Twitter Facebook Linkedin

Moving from Run to Grow: The New Model of IT Operations – By Chris Bedi

To share...Tweet about this on TwitterShare on FacebookEmail this to someoneShare on LinkedInShare on StumbleUpon

Author: Chris Bedi, CIO at ServiceNow

Competition is brutal. Regardless of industry, if you can’t innovate and operate faster, your company is destined to be outmaneuvered. But too many companies are stuck in the old way of doing things where 40 percent of skilled workers’ time is consumed by administrative tasks. Their time is spent on running the business instead of on changing the business. It’s the CIO’s job to break this cycle and help our companies be free to focus on accelerating business velocity. The alternative is getting outflanked by your competition.

The CIO has the ball

Rethinking how work gets done rests squarely on the CIO’s shoulders. CIOs are uniquely positioned to lead our companies to achieve the highest levels of speed and productivity. IT platforms drive every business process, employee interaction, and customer experience. IT is also one of the worst offenders when it comes to balancing run-grow; Gartner reports that IT organizations spend 66 percent of their resources on day-to-day operations— purely keeping the lights on. So, the transformation needs to start with IT’s own operations.

No one else will make this happen, but we can’t just dip our toes in the water. Incremental improvement is not going to achieve the level of speed that our companies require to stay competitive. We must jump into the deep end by automating EVERYTHING. The new model of IT operations is “the best service is no service.”

Lead through change

A short while ago, I looked at our IT spend and decided enough was enough. We were spending 74 percent on running the business and just a quarter of our resources on innovation. My IT team was bogged down with unstructured processes, fragmented data and manual reporting, and we had no time to even think about innovating.

Deep down, we all know that throwing more bodies at this problem won’t solve it—intelligent automation will. McKinsey estimates that about 60 percent of occupations could automate at least one-third of their activities with technology that already exists today.

In plain English, we looked at everything we do and said: Let’s automate the !*?# out of everything we do. What we discovered is that 84 percent of what we do has potential to be at a higher-level automation, so we set out to automate all of it.  The real number has to be 100 percent, but like most, we are likely constrained by our own imagination.

My team was skeptical. There isn’t a good answer to what IT operations is supposed to do once we’ve automated everything. The truth is that the path to a “no service” model is littered with questions for which we do not have all the answers, yet. CIOs need to be open about this. We don’t know what’s on the other side of this hill, but we’re going up it. The IT team will lead through change or wake up one day and be ”yesterday’s” IT department.

Embrace the long game

Getting to the point where the organization is truly “no service” is a process. Service automation can’t just be about quick wins and incremental improvement. It’s about a service automation strategy that creates a competitive advantage over the long haul. You won’t jump from manual to machines completely managing every service in one step.

IT needs to begin with the processes it knows best—IT processes. Start with end-to-end processes were structured tasks are abundant and automation will alleviate the most workload for the IT team. This may be provisioning VM’s, patching machines, installing. Automate these entire processes—the provisioning, the management, the reporting, the scaling up and down. Learn from them, then tackle more.

Using service automation, we’ve been shifting our IT model from running the business to growing the business. In 2016, this shift resulted in 43 percent of our IT resources going to innovation. The time we have freed up in our IT team allows us to experiment with how we can help the rest of the organization apply intelligent automation to their own business processes. For instance, financial close, procurement, on-boarding, marketing activity planning and contract execution.

Mind what you measure

When you move from humans to bots doing the routine tasks, and eventually managing entire processes on their own, what you measure must also change. CIOs still need to measure productivity, velocity, financials and the user experience, but the metrics will be different.

Where productivity used to be based on the time spent running the business vs growing the business, your metric will now be the percent of work eliminated or fully automated. Velocity was measured in issue and request cycle times; now it will be percent of work proactively executed or percent of processes with a cycle time of near zero, i.e. nearly instantaneous. CIOs measured financial impact as IT OpEx as a percent of revenue; now it will be margin contribution. And the standard for user experience was the Net Promoter Score; now it will be the percent of issues resolved via self-service.

Overall be bold in what you take on, but realistic about how long it will take to achieve your goals. We’re not yet at the point where IT can spend 80 percent of our time on innovation and 20 percent on running the business. But we’re close to a 50/50 split and we’re continuing down the path of relentless automation.

To share...Tweet about this on TwitterShare on FacebookEmail this to someoneShare on LinkedInShare on StumbleUpon
Chris Bedi

About Chris Bedi

Chris Bedi joined ServiceNow in September 2015 and currently serves as ServiceNow’s CIO. Prior to joining ServiceNow, Bedi served as CIO of JDSU from August 2011 to March 2015 where he was responsible for IT, Facilities, and Indirect Procurement. Prior to JDSU, Bedi held various positions at VeriSign from April 2002 until August 2011, including CIO, VP Corporate Development, and VP HR Operations. Bedi began his career at KPMG Consulting from June 1996 to April 2002. He holds a Bachelor’s degree in Computer Engineering from the University of Michigan.