For Analysts And Managers, These Are The Top Data Management Practices Your Team May Be Overlooking
Every company managing data – especially data associated with Geographic Information System (GIS) analysis and technology – needs to have a strategy in place for consistently maintaining its assets. This is true no matter the size of the data stores. What practices might you be overlooking in your own data management strategy? Let’s explore below.
1. Simplify access to traditional and emerging data.
Being able to quickly and easily access files is a key component of every data management strategy whether we recognize this or not. Why else would companies take the same to organize and clean an enterprise network of outdated and erroneous data? Having files spread across a network can slow a workflow down and complicate decision-making in cases where a user may not be familiar with where necessary is stored.
Simplifying access to spatial data is an important point in every data management strategy we don’t want you to overlook. Companies can accomplish this by keeping their network as condensed as possible, limiting the number of drives across which data is stored. They can also employ specialized filing systems that seek to categorize data based on ways users are most likely to use it, such as by region, project, etc. Our own team has also found success is managing data across networks by employing customized data registries that allow file paths for spatial data like Layer Files to be defined in a single configuration file or spreadsheet and pulled directly into ArcMap with the click of a button.
2. Don’t underestimate the value of categorizing your data.
Organization is not just a buzzword in lifestyle magazines. It is also a key component in properly managing enterprise data. Organizing data can involve any number of aspects, like storing data based on region, owner, date, or relevancy to a project. What all of these systems have in common though is that categorization is a catalyst for successful organization.
Incorporating categorization in a data management strategy allows you to better analyze the context of the data as well as better assign responsibility for maintenance of its quality.
You do not necessarily have to physically categorize these files – simply attributing classifications to files can help to streamline data management practices and increase comprehension of available resources.
3. Make note of metadata and make it a priority.
We will be the first to admit that metadata is not the most glamorous aspect of data management or data analysis. With that said, it is one of the most useful aspects of data outside of the actual content it represents. Metadata makes it easier for professionals, teams, and companies in general to identify and reuse data correctly. By noting the when, where, why, and by whom it was created – as well as any other pertinent details – the process for both using data as well as maintaining its quality becomes much simpler than if it were lacking that information.
Many files lack appropriate metadata because it does take time and genuine effort to include these details. However, the payoff is well worth the work. This is especially true if your company has high turnover rates and finds that a single file can pass through multiple hands during its life cycle. Knowing just the basic characteristics of data can assist new users in properly applying it to their projects.
4. Focus on the quality of data – and not just the quantity.
Data employed throughout a company can originate from almost anywhere. It often comes from third-party vendors or is created internally by geographers, analysts, or the like. Regardless of where it comes from, companies need to remember that maintaining the quality of this data is serious business. Inaccurate or outdated data could result in non-standard formats, misplaced assets, spelling errors, etc. If these inaccuracies are caught too late, teams risk compromising projects for which this poor-quality data has been used and inevitably falling behind.
Cleaning up data regularly by implementing strict data management plans, assigning individuals and/or teams to maintain this quality, and being aware of the data that is actively – and not so actively – being used throughout an organization are all essential leaps forward in supporting and improving the quality of these digital assets.
5. Determine what is not being used and put it on a shelf.
We all have a habit of taking on a hoarding persona when it comes to the data we keep on board.
What if disaster strikes and I need these files? What if we need back-ups of these back-ups’ back-ups? What happens if I get rid of this data and end up needing it in a month? What if?
It’s common to have these fears and to feel as if having this information around will automatically mean it is useful in some way. However, the truth is that much of the data we have available to us is not actually being used productively. It is sitting untouched or worse, still being maintained despite it no longer being needed.
Identifying the data that is not being actively used or at least has not been utilized in a project in x amount of time is the first step to getting a grasp on space that can be cleared up for more fruitful endeavors. While it is not recommended to completely remove this data from the system, it is encouraged to develop an archival system that would allow you to tuck unused data on a shelf to be tossed back into the fray of things if needed. It allows for easy access while discouraging use of outdated or incorrect information.
6. Have a strong recovery plan in place.
In the event of a real disaster – and not a perceived one that convinces you that you still need those files despite their not being touched since 1995 – companies need to have a recovery plan in place. As technology has become more advanced over the years, our reliance on data and the systems in which these files are housed has increased significantly. Resources like cloud technology have made it easier to manage back-ups of our systems, but they are not enough without a concrete plan in place.
Data management strategies should include clear plans for backup and recovery, outlining how often different types of data need to be backed up and what steps should be taken to recover them.
7. Identify what practice is not working and try to figure out why.
While it is imperative companies have a data management strategy in place, this does not accomplish its full potential if the steps outlined in it are not being implemented.
As for those practices that fall through the cracks? They can tell us a lot about our data, our teams, and our needs.
Identifying those data management practices that are not currently working within our company can be hard. It isn’t because we cannot see them – that is easy enough. The difficulty lies in admitting these steps are not working and figuring out what can be done to fix (or even replace) those practices effected. Sometimes the solution is as simple as tweaking your schedule for review. Other times it may be more complicated, involving identifying and addressing training needs of your team.
8. Long-term plans are good, but don’t overlook the short-term.
Around here, we are a fan of goals. Many of the companies with which we work set goals for their data governance and software systems like moving the network from Oracle to SQL Server or moving their Geographic Information System (GIS) professionals from ArcGIS Desktop to ArcGIS Pro. These are the results we want in the end, but we need to remember that the steps along the way are just as important as what we are trying to accomplish.
For long-term goals, breaking them into bite-sized short-term milestones make the task both easier to digest and more likely that your team will stay on track. For example, a move from ArcGIS Desktop to ArcGIS Pro could include inventorying your network, finding and fixing or removing broken data, moving projects that are not solely ArcMap dependent to ArcGIS Pro, and then creating a plan for moving ongoing ArcMap-heavy projects to this new platform. Within each of these milestones, it is important to note the potential for including short-term goals as well – such as the opportunity to archive outdated ArcMap Documents (.mxd) during the move or the potential of bolstering your metadata during inventory.
Another important aspect of a company’s short-term and long-term planning in regards to data management strategy is establishing a clearly defined maintenance schedule for reviewing the data on an enterprise network. While this process may be deemed more short-term, it is necessary to keep long-term projects and projects in working order. Having a plan for it all is essential to success.