When the Open Cloud Manifesto was unveiled recently by IBM et al I wrote "The (Cloud) Bastards say, Welcome"
And I invited several cloud pioneers who have been at it - delivering cloud based products and services or helping evaluate and nurture them for several years - to discuss the manifesto and what they have learned in Cloud Computing over the last few years.
This time it is Eric Dirst who, as CIO at DeVry and previously at Sirva, has been a passionate adopter of Cloud solutions.
“My epiphany about Cloud Computing came in 2002. For years I had been the victim of countless business rants that IT is “not nimble enough”, “not focused on speed to delivery”, and “unable to turn on a dime” with the changing business landscape. When I stopped trying to defend IT, and looked at the business complaints objectively, I realized they were correct.
IT spends a ridiculous amount of each calendar year working on things for which the business finds zero value – upgrading package software releases; testing and deploying package software patches; testing and deploying upgraded hardware; testing and applying security patches or changes to comply with the latest vulnerabilities or changing audit landscape (SOX, PCI, HIPAA, etc.); etc. etc.
What did all these activities have in common? They are all IT “plumbing” activities for which the business users derive little to no enhanced capabilities. Having the plumbing is of value, but it’s not something the business actually “values” – they expect that level of delivery. They want more. “More” to the business means quick ability to change workflows, scale for growth and shrinking times, and adding features/functions on a regular basis to reduce costs or enhance revenue. Of course, some of you would argue that the software upgrades delivered value, but I’d ask you to go back and audit your last major software package upgrade and focus on what additional process, efficiency, revenue enhancing or cost reducing functionality you were able to deliver through that upgrade. I’m guessing you already know the answer, and so does the business – nada, zilch. Most upgrades are just done to keep the company “in support” with the vendor, sadly enough.
So, what to do about this situation? In 2002 I realized that this kind of model was ridiculous and there had to be something better. Thankfully, that was when one particular form of Cloud Computing called Software as a Service (SaaS) was just starting out. [By the way, I hear some of my peers and many vendors still using SaaS and ASP interchangeably - they really should study SaaS multi-tenant architectures closer and leverage the resultant cost savings]
Over the next few years I worked with my business partners to move us to the SaaS model for a number of business functions: Sales Force Automation (SFA), Customer Relationship Management (CRM), Payroll, Compensation Planning, Talent Reviews, 401k management, Travel Booking, Expense Reporting, Purchasing, Training, Knowledge Management, Security Penetration Testing, and probably a few more I’ve forgotten by now.
What did this net my company? Well, I eliminated the following:
- Hardware/software for development, test, staging, disaster recovery and production environments X the number of systems above (14+)
- Hardware/software maintenance, and the associated resources that would have had to do this maintenance
- Resources needed to patch testing/deployment, software upgrade testing/deployment, hardware upgrade/deployment, performance testing, etc.
More importantly, the elimination of all those activities freed up my staff to focus on configuring these SaaS environments to meet the needs of my business customers. Instead of dealing with the plumbing, we were now focused on the business process activities…because that was all we could configure in these SaaS environments. The business was elated, because now we were talking a language they understood – namely business process and what is the UI, workflow, and reports delivering.
Of course, I still had the Integration IT efforts to manage, and the associated internal hardware/software issues to deal with for my Integration environment. However, the integration was easier since the SaaS environments all supported standards such as Web Services with published API or SOA environments. Even better, I contracted all our vendors to conduct annual SAS70 Type II audits. These reports were used by our SOX auditors to reduce our audit fees since they could now limit their testing based on the vendor results.
My only gripe was that I couldn’t move more environments, faster, to the SaaS model. SaaS solutions are still uneven when it comes to areas such as Finance (AP, AR, GL), billing, pricing configuration, imaging, and customer/partner portals. However, I predict this will dramatically change in the next 2-3 years – if for no other reason than the current SaaS vendors will have to look for new revenue streams, having saturated the SFA/CRM, Purchasing, and HR markets.
I’m often amazed to hear of CIOs who won’t look at SaaS solutions. They must derive some satisfaction I never found in managing the infrastructure for all those development, test, staging, disaster recovery and production environments. Many of my peers will cite risk areas like security, service level agreements (SLA), data ownership, recovery point objectives (RPO), and vendor longevity. These are all valid concerns, but I’ve found that good contracts can minimize these issues to manageable risks, or eliminate them altogether.
I’ve also found that often the vendors are better at security, SLAs, and disaster recovery than my own IT shop. The good vendors already store data encrypted, do encrypted backups, transmit data over the web encrypted, report on numerous SLAs, deliver >99% availability, and regularly test disaster recovery by actually failing over the production environment. And I am pushing them to seek out clouds for their own processing and storage needs. So, their environments get better each year, from a functionality, security, and resiliency factor – all without me paying an extra dime (well, until contract renewal usually at 3 years – for which I’ve already negotiated a max 10% increase).
What could be better than that? I know the answer. Nothing. SaaS is here, and anyone who doesn’t take advantage of it is costing their company money, time, resources and limiting their business customer’s agility.
My fellow CIOs – better to hear that from me than from your CEOs.”
