In most organizations, IT holds a lot of data – but that data isn’t always effectively shared among all divisions that need it. In this guest post, tech writer Aidan Grayson has some advice for how companies can tear down dreaded data silos.
The much maligned data silo occurs when one business unit within an organization has a lot of data that another one doesn’t and vice versa. This situation creates inefficiencies because Department A doesn’t know what Department B could easily help them with.
And while we’re all aware of the problem, fixing it is a different story – mostly because it’s hard. Divisions have protocols in place and everyone’s supposed to follow them. Getting everyone to change, therefore, is easier said than done.
But if you’re dead set on destroying data silos at your organization, here are four ways to fix the problem once and for all.
1. Create a culture of cross-divisional communication
OK, we all know it’s a cliché to say something’s “lost in translation.” But that’s exactly what happens when marketing wants to know something IT has had a handle on for years.
While it’s true that educating these disparate departments on terminology and processes may not be 100% possible all the time, they should at least know that it’s ok to share information when others come looking for it. Not only that, they should know that it’s worth taking the time to explain – and, conversely, listen and ask questions – when someone from a different department comes hunting for information.
Just knowing that data collaboration is part of running a productive, competitive organization is an important first step toward tearing down silos. By no means is it the end-all, be-all solution, but it’s definitely necessary.
2. Continuously plug knowledge gaps
Many companies nowadays take pride in rigorous employee training programs that teach all new hires the basics by which every department operates.
Marketing hires learn what IT and QA are up to. Development learns how marketing keeps track of leads. Departmental jargon becomes company jargon, and everyone is able (mostly) to speak the same language.
In institutions where comprehensive training is the norm, there are fewer knowledge gaps. And fewer knowledge gaps mean better communication and less data exclusivity among business units.
Of course, if you’re like most organizations, a training program like this may not be standard procedure for new hires – at least not yet. That’s why it pays to determine what the knowledge gaps among departments are and come up with ways to “plug” them.
If not through instituting an employee training program, it could be through periodic, mandatory seminars that address common problems. The better your divisions are able to communicate, the more data they’ll be able to share.
3. Build a comprehensive IT blueprint
Think about the office where you work. Building it involved a concerted effort among carpenters, electricians, plumbers, and others. But would you have hired them to build it if they couldn’t produce a blueprint of how everything was going to fit together?
Of course you wouldn’t, so why do you let your organization run without one? Creating an IT blueprint is essential, and it should do more than just address knowledge gaps. It should seek to prevent them from occurring at the outset.
A good blueprint will involve every business unit and define the methods by which they work together to share and integrate data.
4. Build integration into the process
Speaking of integration, it should be a part of your process from the beginning.
The point of the blueprint, after all, is to create a plan that doesn’t allow data silos to develop. We know they’re bad, and we know integration can kill them. You just have to make sure all your business units understand the value of approaching each task as part of a cohesive, company-wide effort.
To achieve this end, organizations will need to modify how they tackle new projects. What’s a project for IT shouldn’t just be a project for IT, but something marketing is aware of so they can blog about it or use it when creating new collateral.
By the same token, development should be on the same page as sales. What better way to sway prospects that are on the fence than to assure them that the new functionality they’re after will be available within three months?
Of course, these kinds of things should happen automatically. They should be endemic, and it’s data silos that prevent them from being so. The reality, however, is that you do have the power to destroy those silos and use better communication to increase market share.
It will take time, sure. But using time wisely delivers positive ROI.
About the author: Aidan Grayson writes about enterprise software, business IT issues, and B2B cloud computing. He contributed this post on behalf of the R language and big data brain trust at Revolution Analytics.