The Difference Between System of Record and Source of Truth

Previous Post: Organization Dynamics and Data Strategy

In our earlier posts, we discussed the three main components for adding value to data: Business Processes and Analytics which are both supported by Data Services.

As we design and architect the data strategy components, we will invariably face this critical question:

From where should we source the data?

To answer this question, we usually comb through the inventory of all the applications in our enterprise trying to identify the single data source. In doing so, we are often faced with two challenges:

1. In some cases, there is no single system that has the complete data set. Bits and pieces of the required data is stored in multiple systems. To make matters worse, we cannot easily stitch the data together to make a complete meaningful data set.
2. In other cases, the same data with the same attributes are available in multiple systems. But, the data values are different and we are not sure which among them is accurate.

If we start designing integrations or building reports or architecting applications without answering the data source question, we will deliver sub-optimal solutions in siloes. Each solution will tell a different story because the underlying data is not the same. And, we lose credibility with our business users.

To answer the data source question, we will have to understand the concepts of System of Record and Source of Truth and how they are same or different.

This distinction is critical because the below Data Services are dependent on this definition:

• Integrations
• Conversions
• Quality
• Master Data Management (MDM)
• Governance
• Discovery
• Coverage

Definitions:

System of Record (SOR): A system of record is the authoritative data source for a given data element or piece of information.

Stating it a bit differently, it is the data repository where the data object, as a whole or specific attributes of a data object, are maintained. This maintenance includes data creation, updating, modifying and deleting.

The SOR is the data source that is used for audit or regulatory reporting purposes.

Source of Truth (SOT): The source of truth is a trusted data source that gives a complete picture of the data object as a whole.

We will discuss a couple of scenarios – one where the SORs and SOT could be different and another where both of them could be same.

SOR different from SOT

Let us take the example of items (or parts). An item is a component of a product that our organization designs, manufactures, and sells to the customers.

In the product life-cycle, an item is born in a design/engineering system like PLM and goes through the business processes of manufacturing in an ERP application, sales in a CRM application, and finally shipment in a TM application. During each and every step of the business processes, new attributes could be added or existing attributes could be modified.

So a single data object like an item has attributes stored in different systems. There is no single system that gives the complete view of the item, i.e. we do not have a single SOT.

At this point, we may hear traditional MDM proponents say that we need to create an item master MDM system where all these attributes are managed centrally. For the past couple of decades, many organizations did embark on this journey to create a central item master.

Having said that, if our organization portrays characteristics of a type 2 organization as explained in our earlier post on Organization Dynamics, this type of central MDM solution rarely works. Due to the dynamic nature of our organization, it is not possible to sustain this MDM solution. This solution has significant governance overhead and cannot quickly respond to ever changing business environment and imperatives.

The best option in this case would be to create a SOT by compiling item attributes from the different item SORs.

It is this SOT that would be the source of data for BI/Data Science and Data Integrations.

This is what Gartner refers to as a Consolidation MDM.

SOR same as SOT

For this scenario, let’s consider another data object like the chart of accounts (COA). The COA is normally maintained in the financial/general ledger module or a dimension management system.

This data is normally managed centrally by a select few team members. The data is published to all the subscribing processes and applications.

In this case the System of Record (SOR) and Source of Truth (SOT) are the same.

Delivering Sources of Truth

Most of our users work on business processes that span multiple applications. The expectations from the business users are:

• Clearly identify all the SOTs
• Single system to access all the SOTs

This is where a data repository helps us in delivering SOT data to the business users. This could be an operational data store, data warehouse, or data lake.

This is also known as an Analytical MDM.

We have discussed the nuances of SORs and SOTs. If we recollect our earlier post on “Data as an Asset Strategy” and it various components and sub-components, we can now conclude that the Analytics Data Services is critical to delivering SOTs to the business.

The operational data store or data warehouse or data lake is part of Analytics Data Services.

Whether it is Business Intelligence, Data Science or Data Integration to downstream systems, the Analytics Data layer plays a central role in this strategy.

Next Topic: Analytics Data Services