Dynamic Communities Magazine

Data. Arguably the most valuable commodity of any organization today. No small wonder then that companies and employees tend to hoard it. Not so long ago this was a major issue as IT departments struggled to keep up with the ever-increasing demand for storage. An entire IT practice started up around the notion of keeping data on the right tier of storage to optimize price versus performance. This is called data management, and generally moves data around based on last access date.

This article is not about data management. Where data management generally fails is in categorizing data. Like most company assets, different types of data have different values. To differentiate, we look at the information that is contained in the data, and so the practice of applying value to data is often referred to as information lifecycle management (ILM). Sometimes it will be called data governance or simply information management.

In recent years compliance requirements have brought ILM to the forefront of IT departments’ concerns, more around securing sensitive data. Many companies decided to treat all data equally and ensure it is all secure. Not a bad practice, though these companies are missing out on the perfect opportunity to start an ILM journey. And ILM is a journey. Every company’s data is in a constant state of evolution, and ILM should not only be involved, but it also should help guide the change. In the case of security, protection generally adds cost. Does spending additional money to secure the non-sensitive data—which is generally the majority—really add value or just overhead?

Maybe a little more relevant to the Microsoft Dynamics 365 for Finance & Operations and Microsoft Dynamics AX world, preparing for an upgrade—though not in the middle of one—is another great time to start ILM as there will already be focus on the data.

On a side note, there is a good reason why Microsoft does not recommend, and their tool does not support, upgrading all data when moving from Microsoft Dynamics AX 2009 to Microsoft Dynamics 365 for Finance & Operations. The schema change is so drastic that the level of effort in simply mapping the data is prohibitive, especially when the volume of customizations implemented in the customer base is considered. The value of data diminishes drastically when a significant effort is required to make it available or maintain it. A data warehouse tool will often already have connectors to ingest the data natively and may even provide self-service reporting, increasing the ease with which the data can be consumed.

Back to the topic at hand. Anyone who has worked in IT intrinsically knows that the only thing worse than no data is the wrong data. Ever show up for a party on the wrong day? Information that cannot be trusted has no value. So why keep it? The hoarder mentality tells us that we keep things because they may someday be useful. I guarantee, that list of people who bought typewriters 20 years ago will never have value again. On the flip side, a list of people who bought cars five years ago may have value, unless you sell hats.

These examples bring out two characteristics of information that can be used to start classifying data: criticality and lifetime. Data that is outdated only serves to bloat databases, resulting in performance issues and the need to sift through “noise”. In the same vein, data that is not directly related to your business may have diminished value. This is not to say that all non-critical data should be deleted. That would lead to scenarios where there is no data, which is still a very bad place to be. Ever find an invitation to a party in the stack of mail, two days late?

New compliance requirements, while very compelling events, are not the only times ILM becomes important. System upgrades, integrations, and migrations are all affected by the volume and quality of data. What is the value of spending an extra five hours on a weekend migrating data that will never be looked at? It would be much more productive to move that data to archive media or a data warehouse. That way the data is still available, say in case typewriters suddenly come back into fashion, without impacting the reliability and performance of more relevant data.

So where to start? To start, IT should not be making the decision on what data is important to the business. Though IT will likely be involved in the ILM process, they are not the driving force. The start, then, is identifying what data belongs to what business owner. Don’t do this at the field or table level; that will be too time consuming and may scare off the business owner. Start at the entity level. Who owns customers? Warranty information?

The knee-jerk reaction may be to keep it all. While there may be valid reasons for this, there also needs to be justification. Buy-in at an executive level may be needed if this is a company wide initiative.

Next, the decision about what attributes will be applied to the data to help with the process must be done. Some possibilities are: criticality, lifetime, sensitivity, legality, and compliance. If GL transactions need to be available to auditors for seven years, don’t waste any time arguing if, just decide where and move on. Legal has a special place in any ILM plan, even if not considered as an attribute of the data. Issues like e-discovery may bring their own requirements, for either retention or deletion, that are not directly related to the business value of the data. Always check with legal before deleting any data.

Now that we know who and what, it’s time for IT to get involved. Just moving or removing the data outright is a bad plan. Create reports showing what will be moved/removed and have it validated. Keep backups. Note that an updated résumé is not a backup plan.

IT may also be able to help business owners understand what data is really being accessed. Adding temporary logging to identify data that is not touched is a great way to locate stale data. Helping business owners understand what a data warehouse really is and how it can be used to access current data as well will go a #pitchthepaperclips | #touchlessinvoiceprocessing www.metaviewer.com
MetaViewer Paperless Automation is designed with your budget in mind, robotically capturing information and automating workflow while offering superior functionality and user experience in a tailored and affordable solution. Paperless Automation for Microsoft Dynamics AX  |  D365O  |  GP  |  SL  |  NAV  |  CLOUD
Are paper-based processes “eating” your profits? long way in reducing resistance to change. Remember, ILM should be something done for the company, not to the company.

Although already mentioned, the importance of data cleanup as part of an ILM process cannot be understated. All versions of Microsoft Dynamics 365 for Finance & Operations and Microsoft Dynamics AX make heavy use of temp tables, some of which are not selfcleaning. Regularly running the provided cleanup utilities should be a best practice. It is also worthwhile adding in some new ones, depending on which modules are used. How valuable are four years of batch job history? If the answer is “not very”, consider the fact that in most installations, the batch job history table is in the top five largest tables.

Every company, whether they know it or not, is already on an ILM journey. Every report and integration are a step along the path, so why not make it official? It may seem daunting at first, but every future upgrade, merger, integration, and report will be that much more efficient to undertake. And you may even save some money on storage.