A while ago I had the opportunity to sit down with Microsoft’s CIO Tony Scott to talk about “mission critical” applications. (See “Interview with Tony Scott on Mission Critical in Today’s Enterprise.”) Since that interview, I’ve done some additional thinking on the topic and wanted to share that here. Tony offered some great insights and, though his comments were mainly directed at CIOs, I think they are appropriate for the broader audience. Let me start with three key takeaways:
· Understand the cost and the benefits when defining something as mission critical
· Keep the infrastructure refreshed and current
· Don’t just look at the application—look at all the dependencies
Now let’s take a look at the term itself. It is a part of the language of the enterprise but are we all in agreement on the meaning?
What is Mission Critical and Who Gets to Define It?
Wikipedia defines mission critical this way:
Mission critical refers to any factor of a system (equipment, process, procedure, software, etc.) whose failure will result in the failure of business operations. That is, it is critical to the organization’s “mission.”
I see two problems with this definition. First, who gets to define which applications qualify as mission critical? And second, what exactly constitutes failure?
The answer to the first question is easy in my view. It’s the business that defines what is mission critical because, frankly, it’s the business that will suffer if something goes wrong. Tony alludes to this in the video when he talks about the fact that, in so many cases, there hasn’t been a check-in with the business in a very long time. What he’s really saying, though, is that this is about partnership between IT and the business.
What Does Failure Really Mean?
The second question is more problematic, though. If I am a marketing manager, kicking off an email campaign is critical to my business. If something happens and emails are delayed, does this constitute a failure? What if the campaign is tied to a product launch?
How about this? I am in charge of catalog sales. Something happens and suddenly my website is not accessible for a period of time. Is this a failure? Well, it might be if this happens during the holiday season.
Everything Comes at a Cost
As I said earlier, there are three very good takeaways in the video, and all have to do with the partnership between IT and the business. First, you need to have a common understanding around the cost and the benefits when you define something as mission critical. Tony cites the example of IT being able to recover something very quickly but the actual business practices don’t support that.
Second, it’s really important to keep the infrastructure refreshed and current, and ensure that things are working together as designed. CIOs are often asked to reduce cost but they may reduce it in areas that will affect critical applications.
Finally, when considering new applications, you must make sure that there is alignment between IT and the business. Part of this requires that you look closely at all the other dependencies that go along with the new app (what the cat dragged in). Sometimes these dependencies are not well understood in the initial review.
Cloud to the Rescue?
There’s no question that deploying and maintaining mission critical applications is hard. And now the choice of moving to the cloud presents more challenges and more opportunities—challenges in that developing for the cloud is unfamiliar and opportunities in that you could realize cost savings, greater performance, scalability and availability.
How does your business determine what is mission critical? I’d love to hear your comments.
The opinions and views expressed in this blog are those of the author and do not necessarily state or reflect those of Microsoft.