5 Top Reasons Why Cloud Computing Projects Fail

Leave a comment

  • Sharebar

Benefits of cloud computing are obvious to many decision makers now. It is a rage with large enterprises, SMBs and startups equally, mainly because of the cloud’s ability to provide the size-specific IT as a utility service.

Cloud adoption takes place in various forms in enterprises and can be for any level in the cloud stack. Projects such as integration and migration of data & systems are most commonly happening in the enterprises in order to switch to cloud. Or, even the core application development projects on cloud platforms (PaaS) are increasing now.

There is a frenzy to switch to cloud computing. The global cloud computing market is expected to grow at a 30% CAGR reaching $270 billion in 2020, concludes the latest research report Global Cloud Computing Market Forecast 2015-2020 covering the cloud computing products, technologies and services for the global market. What was once seen as knowledge gathering process has now turned into real-time implementation projects.

In the process, some common mistakes are made which leads to failure of cloud computing projects. Here are top 5 reasons for the cloud computing project failures:

1. Failure to manage and monitor applications

There is a myth that by switching over to cloud computing, the operations part of the applications are done away with. If you switch to PaaS, then the developers start coding almost immediately without having to bother about setting up the development infrastructure. They also do not have to bother about the ops part. True that cloud does change the ops scenario for developers, but it only changes the nature of the ops part. With cloud computing, you may not be bothered about provisioning, testing and/or deployment, but still you need to monitor and ensure that everything is alright with your application. After all, you will be accountable to your customers if your application fails.

Since in traditional on-premise development scenario, 80% of the developer’s time is spent on monitoring and management of the application and other IT infrastructure, it comes as a big jolt when the team and the management are not prepared to accommodate managing and monitoring of the application even when they switch to cloud computing. Unforeseen costs of operations and lack of proper application performance management process can lead to cloud project failure.

2. Failure to understand technology and legacy data

Migration to cloud computing is bound to have many integration requirements. Failure to understand the legacy data systems properly will lead to total meltdown of the cloud projects. Similarly, when developing a new application in cloud, the management or the development team may select the wrong technology for the development and hence end up selecting the wrong cloud vendor too. Naturally, if the wrong technology is selected for the application, the end result will be less than satisfactory.

3. Failure to understand requirements upfront

This is directly related to the reason #2 and is a common reason for project failures even in the traditional development situation. If you do not understand the project requirements upfront, then you will end up in either choosing wrong technology or develop something totally tangential to what your customer needs. Failure to understand requirements upfront will cause problems right from architecture design to database schemas, making test-cases and provisioning.

4. Lack of right skills-set

Cloud computing has created its own need for skill-sets, which should be specific to integration projects and development projects. Since cloud computing is relatively new naturally, the available skill-sets will not be so easily available. Moreover the available talent may lack the experience. All these put together increase the risk of project failures.

5. Failure to scrutinize the vendors

Cloud computing is relatively new and suffers from certain inherent glitches such as lack of standardization of SLAs, failover remedies etc. Failure to scrutinize your service provider can lead you to dire consequences. For example, you must understand where your data resides, what are the data security policy that your vendor follows, what are the back-up provisions, what are the disaster recovery processes etc. with your vendor. If you ignore such important issues in your SLA, then you may have to cut a sorry figure with your customer especially during a disaster situation. If the back-up policies are not robust, your project is at high risk for failure.

 

Image courtesy: http://www.improvizations.com/kronosblog/bid/28249/APPROACH-is-important-in-Workforce-Management-Implementations

Leave a Reply