Unified Data Model and Introduction of Buying Groups
A Unified Data Model (UDM) serves as the foundational, customer-centric physical schema for the Fusion Unity Data Platform within Fusion Marketing. Its purpose is to provide a single structure to ingest disparate source data, unify identities, compute derived metrics, enable precise segmentation, and drive consistent activation to downstream systems. UDM is aligned with the Oracle Fusion Data Model enabling efficient transfer of data from Fusion Applications.
Primary Components:
- Data Object Groups - Profile, Behavioral, Transactional, Product, Other: Logical groups of tables/objects that keep profile attributes, event/behavior records, transactional data, product catalogs, and miscellaneous look-ups organized for performance and traversal.
- Master Entities — Golden records that represent unified individuals/accounts; created through identity resolution and used as the canonical anchor for 360° profiles, joins, and aggregations.
- Events — Time stamped behavioral records that capture who/what/when/channel/context (clicks, purchases, page views, etc.); events link back to profiles/masters for behavioral analysis and real time triggers.
- Intelligent Attributes — Pre-computed or derived metrics tied to master entities (examples: propensity scores; recency/frequency; loyalty tiers; Customer Lifetime Value (CLTV) and Average Order Value (AOV) style aggregates) used directly in segmentation and scoring.
UDM is the foundation that supports advanced analytics, AI/ML: designed for scalability, high performance, and governance, making it possible to support advanced analytics, AI/ML, and real-time orchestration with a single source of truth.
The main change in this release was to introduce the concept of the Buying Group. This required a set of physical tables to create and their relationship to other existing ones. As such with 26A, the following UDM objects where changed:
| Change type | Data Object Group | Table Name | Comment |
| new table | Other | Marketing Message | This data object stores the information about the Content that a customer sees. For example, an email message or a text message. |
| new table | Other | Marketing Web Page | This data object stores the information about Marketing Web Pages. It has attributes like Name, title, type, description etc. |
| new table | Other | Product Recommendation | This table is a metric table designed to hold information about recommendations for certain products associated with accounts. |
| new table | Other | Product Usage | This table is used to store information about the usage characteristics for a particular subscription, product and/or account. |
| new table | Other | Tenant Preference | new table. E.g. for storing the tenant corporate currency code. |
| new table | Profile | Buying Group | A Buying Group is linked to a specific account and product. It can have multiple Contacts (called members) associated with it. |
| new table | Profile | Buying Group Buyer Role | This object describes the anticipated role(s) that members of a Buying Group play during a purchasing process. |
| new table | Profile | Buying Group Demand Type | Demand Type lists the different values used in the GTM process (Acquisition; Retention/Renewal; Cross Sell, Up-sell; etc.). Depending on the circumstances of the Customer, such distinctions may not be applicable or additional dimensions or Demand Type values might be necessary. |
| new table | Profile | Buying Group Member | Buying Group Members Table associates contacts to Buying Groups based on the criteria established in the Buying Group Profiles. |
| new table | Profile | Buying Group Persona | Persona table to store Persona list to be selected while creating the Buying Group profile. |
| new table | Profile | Buying Group Persona Definition | This table stores all BuyingGroupDefinition information like ID, Name and Description. |
| new table | Profile | Buying Group Profile Demand Type | This table associates BuyingGroup to Demand Type. |
| new table | Profile | Job Department LOV | This table stores user's or customer's Job Department Taxonomy. |
| new table | Profile | Job Function LOV | This table stores user's or customer's Job Function Taxonomy. |
| new table | Profile | Job Level LOV | This table stores user's or customer's Job Level Taxonomy. |
| new table | Profile | Title Repository | This table is a centralized data object designed to store the standardized job titles generated from both 1st party internal and 3rd party external title mapping processes. |
| new table | other | Account Product Fit | Scores each accounts fit for each product using firmographic, subscription-history features; labels are derived from subscription product IDs in the query. |
| new table | other | Visitor | This table stores information about anonymous visitors in early interactions (e.g., Mobile App or Web Push) when contact details are not yet available. Once user information is identified, the records can be updated or linked to customer, contact, or contact point records. |
| new column | Other | Organization | One new column BrandIdentifier. |
| new column | Product | Category | Three new columns for Level10Cateogory to extend the hierarchy with the 10th level. |
| new column | Profile | Account | Three new columns HealthScore, Industry, Region. |
| new column | Profile | Customer | One new column Region. |
| new column | Profile | Master Account | The same three new columns HealthScore, Industry, Region as in account. |
| new column | Profile | Master Customer | The same one new column Region as in customer. |
| new fk | Behavioral | Event | Three new foreign keys to FusionUser, MarketingWebPage and Visitor tables. |
| new fk | Other | Marketing Message | New foreign key to Organization table. |
| new fk | Product | Product | No change three new foreign keys are pointing to product from new tables: Buying Group, Product Recommendation, Product Usage. |
Business Benefits:
- Single Customer View: Builds comprehensive 360-degree profiles by merging structured and unstructured data, enabling deeper insights into behaviors, preferences, and histories across touch points.
- Eliminated Data Silos: Centralizes fragmented data from CRM, e-commerce, and other systems into one authoritative source, ensuring consistency and reducing redundancies across teams.
- Foundation that supports advanced analytics, AI/ML: designed for scalability, high performance, and governance, a single source of truth.
- Enhanced Segmentation: Supports granular, real-time audience segmentation based on demographics, transactions, and behaviors for precise targeting.
- Improved Personalization: Powers tailored experiences and campaigns using unified insights, boosting engagement and conversions.
- Operational Efficiency: Automates data unification and quality control, minimizing manual efforts and accelerating workflows for marketing and sales.
- Consistent Omnichannel Delivery: Maintains context across channels like email, web, and in-store, preventing disjointed interactions.
- Stronger Data Governance: Centralizes privacy controls, compliance, and accuracy in one model, simplifying audits and security.
Steps to enable and configure
You don't need to do anything to enable this feature.
Tips and considerations
- Architect checklist (quick items a data architect should verify):
- Clear master record strategy and stable canonical identifiers for joins and aggregation.
- Event model that preserves context and links reliably back to masters.
- Availability of precomputed intelligent attributes for performant segmentation.
- Support for export/APIs and low latency endpoints if real time activation is required.
- Review the tenant’s default data model
- Use the Data Model configuration UI page browse and review Unity UDM predefined physical data model data objects, attributes and relationships to see whether you need to add anything.
- Use the Master entities UI to review the predefined master entities (Master Customer, Master Account) de-duplication and promotion rues, to se whether they fit your data quality reality
- Do the “first publish” (and review what becomes locked)
- Oracle explicitly calls out that before your first publish job, you should review items that cannot be changed afterward (for default objects), including:
- Bucketing strategy and partition strategy
- Default relationships (can’t be edited after first publish; you can create new ones later)
- Default attributes/objects can’t be deleted; ID fields can’t be changed
- Also: you need to run the first publish job before creating Data feeds (ingest jobs) and Segment Deliveries.
- Oracle explicitly calls out that before your first publish job, you should review items that cannot be changed afterward (for default objects), including: