DTSS and Matching
DTSS matching is the process of matching a record in a Data Tenant (DT) with a record in a Customer Tenant (CT).
A data tenant is a shared resource that can be accessed by multiple customers. Whereas, a customer tenant stores a customer's private data. Matching is the process of searching a data tenant for opportunities to enrich the data in the customer tenant.
- Automatic - Automatic matches are matches where the degree of confidence in the match is high enough to merge the records without human intervention.
- Potential Matches - Potential matches are matches that have been identified as possibly being a match but a data steward is needed to review and confirm the match.
DTSS supports both Automatic and Potential Matches. The match rules used by DTSS are defined in the data tenant and they cannot be changed by customers or partners. They are derived from the application of Reltio’s best practices developed over time to create baseline matches between your data and the data in the data tenant.
The DT match rules should satisfy most customer business scenarios and will not require any changes. But there may be situations where you need to modify the behavior of the DTSS match rules. This can be done by configuration changes.
The CT and DT are contained in the Reltio’s API ecosystem; the dotted lines indicate the boundaries of the CT and the DT.
The data tenant L3 contains the data model and match rules for the DT and the customer tenant L3 contains the data model and match rules for the CT. The match rules that control cross-tenant matching are defined in the data tenant L3.
The customer tenant match rules only perform matching on data in the customer tenant only, not outside of that. The data tenant generates match tokens in the same way as customer tenants, but they're not compared with the Reltio API. These are distinctly separate data groupings.
How do we cross match between the data tenant and the customer tenant?
This is where the DTSS APIs come in. The DTSS API enables matching between the customer tenant and data tenant by comparing the customer tenants tokens with those of the data tenant.
These are an entirely separate set of APIs that are not part of the Reltio APIs and these must be enabled and configured separately. Reltio provides four pre-built-based tenants: DEA, NPI, 340B, and D and B.
Each supports a limited set of entities and entity types such as Healthcare provider (HCP) and Health Care Organization (HCO). These entity types must exist in both the customer tenant and the data tenant to enable matching and all the fields that are in common must be modeled the same way. For more information on Pre-configured solutions, see topic Reltio DTSS set up.
Using a data tenant using the Search page
After your CT has been subscribed to the DT, the data tenant data becomes available on the left side of the search page under the Tenants section.
The data in the DT can be viewed by selecting the right tenant. Potential matches from the DT can be seen on the CT Potential matches view.
The potential matches view enables you to compare key attributes across the potential matches and make a decision on whether to merge or not to merge. A Data Steward can analyze the options for potential matches and make a decision on which record is the correct match.
In order to enable DTSS matching, the data tenant and the customer tenant must both be registered and the subscription must be created and applied. This is done by Reltio when access to the Data Tenant is licensed.
The behavior of the DT match rules can be configured to some extent in the data tenant subscription. The DTSS match rules to use for matching with the CT can be specified on the synchronization configuration section of the subscription.
If the DT and CT data models are different, the mapping section of the subscription enables you to accommodate the differences. All attributes used by match rules must be available in both the CT and the DT. The AUTOSUBSCRIBE option must be set in the subscription to enable updates on the DT to be automatically evaluated from matching with the CT.
There are several API requests that can be used to work with, troubleshoot, and manage the data tenant subscription. Check out the related topics in the sections below.