How to Fix a Cloud Deployment That’s Losing Money

Cloud computing will save you money. 
Ten plus years ago that was a familiar refrain when the technology hype machine touted cloud computing as the best thing since sliced bread. Today we have over a decade of cloud deployment data that often disproves or redefines the context of that statement. 
NetEnrich recently conducted a survey of 100 IT decision-makers in companies with 500 or more employees. Amongst this group, 85% claimed either moderate or extensive use of cloud computing infrastructure, while 80% stated that they have moved at least a quarter of all workloads to public cloud providers.
Now the bad news. The survey identifies security as the top cloud computing issue (68%), followed by IT spend and cost overruns (59%), day-to-day maintenance (36%), and root-cause analysis and post-mortems (22%). Some 48% also claim that the cost of recruiting cloud skills is an ongoing issue in their IT organization, with an estimated 10 open positions that chase every one qualified candidate. 
All in all, cloud computing is not the cost-savings slam-dunk once promoted. This is especially true about delivered cloud value, in terms of business agility, compression of time-to-market, and instant scalability. In some situations, enterprises see negative values for every cloud computing dollar spent rather than the business efficiencies they expected. 
Our decade of cloud implementation data indicates that enterprises often need to take operational cost savings off the table as value justifications of cloud computing, which leaves soft value savings such as agility and speed-to-deployment. If those soft savings are not there, or not valuable to a specific enterprise, the enterprise will see negative value in cloud computing. This is a concerning situation for enterprises and cloud technology vendors alike. 
So, what can be done about this problem? 
Also see: CloudOps: How Enterprise Cloud Migration Can Succeed
The Choice That Isn’t a Choice: Cloud Isn’t Optional
If cloud computing delivers a negative value, you might think the best solution is to push off future cloud migrations and/or pull cloud applications and data back to traditional systems within enterprise data centers. Unfortunately, those ideas won’t work. 
The major enterprise technology providers dedicate 80% of their research and development (R&D) budget to build technology for cloud-based platforms or to build clouds themselves. That means R&D dollars are not spent on the upkeep of existing products or development of net-new products for more traditional enterprise systems. 
Enterprises already see fewer software updates and upgrades for non-cloud products and dwindling updates and upgrades for security and database systems. The larger enterprise technology providers know their future is in the cloud, and thus, they spend their innovation dollars in the cloud. That money will naturally come from the budgets of their more traditional business software and services. 
At the end of the day, those who want to avoid value delivery issues around cloud-based projects may discover even less value by staying put or repatriating systems back into the data center from the cloud. So, you can’t avoid the pull of cloud platforms to move forward, and you can’t save any money if you move backwards. At this point, it looks like cloud is the only game in town.
For Cloud, Planning is Everything
Most negative cloud values arise when an enterprise does not correctly leverage cloud-based resources. Typically, this means someone did not pick the best platforms, tooling, and processes for the project, and those choices are not or cannot be optimized. Those incorrect choices almost always track back to a lack of planning in the initial phases of a technology project, cloud or not. 
A universal truth: If you don’t have a thorough understanding of the problem, you will only have a partial understanding of the solution required to solve the problem.
The somewhat good news? There is no immediate pain to an unoptimized cloud operation. Most of the time, moving systems to the cloud works just fine and, initially, appear to be successful migrations.
However, even though it “works,” the team will soon realize the migrated or net-new cloud-based systems do not deliver the value defined and promised in the business case. This includes value metrics in terms of how the business value of cloud computing can be measured. Here, many cloud projects get an F. 
For example, look at a generic enterprise that moved their inventory management and inventory data to the cloud over the last five years. There are about five variations of cloud technologies they could have deployed, and an exponential number of configurations within those variations. Most of these solutions and configurations work, but these enterprises will not realize their peak hard and soft savings because the defined value metrics are misaligned with the chosen cloud solution. Therefore, the enterprise will experience some degree of negative cloud value. 
In the case of our inventory management system, let’s say the enterprise defined a cloud-based database that won’t store complex data and/or can’t deal with nesting as needed for advanced predictive analytics. It turns out these features are required to create automations to optimize inventory, such as just-in-time (JIT) inventory and automation across a supply chain. 
Let’s also say the enterprise leverages a cloud-based user interface development system that does not support mobile computing platforms. To top it all off, the applications and data storage systems were defined without regard for operations, security, and governance during initial planning. Thus, the system must deal with the inefficiencies created when those parameters are added into the system just prior to deployment.
Now is the Time Improve Cloud’s Business Value
The trouble with the slam-dunk cloud conclusion – this cloud deployment is clearly a success or a failure – is that it doesn’t account for real-life variables, such as skills shortages and an almost chronic lack of planning that’s necessary to make good technology decisions.
The goal is to build better systems with better thought-out cloud technology and configurations to find the desired value. That’s a hard goal to reach if you can’t find the skills required to correctly complete the first step or if you forge ahead without those skills. 
Some would argue that cloud computing is so new we’ve yet to build up sets of best practices to cover every cloud situation, and these types of issues are just a part of the growing pains. It’s now a proven fact that we can’t go backwards. That means we must find a way forward to establish business value for every cloud project. In short, we have to learn from our mistakes and reconfigure our cloud deployment. 
If you think about it, we can almost always trace a cloud project’s negative value issues back to substandard skills at the outset of a project. Whoever made those calls with faulty information often found that their choices led to negative values, even when the system “worked.” We clearly need to fix the planning problems ASAP and define clear paths for our IT staff to gain the knowledge required to close the skills gap. 
The good news: now that we’ve been down this path, we know enough to step back and rectify our earlier decisions, to use a now-informed cloud strategy to alter key variables to move toward a profitable cloud deployment.
Here’s the Takeaway: It’s Never Too Late to Reconfigure
An optimized and high-value cloud solution is almost always possible, but most enterprises will fall short. If your new cloud-based applications and data sets lack the value the enterprise assumed was a guaranteed outcome of the move to the cloud, it’s time to go back to the beginning.
Take a hard look at the initial planning processes for current and future cloud migrations and look for opportunities to optimize existing migrations. It’s never too late. 
The post How to Fix a Cloud Deployment That’s Losing Money appeared first on eWEEK.