In this series of Insights, we explore why data cleanup efforts so often fail, despite organizations' desire to eliminate unnecessary data.
Organizations face real and significant challenges, including legal and regulatory drivers like fines, and there is a wider, growing cultural assumption among customers and employees that organizations are merely the stewards, rather than the owners, of their personal data.
What Are the Main Challenges Preventing Effective Data Cleanup?
Although every organization is distinct, the following five reasons most commonly prevent organizations from effectively implementing data cleanup:
The order of this list may at first seem reversed—instinctively, technology might seem to be the main reason why data cleanup doesn't happen, followed closely by culture and process. However, as this series aims to clarify, when accountability and buy-in are taken care of, the other three fall into place and are much easier to tackle. If the first two are left unaddressed, as they typically are at most organizations, data cleanup doesn't happen at all. With this perspective in mind, we will examine the final reason why data cleanup fails: technology.
What is Data Cleanup Technology?
Unlike Enterprise Resource Planning, Human Resources Information Systems or Customer Relationship Management, there isn't a discrete set of features or capabilities that comprise technology useful for data cleanup. It's a dog's breakfast of solutions, nearly all of which were originally created to address other needs (such as eDiscovery, file migration, storage management, privacy or information security), but have evolved into use cases that support data cleanup.
It's important to remember that unless an organization is dealing with the small subset of technology solutions that were expressly developed to support data cleanup, the original purposes of these technologies influence their effectiveness in data cleanup.
For example, there are many tools for data cleanup that began as eDiscovery solutions, which makes sense—data cleanup requires the kind of powerful file analytics capabilities that are the bread and butter of eDiscovery. But many of these tools still retain their eDiscovery orientation around matters and managed review, which is not the most efficient or effective way to address data cleanup.
For example, rather than scanning a shared drive to find all types of sensitive data, they may require users to create a matter to handle categories of sensitive data (e.g., a matter for social security numbers, a matter for hospital billing codes or keywords, etc.). And once the scan associated with a matter is complete, the results tend to be optimized for human review of the data set rather than applying broad brush actions, such as delete all duplicate ePHI or move all files with social security numbers created more than three years ago to a secure location.
This isn't a showstopper by any means but will impact the experience of using the tool to effect data cleanup. And if organizations are considering tools that were developed to address different needs (say one coming out of eDiscovery, one out of privacy and one out of information security), getting an apples-to-apples comparison can be tricky.
What do Organizations Really Need?
Given the complexity in the technology space around data cleanup, how do organizations cut through the noise to determine what they truly require? We suggest the following as a list of must-have considerations when selecting a data cleanup technology partner:
- Scope – does it address unstructured or structured data (or both)?
- Speed – can it scan petabytes per week (not terabytes)?
- Interoperability – how many systems can it scan successfully?
- Cost model – is pricing by volume, per user or system connectors?
- Adjacent capabilities – can it also address other needs, such as eDiscovery, information security or privacy?
For organizations that have spent any time evaluating the market for technology to help with data cleanup, they'll likely know that very few (if any) tools currently available are strong in all these areas, which means that for most organizations, a suitable approach will require more than one tool.
As organizations approach tool selection, M365 is the 800-pound gorilla in the room. If Microsoft can do it serviceably well—or will be able to do so in the next 18 months—don't bother buying a solution to address the same thing. Why pay for capabilities the organization already owns or will own by the time they implement a point solution?
Technology is Important, but Not the Whole Story
In the final analysis, technology isn't the foundation of successful data cleanup, it's the icing on the cake. Although it's not sufficient to effect data cleanup, it's 100% necessary. Just know that what's required for truly effective data cleanup is unlikely to be delivered by a single technology solution; will overlap with what organizations already pay Microsoft for; and will only get them so far. As should be clear from this series of Insights, the people and process aspects of data cleanup are far more predictive of success than a silver bullet technology solution. The right balance of all three, and the adoption of the approach we've laid out, will give organizations the best chance of success in data cleanup efforts.
Now, with knowledge of the intricate dance of technology, people and process in data cleanup, it's time to review the organization's approach. Are they balancing all three effectively? If not, consider this a wake-up call.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.