Skip navigation

The more I think about what I wrote on March 3 about herding cats and such, the more I realize how personas (more about development personas some other day!) or Archetypes in enterprises have fueled this Oklahoma Cloud Rush. I was stuck in traffic this morning, and I found myself typifying the personas of the drivers. You have The Jammers…they are the people who drive all the way up to where the left lane merges into vapor at full speed and slam on their brakes and head fake (using right bumper and fender) drivers into letting them break in line. I wish there was a huge iron boot that would drop out of the sky and crush them and their car and then kick the steel (and plastic now) waffle all the way to the scrap yard! Then you have The Switchers..they are the people who constantly switch lanes in traffic that is moving at the same exact tempo trying to pass the entire herd of commuters, often driving 20 miles to get 5 miles down the road. I think they are already in the purgatory of self induced frenetic disillusionment, so that’s swell enough for them. THEN, you have The Turtles…actually dangerous because they do everything like they have their heads hidden within their bodies…Once again, punishment enough, I just wish law enforcement would enforce the minimum speed limits in major metropolitan areas..

There are personas, yes archetypes, in enterprises as well. They not only impact cultures, they actually dictate where and how change in enterprises is instantiated. I love that word! I will explain:

  • The “Informal Buyer”- YIKES! These people are called Maverick Purchasers in the other building where the CFO is king and Purchasing is the Royal Guard. One of the “Killer Apps” some analysts are looking for in the “Cloud Management Suite” would result in an outcome that would put these folks on the bench, and take their locker keys away.(More about that in another future note) BUT, the flexibility and responsiveness (speed of fulfillment) of a well executed “cloud life cycle” would provide necessary and sufficient order while allowing this psychology to persist (it is beneficial in “evolutionary” cultures) long after the freedom to “stand up” anywhere your credit card reaches goes away…AND the same well executed “cloud life cycle” would help them make the transition from outlaw to champion as “Cloud Evangelists” with some cloud based wins to point to. Mavericks to Champions…love it.
  • The Formal Buyer- Critical role and actor in an evolutionary, revolutionary or a “greenfield” organization (I now have to write something about organizational archetypes….later #3). In an evolutionary organization, these people WANT to exploit and expand, in a planned and governed manner. In a revolutionary organization, these folks need to balance the culture of liberal enablement with good governance and “day two” operations. In a cloud “greenfield” organization, these are likely director level infrastructure or application owners. In every organization, The “Governator”. They already understand the anxiety of being between the LOB demander and the Infrastructure provisioning logjam. These folks would WANT to be involved in every aspect of the cloud implementation, from discovery and classification, to service catalogue “creation and maintenance”, to provisioning and day two operations. These folks WANT a well executed “cloud life cycle”, and know what one looks like.
  • The Architect- I know….the creepy arch villain in the Matrix trilogy- strangely similar. They have been involved with every cycle of evolution, and get engaged eventually in every revolution (weird…Matrix Revolution..Matrix Evolution..I may be febrile again…). They are the penultimate creator/influencer (and may be the ultimate creator/influencer where the CIO has “KPI/scorecard myopia”) for cloud exploitation in any organizational archetype that they exist in. They will want to be involved or dictate the definition of every aspect of cloud implementation, and are valuable in that role due to the way their circuit boards are cross wired into BOTH the existing Enterprise Architecture, and the FIVE YEAR Technology Plan, and they actually know the latter is an oxymoron.


Yawn, traffic is about to start moving (yes, I just blasted this on my BleakBerry waiting on the wreck that occurred when the Jammer and the Switcher decided to try automobile fusion on the ramp up to the Beltway)..I would be GLAD to argue about this with you, but you know that already…cross town traffic, it always makes me grumpy!


To weigh in, comment below! I'll begrudgingly read your thoughts. To suggest a rant - email me at


Sometimes, I find myself wandering through a Home Depot on Saturday, and I see something that catches my eye. Perhaps a seed packet for spinach. Perhaps a new pair of garden shears. I pick it up, and decide to treat myself. Sure, they are useful. But, I’m not sure they are necessary. They are impulse purchases.

I find it hard to believe that anyone would buy cloud computing on impulse. I’ve talked to dozens of people and companies about their cloud purchases – and all are highly premeditated. Rare is the case where someone opts to burst into the cloud on a moment’s notice.  So why all this bursting? Are people not planning, even a little?

How do you plan for external public cloud usage? It’s possible. It’s being done by enterprises everywhere.

First, in the past, we’ve talked about understanding who your users are, and what services you’d like to offer.

Then, we’ve talked about estimating the capacity you’ll need to address these requests – and the onslaught of new ones you’ll get when people realize just how easy it is to get what you need from this new-fangled cloud thing.

So, how are you going to fulfill all those needs? Well, you can estimate what you have on-site by way of resources. Not enough? Think you might need more for certain times? Well, take that list of services you’re offering and categorize them as “must be in my datacenter” and “could potentially move outside.” The criteria include:

  • Compliance requirements – both operational and regulatory
  • Security requirements
  • The complexity of configuration needed
  • Proximity to large data sources (moving lots of data tends to be inefficient, both financially and practically)
  • Splotchy consumption of resources
  • The price of external resources

Having isolated the cloud services you can potentially send to the public cloud, it’s time to investigate options. My experience in the past few months is that service providers are rapidly increasing the breadth of offerings available – so this is an exercise worth repeating every few months. Map the appropriate services to your needs, and do some financial calculations with your trusty budget team to ensure you’re making sound financial decisions, and implement your plan.

Sometimes using public cloud resources is an ongoing strategy for certain workloads. Other times, it’s overflow. Either way, the time to decide is before you need it. Because we all know, you don’t make great shopping decisions on an empty stomach.

To weigh in, comment below! I'll begrudgingly read your thoughts. To suggest a rant - email me at


Cloud reminds me of the Oklahoma land rush with people pushing and shoving and tripping over themselves to get out ahead and stake a claim. While it's obnoxious to see so many IT vendors out their doing their own land grabs ("cloud personal printing", anyone?), it's more unnerving to see so many IT organizations who are getting caught up in the flow and may be inadvertently allowing their companies to end up with undisciplined approaches to adopting cloud-based computing.


IT organizations are traditionally very good about adopting new technology in a careful and considered manner -- walking that fine line between reaping the business benefits of new technologies and mitigating the security, stability and compliance challenges of that new technology. But cloud computing, as with mobile technology, enables business users to begin making local technology decisions. It's great that public cloud services enable informal IT consumers (such as application developers or business staff) to quickly request and begin consuming cloud-based services without the involvement of corporate IT. It's a great thing. Unless those users are making not-so-great choices.


There are a number of dangers to allowing your company to inadvertently take a pell-mell approach to adopting cloud-based computing. At the very least, you will under-achieve the benefits of moving to cloud if you're over-provisioning servers . And on the other extreme, you could end up with highly sensitive information out in the cloud that should really belong, for compliance reasons, behind your firewall.

So, the challenge is, how does IT take an appropriate pace and strategy to adopting cloud-based computing while ensuring that business users aren't doing an end-around? Here are a few ideas.

  • First, don't allow a situation where business users are by-passing enterprise IT as it relates to public cloud services. Business users (or even those application developers who need a short-term QA box or short-term SharePoint server) are inevitably going to find that public cloud services are a very attractive offering for a quick, informal purchase. You know it's going to happen, so get involved. Show IT leadership by helping to negotiate and oversee those public cloud requests from the line of business. Work with a preferred service provider to set up good default self service options for those customers. It's a win for the business folks because they'll have access to these services and IT will already have done all the heavy lifting. And it's a win for IT because you'll be channeling all that potential hidden and "unsupervised" usage into a place where you can monitor what's going on.
  • Establish clear cut policies around what is and isn't appropriate for public cloud usage. What kinds of applications and use is okay and what needs to remain behind corporate firewalls for policy or compliance reasons? Developing an in house application and need a place to test with live data? Probably not okay.
  • Clearly communicate what your overall cloud adoption strategy will be. Are you looking at an evolutionary approach to cloud, where you'll first focus on automating datacenter and then consolidating servers via virtualization with an overall plan to move to cloud? Or are you more transformational, with the intent to implement PoCs and or even introduce a private cloud for particular domains or types of internal servers and applications? Sharing this roadmap with the business helps them understand that IT is taking a leadership role around cloud and it gives you the ammunition you need to manage customers expectations about what you're doing for them.


Unless you get out ahead of these hidden, informal usages of public cloud, you'll never really effectively manager your company's adoption of cloud-based computing. It will be like herding cats - except that you won't even know how many cats there are to begin with, so you won't ever really be able to tell if you're even close to succeeding.

To weigh in, comment below! I'll begrudgingly read your thoughts. To suggest a rant - email me at