Cloud computing gives enterprise users more flexibility than ever before while putting more pressure on the enterprise IT. End users want better applications, high performance and absolutely intolerable to downtime, development teams need rapid access to test and development platforms, whereas executives want justification of the all cloud related expenses and want to see the impact of the spending. Existing platforms cannot keep up with all these requirements and the corporations seek the solution in the cloud.
Despite the need, truth be told, many cloud projects fail. You may be tempted to think that a good plan, clearly defined use cases, carefully chosen, right project members and SMART success criteria is a clear-cut recipe for success. I can say that lacking these elements is a sure way to failure but I cannot say that having all these is a guaranteed success.
Reason 1: The cloud is technically perfect… except nobody is using it. I do agree that from a technical point of view, cloud is an excellent technology. From an infrastructure perspective, it is possibly one of the wonders of the enterprise IT; it is a marvel. However it doesn’t mean anything unless it delivers business value. I agree that such a wonder can only be built by highly skilled, highly technical people. But if IT is done for the sake of IT, meaning that if the cloud is built just for the technical superiority without considering the business value it will end up sucking up electricity and maintenance time where nobody is using it. To avoid such an outcome, it is important to include higher management in the project team right from the very beginning. Make sure that they clearly, openly define the scope and state their success criteria.
Similarly, the cloud needs to address and answer some basic questions: what problem will it solve? For what will it be used for? How will it be used? Who are the users and what are their roles? As I have discussed above, if the business use scenarios do not solve problems or create value there is no chance that the cloud adoption will take off. In such a case, where the business value is questionable, cloud deployments are likely to suck up resources without any foreseeable returns.
Reason 2: The cloud initiative will enable our enterprise to be more agile in today’s fast-paced business environment… And similar buzzwords which are too vague. Once again, the cloud should either solve a business problem or add a business value and be measurable in all its objectives. Defining unmeasurable, vague objectives or deploying a cloud just because it is a hype –everybody’s doing it, so it should be something good- is meaningless and the road to failure. Without specifying measurable objectives and success criteria, there is no way to determine if the objectives are met or not. One can say/observe that the cloud deployment indeed delivered speed, agility and flexibility but even if this is the case, it is impossible to measure the level and in turn, make sure that the chosen deployment method is the right method for the business. Asking simple questions as if and how each object is measurable, who are the parties interested in each objective and how meeting the objective contributes to project success will help the parties understand “success” better.
Reason 3: We can’t leave this feature to the second phase!.. And that’s why the deployment is behind schedule. Companies want to have all the features at once and want to start using the “investment” as soon as possible. This is human nature and understandable, however it is one of the reasons why many projects, not just the cloud projects, fail. No matter the size of the deployment, it should be broken down to bite size pieces and realistic milestones are established. Ensuring the project is defined with those success criteria means that the progress on the project can be tracked, stakeholders are promptly informed, the results are delivered as previously agreed, confidence is not eroded and the project is not left to die quietly on its own.
Reason 4: Of course our cloud project is successful!.. but we cannot simply show it. In line with reason 3, sometimes project success criteria are laid out very clearly but without any methods to measure it. Or sometimes the criteria are so impressive yet almost impossible to measure. Business use scenarios could be very specific, such as reducing the time it takes for a customer to place an order or reducing the time it takes to deliver a server to the development team. Whatever the scenario is, there should be tools and processes in place to automatically measure (and preferably report) the defined criteria. The results should be compared to the before-cloud time to see where there is a progress and where there is not. So that it could be decided whether to improve on the points where the progress is slower, drop the scenario entirely or continue as it is.
Reason 5: We use the cloud a lot… but we don’t know what it means in monetary terms. Generally IT projects invite finance people at the very beginning (to ask for funds) and at the end (to show them where the funds are spent). In terms of cloud projects, this is almost always the same. Despite the fact that one of the main reasons for cloud adoption is to lower IT costs, many projects fail to address the financial success criteria and focus on the technical criteria. In this case, companies reach their technical goals somehow but the costs become significantly higher compared to the existing infrastructure, to other vendors or to competing cloud models. In order to avoid such results, it is best to include a member from the finance team at the beginning of the project and have the financial success criteria clearly defined. This will enable to build a “before and after” model and enable to report the outcomes to the higher management by the numbers.
Here are my top reasons for the cloud projects to fail and my tips on avoiding them. What are your success criteria? Do you have any failures in your cloud deployment? Share with us in the comments!
- Featured image: https://dribbble.com/shots/123902-Fail-Cloud