Blog post

Does IT Matter, Again?

By Andrew White | September 22, 2022 | 2 Comments

OrganizationData and Analytics

The question of, “Does IT matter” has been responsible for pages and pages of argument, both suggesting it does and it does not. It is interesting in its own right that the question remains present in general conversation. Nicholas Carr is probably tickled pink that is book title continues to generate interest (and royalties, I suspect).

Getting Rid of IT

I was moved to blog after spotting yet another article on the same topic in Monday’s copy (September 19th, 2022) of the Wall Street Journal. The article was called, “Get rid of the IT department? Some people think otherwise”. It was inside the C-Suit Strategies periodical. It was based on an earlier article (It’s time to get rid of the IT department) written before by Joe Peppard.

Now to be clear, the original article and this new one was not really about whether IT matters. But the message played on that emotive and eye-catching angle. The original premise of the first article and this new update is that IT as a centralized function is not very useful, and that to realize more value from ‘IT’, it should be embedded instead in business functions. The point being the marketing folks, as an example, know more about what they need than some IT specialist that sits in a central team, remote from that business need.

I love the premise and I love the article. But the hype around shutting down ‘IT’ is getting in the way. Our own research speaks to the real argument hiding in these articles. And the core of the real argument goes to the definition of “IT”. For that we have to go back to Nicholas Carr.

Does IT Matter, Really?

His premise was that when certain IT-enabled work becomes standardized and commoditized, they would be less valuable. This is of course totally obvious. Mr. Carr decided to hype the idea that ‘IT’ no longer mattered as a whole, since ‘IT’ was now so widely available to all. His spin was popular, but over simplified. Some parts of the work that goes by the name of ‘IT’ had tended to commoditized status by the time he wrote is book, but much of “IT” had not. Over the years much technology does through the exact same lifecycle. Duh.

But if you know anything about ‘IT’ you would know that in reality we are being lazy and we really should refer to something more akin to information and technology, or even information, communications and technology. ICT is in fact what many economists use to be more precise. But our own industry has gotten lazy, and that laziness clouds (pun intended) the real point.  IT is not one thing or one kind of work.  So while some must by definition evolve to commoditized status, some won’t.

Look at the cloud.  Infrastructure as a Service, or IaaS, is as close to that commoditized argument from Mr.Carr.  Compute and storage is relatively available and cheap.  But what data you use, what analytics you develop, and what decisions or business processes you follow, is closer to the opportunity for unique, differentiated value-add.  That is where Software as a Service (SaaS) and Platform as a Service (PaaS) play a stronger role.  This is why I have said before that the real battle for the most important parts of the Cloud remain in the future: The Battle for Cloud Shifts to Applications and Analytics from Infrastructure.

Putting IT in its Rightful Place

In summarizing Mr. Pepperd’s article he is suggesting, rightly, that some elements of “IT” are more value-generating if people in business functions can use it directly. This is not the same point that Mr. Carr was making but the headline to sell the copy was similar: Shut down IT.  When business knowledge and speed is needed, that IT-enabled work should be local to the skills and need. Think of self-service analytics, or maybe supply chain data steward (governance), or perhaps data engineering (data management) roles.

But any effort to source such capability at the edge of a business will necessarily lead to risk and increased costs. How many other different models will be redeveloped to define customer lifetime value? Coordinating the work of this IT-enabled work across silos incurs a not insignificant cost. Hence some work that would otherwise lead to duplicated effort should be centralized.

Where IT should Matter

For many, a centralized function might be the IT department, or in Finance. Such centralized IT-enabled work should be that which is suited for reliability, resilience, and repeatability. Coordination across the silos-at-the-edge and the central teams still remains a challenge. And increasingly hybrid or mixed models appear in more mature businesses. Coordination challenges and opportunity will likely outlive any evolution of centralization or decentralization.

So the article plays in the emotive idea of shutting down IT. That’s not the real point. The real point is that you should be deliberate about where work should take place. Some work will likely be undertaken by centralized teams focused on the use of information and technology, or I&T. Some work will likely be undertaken by decentralized teams distributed in business functions using I&T. That should be obvious. What is less obvious are the methods, models and leadership skills needed to make this change happen and stick. That’s why we are here.

Relevant Gartner Research for Data and Analytics:

The Gartner Blog Network provides an opportunity for Gartner analysts to test ideas and move research forward. Because the content posted by Gartner analysts on this site does not undergo our standard editorial review, all comments or opinions expressed hereunder are those of the individual contributors and do not represent the views of Gartner, Inc. or its management.

Comments are closed


  • There’s another interpretation of Nicholas Carr’s point of view.

    He made the case for placing greater emphasis on how business technology investment should focus more on business outcomes (competitive differentiation), rather than merely routine IT service operation resilience (5-nines uptime perfomance, etc).

    In fact, many of the CEOs that have replaced their CIO, or added the role of Chief Digital Officer, are an acknowledgment that Carr’s point is still valid, and the bar of expectation must be raised for whoever is the holder of the now essential IT budget allocation.

  • Andrew White says:

    Hi David,

    Thanks for stopping by and leaving a comment.

    I agree. Nicholas Carr did make that point in his article/book for sure. It probably goes without saying but I probably repeat the same point many times a day, every day, since I have been working at Gartner. It is amazing how such a simple idea remains strong and valid. But if you think about it, the point is not specific to IT. Whatever work we undertake, be it IT-enabled or IT-based, it should always be tied to a business outcome.

    I don’t know about you but I never enjoy spending money on a new roof or gutters on my house. Many of us spend a lot of money on such things from time to time. Such expenditures are often large, thankfully infrequent, yet we rarely physically marvel at the work. Most of the time we sit inside the house or in the yard, and would often enjoy more an investment in a new kitchen, or a new bathroom, or entertainment system, or deck. But some elements of infrastructure are far removed from “outcome”, but they remain just as critical as the kitchen, bathroom or deck. So it is with IT spend.

    Much work can be tied to business outcomes, and some is much harder to do so. This is not a unique phenomena to IT. The manufacturing industry solved this challenge a number of years ago with things like Activity Based Costing and JIT accounting. Even more recently we have seen how cloud computing has shifted IT spend from a capital-related conversation to an expense-related conversation. This leads to yet more complexity with spend and responsibility.

    Anyway, thanks again for the excellent comment. It’s all about outcome.