Data asset-based strategies must be reliable, repeatable, and produce beneficial results that are well beyond their costs.
Typically, organizations derive their IT strategy based on known business need at a given point in time. Applications are created to provide answers to specific questions.
When I first learned IT, we started with some basic linear programming languages…Basic, Fortran, etc. Task #1 was to create a logic flow diagram. At some point in time, developers realized that many of the code pieces and parts can be reused both within the current process as well as by other processes. Instead of programming in a straight line, they began conditionally looping. When they noticed that the sub-processes that were being called would also work for other programs, they developed reusable classes and object orientation.
The IT industry grew up focused on process. Getting from A to Z. From single use applications to reusable classes to standardized libraries; IT evolved. However, until recently, the way data was used did not keep pace. IT was primarily bent toward process and building systems to perform those processes.
The growth of the internet as a business platform has spawned a different way of viewing IT. Focus is being re-directed from procedure, and the importance of data strategy is becoming clear.
When we look at all of our IT efforts, data is a common element. Data is the “content” that is shared across the internet as well as the blood that flows through the veins of our business applications. Applications use, generate and transform data. More and more we are realizing that from an enterprise perspective data that can be shared and integrated across processes delivers better value for a business. IT is evolving from being Application-Centric to being Data-Centric.
Business units A, B and C all have procedures for working with a specific piece of data. Each with their own bend for how they can get the most out of that data. Traditionally, each application project is funded and managed independently. Likewise, their infrastructures and databases are developed in a silo fashion. Although many applications need to access the data from individual application databases, those connections are only considered after the fact. APIs are created to link applications or share data as their need is realized. This results in “API Spaghetti” which is complex to manage. Additionally, many applications store like data locally in each of their databases. This data redundancy is costly from a storage perspective and also leads to poor data quality as each application alters that data based on its own needs.
When application development is managed from Data-Centric perspective, Data and content become the cornerstone upon which development projects base their architecture. Principles for managing enterprise level data are designed first. This is followed by engineering development platforms and infrastructures upon which applications are built which can leverage shared data.
If a well considered strategy is in place for managing this shared data, then cost of developing the applications that use them decreases. As business requirements change, and applications are updated and re-written, the data remains viable for use. This exponentially decreases the cost of altering the data when new solutions arise. Additionally, a strong data strategy minimizes data inaccuracy. Data quality is Maximized for enterprise.