Over the last few years, more and more organisations have started to work with managed cloud service providers to deliver a variety of business services. After talking with companies and reading the trade press, it has become clear that not all projects have gone smoothly.
All of the organisations we’ve spoken to had strong reasons to look at, and choose, cloud services. They wanted to increase business agility, reduce costs, enable a focus on the core business, improve business continuity, reduce complexity and bridge the skills gaps that have appeared as IT stacks have become broader, deeper and more complex, calling for increasingly skilled personnel.
Still, many companies have been disappointed and their expectations not met. Instead of becoming more agile, they have been in discussions with their provider about unexpected cost, have had to deal with adjustments to business processes and needed to manage dissatisfied staff that have had their hopes raised when the company finally moved to the cloud.
Two important reasons for these disasters are the lack of a detailed due diligence process before selecting a cloud provider and the lack of a proper cloud strategy.
When considering cloud, and looking to create the right cloud strategy to take the organisation forward, the business needs to focus on the following questions:
- What is the best service model for every specific workload or service? In other words, who manages what and up to what layer?
- What is the best deployment model for every workload or service? In other words, which location(s) do I deploy my services ranging from private to public deployment models?
- Which financial model provides the best fit for every workload or service? In other words, do I want to pay for the service in an OpEx or CapEx manner?
Start with answering these questions and you will have a foundation for your cloud strategy. When answering them, consider these lessons learned.
Keep these lessons in mind:
- Data is strategic – therefore, keep it close. Data has much weight and is not as fast moveable as CPU or memory. As a result, it is wise to keep it close and ensure continuous access regardless of type of platform, private or public.
- Keep your cloud options open. If you run an application in a public cloud now, who knows how long you want to stay there. That is an extra reason to keep your data close.
- An experience level agreement is more important than a service level agreement. Check how your provider handles changes and incidents. Is he focusing on solving the problem first or is he entering in formal discussions on procedures and rules every time you log a call? The end user experience and ability to do their job is key.
- Building a relationship with a cloud service provider takes time. Take that time to get to know each other. This initial investment will pay off as you move along.
- Consider the way you want to work when you want an exit. Many companies forget to think about an exit when they open the champagne bottle at the contract signing. It is better to be safe than sorry and also discuss a possible exit right from the start.
- Put automation and orchestration on the agenda. To profit fully from cloud computing, you will need automation and orchestration. This will help users find the apps and data they need and provide automated workflows that speed up business processes and boost efficiency.
- Make sure your cost models are predictable. Cloud computing is not always the economical alternative for an on premise approach. Therefore, you need to ensure, that you have complete insight in the TCO of both your on premise and cloud platforms.
By answering the three strategy questions and keeping our lessons learned in mind, it is possible to avoid a cloud disaster and find the best suitable solution for any workload you have.