Data Modernization: Turning an Ugly Duckling into a Swan
Data modernization succeeds when it is done within the context of a company's strategic projects.
![Swan with two cygnets swimming on calm water in Staffordshire, England. Swan with two cygnets swimming on calm water in Staffordshire, England.](https://eu-images.contentstack.com/v3/assets/blt69509c9116440be8/bltd857b500264299a8/65a991bf0d8621040a4fc9b2/swan_B5TPEA.jpg?width=850&auto=webp&quality=95&format=jpg&disable=upscale)
On corporate boards and in executive suites, there is verbal acknowledgement that without data quality, security, and integration (i.e., data modernization), there can be no artificial intelligence, which all companies want. Unfortunately, modernizing data is unglamorous, painstaking and often difficult to justify for budget funding.
The reason for those issues is obvious: Data modernization is an infrastructure task, with few tangible business results that anyone can see.
Here are five efficient and cost-effective ways for you to modernize your data, and get corporate buy-off:
1. Incorporate data modernization into important strategic projects.
Any time a mission-critical project involving IT emerges for the company, the project manager on the IT side should look for opportunities to incorporate infrastructure improvements that include data modernization.
If the project involves analytics or AI, hard assets like storage, processing and networks might need to be upgraded, but the IT infrastructure upgrade effort shouldn’t stop there.
As an example, analytics and AI projects require a level of accuracy of at least 70% for general trends data, and that rises to at least 95% if companies are looking for highly specific, actionable results before AI or analytics can be installed in production.
Companies can’t achieve accurate results from analytics or AI if the data they’re using is of poor quality. IT project managers, leaders and CIOs should make this clear to management and their user peers so that tasks like data vetting, cleaning and modernization are included as project tasks.
2. Modernize data for compliance and regulatory purposes.
Regulatory and compliance requirements in healthcare, finance, transportation, and communications all require a cross-section of data for reporting that comes from many different systems. To tap into all that data and combine it in a single data repository for reporting purposes requires system integration. Such system integration requires that data be modernized across all systems into standard forms that can be passed from system to system and consolidated.
To achieve this degree of integration and interoperability, data modernization tasks must be built into compliance and regulatory projects. It is up to the CIO and other IT leaders to explain to management and users why this data modernization work is needed.
It should also be noted that in some cases, outside auditors and industry regulators may want to know from which systems you pulled your data and how you ensured that data from all systems could interoperate and was of high quality.
3. Modernize data for the customer experience.
Customer relationship management (CRM) systems must integrate data from disparate systems owned by many different departments within the enterprise. System integration and data modernization are needed because the end goal of CRM is to deliver to any authorized employee anywhere a uniform, 360-degree view of each customer. For instance, having that 360-degree of view of the customer can give sales a “heads up” before it makes a pitch to a key customer to purchase more widgets. That type of access to data is crucial if service and manufacturing shows sales that the customer has already placed numerous calls to service and has returned the latest batch of widgets because of defects.
4. Modernize data for the supply chain and company acquisitions.
Whether it is adding a new supplier to your supply chain or acquiring an outside company, some of the most painful operational work that occurs is right in IT. This happens because new suppliers and acquired companies run systems that are different from what the home company uses. The end result is that data must be moved from one disparate system to another, and then modernized so the data (and systems) can interoperate and consolidate with each other.
The good thing about these projects is that corporate strategists from the board and the CEO on down have already learned hard lessons about this. They know that although they make the decisions, it is IT that will be burdened with the hardest work of bringing systems together, and modernizing data so it can interoperate.
Realizing the effort that is involved, there will be some upper managements and boards that will just skip these projects and agree to let suppliers and acquired companies continue to run and maintain their own systems for a time. Ultimately, however, the jobs of data modernization and system moving must be done.
5. Modernize data for system rip and replace.
The idea of ripping and replacing legacy systems sounds better in a seminar than it does in practice, because it is in practice where you really see the work involved. The more you look at the amount of work that is needed, the harder it is to justify it.
Nevertheless, there are times when management, users, IT and even the board agree that a legacy system must go. That may happen because its original vendor goes out of business or will no longer support the system. From the board on down, people also know that there is a significant amount of data transformation and modernization that must occur to move data from an older legacy system into a new system. This makes it relatively straightforward to gain management, board and user approval for data modernization, because there is no alternative.
The “catch” for any legacy system rip and replace is that everyone outside of IT (and not just IT) must be strongly in favor of the rip and replace and the effort it will entail before you move forward.
What We’ve Learned About Data Modernization
What we’ve learned about data modernization, or about any kind of hidden IT infrastructure project that improves the usability of data, is that infrastructure tasks like modernizing data can be the difference between companies that are wildly successful and those that are just getting by.
On the other hand, IT leadership also knows that strategic business projects requiring data modernization demand that data modernization be endorsed by all.
It is only then that ugly duckling data becomes the exclaimed, “I never dreamed of such happiness as this.”
About the Author(s)
You May Also Like
How to Amplify DevOps with DevSecOps
May 22, 2024Generative AI: Use Cases and Risks in 2024
May 29, 2024Smart Service Management
June 4, 2024