Modern Infrastructure

Enterprise cloud management: Unlock a strategic direction


Evaluate Weigh the pros and cons of technologies, products and projects you are considering.

Push for centralized IT ushers in cloud-first philosophy at U of Miami

In this CIO Innovator Q&A, University of Miami's Deputy CIO tells how a drive for IT centralization led to a transformative cloud-first IT strategy.

As at so many institutions of higher education, the University of Miami's approach to IT was decentralized and laissez-faire. Let IT freedom reign. If a department had a problem, six different IT teams could be coming at it with five different answers, said Deputy CIO Brad Rohrer.

Rather than lead to new breakthroughs in the delivery of information services, however, the free-for-all ended up hurting IT's many customers -- from the students, faculty, staff and researchers at the school's main campus to the users at Miami's renowned Rosenstiel School of Marine and Atmospheric Science and multi-site medical center. The management part of IT service management was not making the grade.

Brad Rohrer Brad Rohrer

Then two years ago, the university brought in new CIO Steve Cawley and the door was opened to start again. The journey began with an effort to centralize IT service management and is well on its way to transforming the university's IT to a cloud-first organization.

In part one of this SearchCIO Innovator interview, Rohrer outlines the major steps of this IT reorganization, beginning with the consolidation of the university's largest IT units into one central department.

Where do you begin when you want to transform a decentralized IT delivery system into a centralized strategy?

Brad Rohrer: When our new CIO started in June 2011 we started a conversation about IT alignment across campuses. We have three major geographic campuses, and IT was by and large decentralized. Mostly we had campus-based IT, but it was even further 'fractured' within the campuses by school.

So we took our larger IT organizations and basically combined them to form a central University of Miami IT organization. We went from being three or four organizations with 150 to 200 people in the largest one, to a single organization of about 550 people, covering all the campuses. In doing that, what you start to combat is having multiple ways to do the same thing -- different processes and different ways to document incidents; it's inefficient.

With the SaaS model, where you're able to configure but not customize, it forced some acceptance of more of a single best-practice process to service our customers.

Brad Rohrer,
deputy CIO, University of Miami

We wanted to centralize IT services so that we would have a consistent approach to giving people the support they needed. So one of the first initiatives that we did together was look for a service management tool. We were looking for a tool that would help us kind of drive the IT teams together and document and deliver services with some consistency.

Did you bring in a lot of vendors?

Rohrer: We brought in multiple vendors and were already customers of some them. Remedyforce and ServiceNow were the last two we whittled it down to. There was complete consensus [for ServiceNow], and you can't ignore that when it occurs in an organization like ours. Time to implementation was one of the deciding factors. It was cut the PO [purchase order] and three months later you'd have a working incident management system!

Overhauling IT service management is a major project with many moving parts. CIOs have said it's easy to overwhelm IT staff. Why did you start where you did?

Rohrer: First thing we looked at was how we get requests that come in to IT. We wanted to start at a very basic level with incident management. Document consistency was important for two main reasons. The first was to get everybody moving along the same process, and the second was to gauge how productive the staff was that was coming from these multiple different units that existed prior to the consolidation. We needed a way to reassess the volume of work we were doing and think more thoroughly about how we assign work. Ultimately, that enabled us to cross-train a single group of people and become much more productive.

And did you have it up and running as quickly as was promised?

Rohrer: We went through a six-month process of implementation for incident management for all units. I think that was very fast, given the fact that parts of the organization didn't document anything outside of putting lists of incidents in Excel spreadsheets. It wasn't a discipline that was pervasive across the organization. With ServiceNow as the SaaS [Software as a Service] model, where you're able to configure but not customize, it forced some acceptance of more of a single best-practice process to service our customers. That was good because we're still challenged in some ways with the silos we built within IT. So that implementation really got the cross-team effort moving forward.

What were the next parts you tackled and why?

Read more from our SearchCIO Innovators series

Guess? CIO designs app that mirrors work, speeds decisions

CIO's focus on innovation leads to big ideas at IEEE

Agile methodology drives change at Houghton Mifflin

Rohrer: We're in user testing for IT costing and request fulfillment, which will enable self-service. We skipped to IT costing because we have software we use to do chargeback for telecom-related activities -- other departments in the university pay a phone bill to us for auxiliary charges. So to simplify our toolset we went ahead and took the opportunity to retire this old third-party system and move forward with ServiceNow. We're kind of testing in parallel to our old systems now, and it's going very well -- happening on budget and on time! That should be done the first week in December. Change and problem management modules could go live in late January. We almost finished those while we were doing incident.

We'll move on to the inventory modules and hopefully portfolio and project management as well, and we're hoping to get all of that done before the end of our fiscal year.

What's the prognosis?

Rohrer: We've been pleasantly surprised with how easy and straightforward these implementations have been and with how accepting the folks that work for us and the folks that we service are of these tools.

I think it's because they've been kind of starved for this stuff, and so we have a willing group of people using it to get support and a willing group using to provide support!

In the second part of this Q&A, Rohrer talks about the U of Miami's cloud-first philosophy, the people part of the cloud equation and the few things that don't qualify for cloud solutions.

Let us know what you think about the story; email Karen Goulart, Senior Features Writer.

Article 1 of 3

Dig Deeper on Cloud computing for business

Join the conversation


Send me notifications when other members comment.

Please create a username to comment.

Should centralized IT be a goal for every organization?
No. The goals of standardized trouble ticketing and emergency response processes can be attained via a collaborative, shared responsibility model, with the umbrella organization responsible for maintaining and disseminating documentation of the system-wide processes.

Get More Modern Infrastructure

Access to all of our back issues View All