What 'digital transformation' really should mean

We're inundated with exciting new enterprise tech. But for it to amount to more than the sum of its parts, we need to address longstanding problems

What 'digital transformation' really should mean
Credit: Thinkstock

Unlike most buzz phrases, "digital transformation" keeps gaining popularity through the years. This is mainly because from the very beginning, it has invited whatever meaning marketers wanted to slather onto it. Today, digital transformation seems to mean that together cloud and social and mobile and big data and AI and IoT and devops (plus whatever other else you want to throw in) yield a tipping point where businesses discover new revenue opportunities and become qualitatively more efficient. Or something like that.

Sure, today's explosion in enterprise tech is unprecedented, but the promise of transformation almost always disappoints. Part of the reason is that there can be no static endpoint -- by the time you assemble perfection, it’s obsolete. The other difficulty is that the same old problems that have persisted through many generations of technology continue to hold us back.

To thrive amid all this glorious new tech and digitally transform, we need to focus on solving the foundational problems first. And if we nailed just this handful that would truly be transformative:

Shabby security. Lack of technology is not the problem here; human behavior is. If companies patched software promptly, successful attacks would decrease by a magnitude. If users were trained to stop doing stupid stuff like getting phished or downloading and installing malware masquerading as anti-malware, we’d see far fewer breaches. Multi-factor authentication and better event monitoring and new ubiquitous encryption schemes can help. But until enterprises make security basics a C-level priority, we’ll be continued to be hobbled by rolling security disasters.

Data incoherence. The cloud may actually be making this problem worse. As lines of business are empowered to adopt SaaS applications, new data silos spring up. Unreconciled data across the enterprise -- duplicate yet slightly different customer or product records, for example -- remain a plague. Many attempts have been made to implement what was once called “master data management,” but the discipline to maintain it always seems to run out of steam.

Service orientation. The wisdom of deconstructing applications into API-accessible services is almost self-evident, and now the microservices trend is reinventing the idea again. Microservices architecture is an application architecture. But at a certain point when enterprises adopt microservices at scale and start sharing them across applications and departments, we’re going to run into the same problems faced by SOA (service-oriented architecture). Those problems were largely political: who owned what service, what responsibilities came with ownership, by what methods sharing was enforced, and so on. A failure to solve them killed SOA.

Regulatory madness. This may be the biggest roadblock to the cloud. The tangle of regulations in the United States alone keeps organizations from storing data outside the data center, even if the cloud might actually be a safer place for it. State-by-state variations in healthcare regulations, for example, have added all kinds of friction to digital health care efforts. And the differences in privacy regulations from country to country in the EU are legendary. You want transformation? Rationalizing regulations that block transformative technology adoption would be a great place to start.

Software pricing. The lion's share of enterprise IT budgets still goes to the big software companies, whose complex licensing schemes and software audits rake in billions upon billions. Yes, we’re gradually moving to subscription models, but some seem to willfully emulate the complexity of licensing plans. No wonder more companies are adopting open source -- or building their own applications rather than buying.

Inaccessible data. Inside enterprises, access to data needs to be closely controlled, but all too often valuable data remains inaccessible to those who are actually authorized and could benefit enormously from it. Carefully enabling that access requires hard work. In the public sphere the problem is simpler: All public data should be accessible via well-formed APIs. Businesses should have easy access to useful data collected by federal agencies and state and local governments. Today, even policymakers need to dig for it. While we’re at it, let’s make the Web machine-readable.

This is obviously a partial list. I’m sure you have your own favorite sticking points to add, and you’re invited to do so in the comments section.

I don’t mean to minimize the advances we’ve seen -- and not just technological ones. For example, over the past decade technologists have become embedded in lines of business, to the point where even talking about the divide between “business and IT” seems passé. Technology literacy in general, propelled by everything from free MOOCs to clever mobile apps you can’t stop using, keeps rising year over year.

But transformation? You can’t just chant “the Internet of things” and expect wholesale change. I hold out hope that the increasing sophistication of users will help surface old problems that demand roll-up-your-sleeves commitment to solve, because without that we'll never make it to the next level.

From CIO: 8 Free Online Courses to Grow Your Tech Skills
Notice to our Readers
We're now using social media to take your comments and feedback. Learn more about this here.