When the mobile phone in your average employee’s pocket does far more than the clunky computer sitting on that same employee’s desk at work, it’s easy to understand the prevalence of shadow IT and line of business (LoB) purchases made outside the purview of the CIO’s team.
But here’s the thing: Most companies aren’t addressing the core issue. It’s not the technology tools that are the problem. It’s the data feeding those tools. And for the most part, that data is bad. Dirty. Incomplete. Disconnected. Ungoverned.
“We Have 47 Ways to Spell Tennessee”
For example, research done for ASUG’s Enterprise Information Management (EIM) event found that only 17% of finance professionals “feel extremely confident in the accuracy of the data in their reports.” If there’s one department you’d really want to feel rock-solid about the integrity of its data, shouldn’t finance be near the top of that list?
Data management isn’t a very sexy topic, I know. But if data is indeed the oil—or even gold—for modern-day enterprises, as many have suggested, then why aren’t more companies treating their data like a precious resource, rather than a commodity? And if companies are going to continue to purchase new software and tools to add to their enterprise portfolio—introducing even more data silos—it’s likely that there’s a flood of bad data on the way.
It’s not that any sane enterprise sets out to become awash in ruinous data. But everyone knows where the problems are, which can lead to some outrageous workarounds. “We now have a data dictionary of 47 ways to spell Tennessee,” said one respondent to our BI+Analytics Conference study, regarding the costly effects of dirty data.
There’s Gold in Them Thar Hills
The good news is that all this is fixable. Not easy, but it can be done. According to ASUG research of SAP ERP customers across a wide range of industries, there are three main areas that need addressing ASAP. Let’s take a closer look.
Area 1 to Address: Determining data ownership
Here are two questions to start with:
Who does everyone think owns the data inside companies today?
Which is followed by:
Who should actually own the data?
It appears that everyone (except those who work in IT) thinks that IT is and should be the data overlord—the owner, governor, administrator, massager…you get the point. Again, research from ASUG’s EIM eventfound that two-thirds of the 123 respondents (62%) said that IT is primarily responsible for managing their EIM strategy, followed by C-level/executives (30%) and LoB departments (3%). The results suggest that enterprise data- and information-management processes do not have executive-level support (e.g., authority, oversight, resources), which can be an onerous task for IT departments to accomplish all on their own.
As part of ASUG’s Undercover CIO blog series, our anonymous executive laid out the situation and remedy clearly:
One of the biggest misconceptions surrounding data ownership is that it should be the responsibility of IT teams. Except for their own data, IT teams are not and should not be the owners of enterprise-wide data. These teams, instead, should be the data facilitators. They should help people understand where the data is coming from as well as its importance.
In my opinion, data is owned by that person, that team, or that function that put it in the system. They should include not just data that is relevant to their processes, but they also need to enter data that feeds an end-to-end process. It must be thought of as an asset for the entire organization.
Area 2 to Address: Sunsetting the silos
Organizations continue to struggle with integration and eliminating data silos within technology solutions. That’s according to ASUG research’s year-over-year EIM surveys. In addition, we found that connectivity—related to both technology and the people trying to use the technology—is the greatest barrier to the success of EIM governance processes.
Integration gaps across systems, no plans for consolidation of data sources, and expanses of data silos all continue to plague enterprises. This results in unhappy people unable to collaborate.
“Many separate systems and processes for managing the data, in turn, leads to departments that make decisions that impact other areas and do not effectively communicate across channels,” was a familiar theme among the respondents.
Among this universe of SAP customers, ASUG research finds that only 30% of companies believe they have effectively designed SAP user roles and governance within their systems. The financial results can pile up: Gartner reported that poor data quality and practices can add up to an average financial burden of $15 million a year.
Area 3 to Address: Prepping for migration - and SAP S/4HANA
All established enterprises, regardless of which ERP system they are running now, face a migration of their core to next generation systems at some point in the next few years. In an SAP context, we can argue about the exact numbers, but it’s safe to say that there are lots of SAP customers who are preparing for a move to SAP S/4HANA in the near future (preparing being a vague and all-encompassing word in the context of a looming 2025 “go or no go” date). This type of project has huge data implications. Customers should be asking:
- What data should come along for the ride and which shouldn’t?
- What should be archived?
- How will the data you are taking with you get migrated?
- And how will the entire move affect custom processes in place today that employees want to keep using?
According to the respondents to ASUG’s EIM event study, there are still more questions than answers at this point.
- Nearly half (47%) of the 96 respondents did not know which migration tools they are going to use;
- Almost one in three organizations have not yet chosen a data-migration approach;
- And 75% do not have a budget in place to clean and migrate legacy data or are simply unsure of their plans.
Those numbers don’t inspire much confidence.
The good news is that there are peer resources and online and in-person events to help enterprises move forward with this all-important and foundational work that needs to be done. Don’t wait any longer. Your data (and users) will thank you.