By: Pierre Moncassin
A couple of months back, Rohan Kalra and I looked at how modern IT organizations are helping their clients trade control over compute assets for control over business processes.
In the cloud era, we explained, IT can give developers and users immediate access to standard IT services that are accessed and then scaled on demand. By accepting this increased standardization, their users are gaining unprecedented agility at the business level.
For this to happen, though, IT consumers need to understand the business value they can reap by accepting increased standardization. If your job is to help them reach that understanding, here are three other points that might help you make the case:
1. The False Allure of Control Over Physical Assets
It’s good to be clear that in moving to standardized commodity compute resources, the associated loss in control over ‘nuts and bolts’ physical assets is much less of a problem than it might appear.
Traditional IT consumers might balk at the move, despite the real improvements it will bring to agility, because they like feeling in control of their technical environment. They might still feel the need for dedicated, physical servers that they can almost see and touch, for example. Or, they might still want an assigned applications team sitting in a nearby office that they can visit whenever a technical issue comes up.
But that ‘control’ can be both illusionary and restricting. When new business requirements come in and the existing infrastructure needs to evolve urgently, traditional IT consumers can find that:
- They can’t respond to new requirements anywhere near fast enough
- The cost of changing the environment they own is high, and even prohibitive
- The current state of the infrastructure is only partially documented (or worse)
Moving to a cloud infrastructure (and, more generally, cloud services) does require that consumers let go of some control over the ‘nuts and bolts’ side of their IT. Instead of dedicated physical servers, for example, they may have to choose from a pool of virtual servers within a choice of standardized builds.
But what they gain massively outweighs that inconvenience: control over factors that really matter for the business. They now control which services to consume and when, for example. They gain choice over the cost and quality of these services, too. And they are agile in a way they have never been before.
2. The Problem of ad hoc Service Delivery
The move towards standardized service levels means moving away from ad hoc service delivery – where IT staff create tailored solutions to any problem that rears its head.
Ad hoc service can feel comfortable and become an entrenched custom. For example, we may be all-too happy with being able to walk over to the server administrator or service desk analyst from time to time for a ‘small’ request or fix. It can be hard to give up that kind of service – but it’s worth noting that ad hoc service delivery has a considerable downside, too.
Think about service levels. In organizations where the services are not standardized, SLAs might be severely out-of-date, or irrelevant – meaning there are no SLAs to speak of. Consumers instead just go directly to the IT staff for fixes or changes, a habit that makes planning a nightmare.
In moving to a cloud model, the consumer needs to shift from ad-hoc delivery to services with well-defined, standardized SLAs. Indeed, one of the first challenges in the journey to standardization is to accurately answer the question: ‘What service levels are we getting today?’
The primary benefit, though, is obvious. Service levels, commitments, and expectations are clearly outlined and therefore much more likely to be delivered and to match actual needs.
There is also a secondary benefit to standardization. It is more efficient to maintain and support a standardized service (with standard configurations and procedures) than an ad hoc one (sometimes featuring idiosyncrasies accumulated over decades)! I would call this a domino effect – standardizing one service can boost effectiveness and standardization in other areas.
Of course, the IT consumer might find that with standardized service levels, they can’t just call the IT staff directly to resolve an issue like they used to. But they also now know that if there is an incident after the usual IT staff are gone home, they can rely on resolution processes backed by well-defined SLAs.
3. Getting the Service Quality you Paid for
When moving towards a more standardized, cloud based service, it’s not just the true Service Levels that emerge, but also the true costs.
Standardizing the service means you can precisely define service cost versus quality. For example, a popular way to describe the quality/price ratio is to present the services in ‘Bronze, Silver and Gold’ flavors, each with a pricing band.
For some customers this can be the ‘Aha!’ moment where they realize that they may have been paying for ‘gold’ in the past, whilst actually getting ad hoc services closer to ‘Silver.’ And they might just decide the ‘Bronze’ quality does the job fine. Now that is a culture change!
Lesson learned: once the quality/price ratio comes into the light, you may not always need, or even want, to pay for ‘gold.’ It might be just fine to settle for ‘bronze’ once in a while, and to use the resulting savings to add value to your business elsewhere.
Summary
In sum, here are three more arguments for moving away from tightly controlled, ad hoc service delivery and towards delivering standardized services via the cloud:
- There’s less value in controlling physical assets than your clients might think
- Standardized service delivery is easier to plan around and more likely to reflect actual needs
- With standardized service delivery you can make smart(er) budgeting trade offs
For more on this topic, join us at VMworld 2013 for the special session, Balancing Agility with Service Standardization: Easy to Say But Hard To Do (OPT5705).
The panel – featuring VMware’s Khalid Hakim, Paul Chapman, and Kurt Milne along with Dave Bartoletti of Forrester Research and Ian Clayton of Service Management 101 – will explore what works and what doesn’t when it comes to delivering standardized services that truly meet business needs.
Hope to see you there! For more CloudOps sessions at VMworld, check out our highlights of the Operations Transformations track.
