Data integration has come a long way in 15 years. While it used to be a multiyear and hugely costly undertaking feasible for only large enterprises, midmarket companies now have access to the same level of integration capabilities thanks in large part to cloud technology.
However, there is still a long way to go, especially in post 2008 recession times. With expectations higher than ever before to produce value and market pressures demanding that businesses move at a faster pace than ever before, small and midsize organizations especially don’t have enough IT resources, cash or time to devote to their integration efforts.
Seamless integration is critical for organizations in their ongoing quest to do more with less, however certain hiccups can end a partnership before it even has a chance to begin. Following are three roadblocks for businesses to avoid in charting a smooth course for data integration.
1. Misaligned Timeframe
Simply put, unless both partners involved in the data exchange equally prioritize integration efforts, the partnership will be dead in the water before it starts. Data integration requires that both IT departments be in the room together on a regular basis to keep the project moving forward, validating progress and resolving problems. Given the resource constraints that most IT departments are operating within, if another project takes precedence over configuring partner data connections, integration efforts can be placed on the back burner indefinitely.
2. Differing Levels of Technical Capabilities
There is no doubt the integration space is undergoing a significant transformation. However, as with any period of rapid development, there are growing pains. Businesses are evolving to adopt the latest in integration technology at different paces, which leads to the inevitable circumstance in which one partner is using “old school” integration techniques, which require the highest level of technical and professional expertise, and another partner is using cloud-based integration methods that business users can use with minimal IT technical skills and facilitate hybrid integration and human workflow.
To the above point, this is going to have serious consequences in terms of project timeline, as those businesses using older technologies require significantly more time and resources to create data connections as each connection must essentially be created from scratch. More agile businesses cannot—and will not—wait around.
3. Automated Processes—Or Lack Thereof
It’s becoming increasingly important that organizations have the ability to automate certain aspects of the integration process, namely exception handling. As we know, data integration is complex. Businesses and IT departments need to access a number of different integration methodologies in order to seamlessly orchestrate tasks in a process flow; to do this well, automated exception handling and human interaction need to part of this workflow, so as to fast track exceptions and decision-making.
Consider the following. Let’s say a medical billing code was entered incorrectly on an invoice sent to the insurance company. Without automated exception handling, every instance of bad data such as this must be corrected manually—and at $60+ per error to correct, it’s not cheap. The cost and productivity savings when a user is automatically alerted to the error has the ability to correct and automatically reenter into the system is huge.
The key to avoiding these nonstarters is to level set expectations at the beginning of any partnership. Oftentimes as is the case with data integration, assumptions are made regarding requirements and limitations in terms of file format or access to files behind firewalls, to name just a few common examples. Even something as simple as the right person not being present for a kickoff call can have lasting implications on the success of a partnership. As the integration space continues to mature and evolve, these conversations at the onset of any agreement will be key.
