining who someone is on the deal with where the deal stands.
Example use case: Find Decision Makers for opportunities at stage "4. Propose," or Champions for opportunities above a value threshold.
-
Segment Builder path: Profile > B2B > Person Key > Opportunities > Opportunity-Person Relation > Opportunity
-
Canonical XDM traversal: Profile → Opportunity-Person Relation via personKey.sourceKey; Relation → Opportunity via opportunityKey.sourceKey
3. People whose account has an opportunity with specific attributes
Here the opportunity condition doesn't sit on the person directly; it sits on their account. This pattern surfaces people simply because their account has a qualifying deal, regardless of whether the person themselves is tied to that opportunity.
Example use case: Find all people tied to an account that has an open opportunity at stage "4. Propose."
-
Segment Builder path: Profile > B2B > Person Key > Account-Person Relationships > Account > Opportunities
-
Canonical XDM traversal: Profile.b2b.personKey.sourceKey ← Account-Person Relation.personKey.sourceKey; Relation.accountKey.sourceKey → Account.accountKey.sourceKey; Opportunity.accountKey.sourceKey → Account.accountKey.sourceKey
Finding people by behavior, account, or opportunity
1. People with supported experience behavior and a B2B condition
This pattern combines standard behavioral signals—page visits, content downloads with any of the B2B relationship paths above, letting you target based on what someone did and who they're connected to.
Example use case: Find people who visited a product page recently and are associated with a target account or opportunity.
-
Segment Builder path: Profile attributes / Experience Events + one supported B2B path above
-
Canonical XDM traversal: Person → Experience Events plus Person → Account-Person / Opportunity-Person relationship
Other types of B2B audiences
1. People associated with a campaign or program
Example use case: Find people who are members of a specific campaign or program, or who have a particular member status.
-
Segment Builder path: Profile > B2B > Person Key > Campaign/Program Members > Campaign/Program
-
Canonical XDM traversal: Profile.b2b.personKey.sourceKey ← Campaign/Program Member.personKey.sourceKey; Member.campaignKey.sourceKey → Campaign/Program.campaignKey.sourceKey
2. People associated with a marketing program and account condition
This pattern joins a campaign/program membership path with an account condition— useful when you want engaged people, but only from accounts that also meet a business criterion.
Example use case: Find campaign/program members who belong to accounts in a target industry, or accounts with a target opportunity stage.
-
Segment Builder path: Profile > B2B > Person Key > Campaign/Program Members > Campaign/Program plus Account-Person Relationships > Account
-
Canonical XDM traversal: Person → Campaign/Program Member → Campaign/Program AND Person → Account-Person Relation → Account
3. People in a marketing list
Example use case: Find people who belong to a particular marketing list, or lists with a specified status or type.
-
Segment Builder path: Profile > B2B > Person Key > Marketing List Members > Marketing List
-
Canonical XDM traversal: Profile.b2b.personKey.sourceKey ← Marketing List Member.personKey.sourceKey; Member.marketingListKey.sourceKey → Marketing List.marketingListKey.sourceKey
Choosing the right approach
Audience creation
-
Need a list of people tied to accounts? → Build a People audience with account attributes.
-
Need a list of accounts? → Use an Account Audience.
Journey orchestration
-
Need to address people belonging to a specific account? → Use Buying Groups.
-
Need to journey B2B people directly? → Use the AJO B2B Person Journey.
Streaming segmentation
- Working with event attributes on a short lookback window (e.g., 24 hours)? → Pass the data via real-time streaming rather than relying on batch profile updates.
Put the attribute where the traversal expects it
When you need a list of people based on an account-level condition, the filtering attribute must live on the account profile, not the person profile. If the attribute is missing from the account profile, fix it at the data ingestion layer by updating the B2B data model and account schema so Segment Builder can traverse the relationship correctly.