Blogs

Data Governance as a Precursor to Integration Strategy

By Terence Houser posted 10-27-2017 08:30 AM

  

It’s no secret that integrating disparate data and applications across your institution is difficult and there are many factors to consider before implementation of an enterprise integration platform or solution.  In this HEUG Board Integrations Work Group blog series, we’ve learned that institutions need to reflect on inventories of existing data integrations, the execution of business process reviews, and develop an understanding of possible integration architectures within the context of a technical ecosystem.  In this article we’ll explore data governance and stewardship as another pillar to successful data integration strategy.

The value that data brings to any organization is largely based on its integrity, quality and the ease of access to it, yet too often organizations don’t put a lot of thought behind some of the core components and policies required for a comprehensive data governance strategy.  These include data quality, authorship, storage, retention, security, privacy and standards.  All of these considerations need to work in concert to comprise an effective governance strategy. 

Approaches to data governance can vary widely, but there are common components to many that I’ll describe here.  The first is the need to classify your data.  Not all data needs to be governed, and legal, regulatory and institutional factors will play a direct role in that.  Regardless, it’s important to have a clear understanding of what data you store in your systems.  Is the data classified as highly sensitive, moderately sensitive, data for internal use or data that is simply public?  Defining these classifications according to the laws and regulations in your region will help determine what data can be shared (via integrations or otherwise), with whom and under what circumstances.

Another key component to a good data governance strategy is a clear understanding of data domains and stewardship responsibilities.  Many institutions will have data from a variety of systems and services – student, financial, human capital management, medical etc...  These systems are disparate in nature and will often carry with them very different requirements for appropriate stewardship.  It’s important to recognize these differences and define roles and responsibilities for Data Stewards for each.  Data Stewards will often be responsible for protecting the accuracy, integrity and confidentiality of data within their domain.  They may also have final authority to sign-off on the use of data within their domain.  It’s worth pointing out that, in most cases, IT does not play a data stewardship role, but is considered as a Data Custodian – taking care of institutional information, but not dictating how it’s used.

Next, it is important that the context of data across the institution is well understood and agreed-upon.  More specifically, business definitions of data elements need to be agreed upon before they are integrated and possibly, transformed across disparate applications.  This is most often the responsibility of a Data Steward, but can also be delegated to a combination of integration developers, functional analysts and data modelers.  If not done properly, it can have significant negative repercussions for an institution - especially if the data is not well understood, misinterpreted and/or not used appropriately.

With the proliferation of cloud-based Software as a Service (SaaS) application platforms on the market and the shift this industry is seeing towards their adoption, it is critically important to understand that their use does not absolve institutions of governing their data in the cloud.  Gartner recently stated, “Failure to bring cloud-based applications and data sources into this (information governance) process at the appropriate level will significantly erode any benefits in the areas of trust and reusability gained by the proper management of the on-premises information assets...”  It is not sufficient to rely on SaaS cloud providers to govern institutional data and schools should seek to establish data governance processes and policies for both on and off-premise data stores.

Solid data governance is quite complex and we’ve only scratched the surface of what a comprehensive plan might entail at your institution.  What’s clear is it’s a critically important precursor to the development of an integration strategy.  Yet, many institutions will leap to solving the problem of sharing data with a technical solution only to find the platform they’ve implemented either doesn’t meet the needs of the business or is so comprehensive and complex that it isn’t sustainable.

This is the last blog entry in this HEUG Board Integration Work Group series and we hope that this and all of the subjects we have covered have helped you and those at your institutions.  As always we welcome your input, comments and suggestions for future topics.

Best Regards,

Terence Houser
HEUG Vice President of Technology

0 comments
3 views