Kyle is a part time Computer Support Technician, SCCM newbie and student with the college. But The Gute is much more than that. He is the symbol of a PC management strategy that reduced support demand by over 30% last year and pushed IT customer satisfaction through the roof using Microsoft SCCM.
PC Management – Current State
When I took over as CIO just over two years ago faculty and staff were ready to throw all of IT out. Not a day went by the President didn’t hear at least one complaint from faculty or students about the computer problems. IT service sucked as was our team morale.
In a supreme example of penny-wise, pound foolish, IT adopted a PC management strategy that stopped using DeepFreeze to manage the 55 computer labs at 10 campus locations opting instead for a manual GPO approach. And, as a technical college where our lab loads change every semester to support curriculum, IT was regularly failing on those configuration changes which often weren’t resolved until 3-4 weeks into the semester. You can imagine the impacts to instruction and faculty satisfaction.
I honestly had never seen configuration management or a patch management model this bad anywhere in my career. It wasn’t for lack of tools or money. IT had tried Altiris for configuration management years earlier but couldn’t achieve the results common to most every organization using this mature product. Citrix, another mature solution, was chosen but failed to solve the change and patch management issues and only added more. The next approach involved Microsoft System Center Configuration Manager (SCCM) but it was not implemented properly resulting in Microsoft SCCM being nothing more than an impaired version of Ghost.
IT Turn Around
So I challenged our User Services team to develop an improved PC management approach using Microsoft SCCM that would do three things:
- Image any device PC, laptop, tablet or server using SCCM from bare metal without errors
- Accomplish fully automated imaging, monthly patch management, and software deployment using SCCM
- It had to be “Gute Proof”
Why “Gute Proof”
The User Services team had been reduced in size over the years to help deliver the promised ROI of Altiris, Citrix and SCCM which included using more part-time technicians supplemented with temps during our summer project periods. So, whatever they came up with using Microsoft SCCM it had to produce consistent, repeatable results regardless of which technician caught the assignment – and most were students or temps. Gute Proof means it has to allow for a low-skill, no-skill, person to be a champion every time – error free which is more than Idiot Proof. And as my team reminds me – CIO Proof.
Service Improvement Plan
Achieving the goals for PC management were not easy. We started by helping the team see that the vision for change and patch management was absolutely possible and within their ability to deliver. I started with a single first focus – get the McAfee compliance report to single digit every day and keep it their. It took about 37 days to achieve this first milestone in the service improvement plan, but we found and fixed at least one problem every day such as a Bad Pool Caller BSOD in Windows on multiple systems that were all different error codes, and therefore different problems. Morale soared and faculty trust began to return.
The team rallied. We went back to DeepFreeze, rebuilt Microsoft SCCM, implemented Pre-Boot Execution Environment (PXE Boot) and Microsoft Application Virtualization (App-V) and fully automated the bare metal imaging process.
You can see from this short clip of the initial imaging sequence it is totally Gute-Proof. Since “The Gute” only has to know 4 things:
- Press F12
- The PXE Password
- Is this a Lab or Employee Device
- The Device Name
The team packaged all 185 applications for deployment to any desktop with user self service. They implemented monthly configuration and patch management updates which were so successful the server admins stopped work on their solution and piggy-back on the desktop teams work.
Service Quality Success
We can image any device from bare metal in 45 minutes or less from a PXE Boot, remotely. And a full lab in about an hour – unattended. The team has gotten so good we have adopted a true ITSM service restoration model where we re-image or swap the device if it takes more than 15 minutes to troubleshoot. That way the user is back to work now and IT takes care of it offline.
Now the team, Gute included, spend their time delivering high touch customer service and user support. Our part timers now do the re-packaging of apps for Windows 7 and are testing automated user state migrations and enhanced software distribution. Now, our faculty focus on teaching and the needs of our students.
The details of this story will be featured later this month at SunGard’s Summit where I hope to meet many of you. If not, I will post my presentation here afterwards.
PS – And Kyle, he has become one of our rock stars of customer service and doing well in his studies.