Process automation has many benefits, but getting there can be tricky. Start with small, targeted projects.

Mike Fratto, Former Network Computing Editor

December 15, 2011

3 Min Read

InformationWeek Green -Dec. 19, 2011

InformationWeek Green - Dec. 19, 2011

InformationWeek Green

InformationWeek Green

Download the InformationWeek December digital supplement on IT process automation distributed in an all-digital format as part of our Green Initiative
(Registration required.)
We will plant a tree for each of the first 5,000 downloads.


Mike Fratto

Mike Fratto

In our InformationWeek 2011 IT Process Automation Survey, we found that many companies are already automating their IT and business systems. Thirty-six percent of respond- ents say they've fully deployed IT workflow or process automation tools, and 32% say they have a project under way.

That means they're reaping or will be reaping automation's benefits, such as faster response to service requests, fewer configuration errors, and reallocating staff to other projects. However, getting those benefits isn't easy. Implementing automation is extremely difficult with legacy IT systems, resulting in situations where a system requires human intervention to be integrated with the rest of IT. And some tasks are simply too complex or costly to implement.

Sound familiar? If you've ever worked with a network or systems management package, you know that early successes occur when everything--all the software and operating system versions and configurations--neatly lines up with the vendor's recommended requirements, which usually cover a small subset of current software, never mind older versions.

More often, you can quickly integrate a portion of your systems, and then you're left doing the remainder on your own. The vendor's product will take care of the easy 80%, leaving you to focus on the harder tasks. Sadly, that remaining 20% can be difficult, and that's where you really need the help. There's a reason automation software is often sold with professional services. It's plug and pray, not plug and play.

You don't have to automate your entire IT operation. Start with small, targeted projects such as deploying a database by automating the deployment and configuration of the server, storage, networking, and RDBM software into one step and grow from there. Do your homework, including a thorough process analysis documenting existing processes and systems. Then do a gap analysis, comparing where you are now with where you want to be. Now you're ready to proceed with an automation plan.

Nobody likes doing planning, inventory, and analysis. It's expensive and time consuming, and most people think they already know everything there is to know about their processes, so why waste time documenting them? Well, you likely don't know everything. Unless your organization has an ingrained process-oriented culture, your IT processes are probably semiformal at best, and you can't automate a workflow semiformally. Unlike people, computers do only what you tell them.

You must inventory your IT systems, their APIs, management interfaces, and input and output formats. There are few meaningful integration standards. Having an XML interface isn't sufficient. Knowing the integration features means you can determine how each system can integrate with the others and where to focus.

You could throw caution to the wind and hack together an automation system. It might even last for a while, but eventually it's going to become fragile and complex. Do yourself a favor and do a thorough systems analysis first. The time and money invested will be worth it, because you'll be able to buy products that fit into your environment rather than having to adapt to a vendor's requirements. You'll thank me later.


Download the InformationWeek December digital supplement on IT process automation.

About the Author(s)

Mike Fratto

Former Network Computing Editor

Mike Fratto is a principal analyst at Current Analysis, covering the Enterprise Networking and Data Center Technology markets. Prior to that, Mike was with UBM Tech for 15 years, and served as editor of Network Computing. He was also lead analyst for InformationWeek Analytics and executive editor for Secure Enterprise. He has spoken at several conferences including Interop, MISTI, the Internet Security Conference, as well as to local groups. He served as the chair for Interop's datacenter and storage tracks. He also teaches a network security graduate course at Syracuse University. Prior to Network Computing, Mike was an independent consultant.

Never Miss a Beat: Get a snapshot of the issues affecting the IT industry straight to your inbox.

You May Also Like


More Insights