The CIOs and technology managers surveyed said that cloud infrastructure was their top IT investment for 2013 and for the long-term. The survey also shows that concerns about the cloud have matured as well.
In contrast to early cloud objections around security, current-day concerns of IT managers involve line of business issues. The three top cloud concerns of CIOs, according to the survey:
- poor user experience due to performance bottlenecks,
- the resulting impact on brand reputation and loyalty, and
- revenue loss due to non-availability, poor performance or cloud service troubleshooting costs.
This became very clear last week when cloud outages plagued Amazon, Apple and Microsoft.
Managing the cloud is no longer just about servers and racks, as the majority of companies polled already use the cloud for customer facing, revenue producing applications. There’s also a growing focus on innovation, especially with larger companies which want to emulate the success of upstart competitors. It allows them to speed up development with infrastructure and services on demand, which shortens new product cycles and reduces the barriers to innovation.
With so much more at stake, the reduced visibility into cloud providers is becoming a greater concern for IT managers accustomed to a very tight rein on critical systems. Most cloud providers do not offer guarantees of response time levels and the typical red-yellow-green dashboard metrics of server health are not very meaningful in a world where customer experience is the real measure of success.
So as the cloud reaches its maturation phase, it’s clear the core survey takeaway is the business concern around loss of control.
Add in the growth in BYOD and mobile and you can see just how far removed cloud-based businesses are from managing their customers’ experience. Teams are left guessing when issues arise, wondering whether to point the finger at third-party Cloud services such as social media, ad servers, video, analytics and more. And, when you rely on a hunch instead of proof, dollars are wasted and the threat of a looming problem is only delayed, not prevented. You can see why the cloud keeps some CIOs awake at night.
So how can a business in this new cloud era take back control? We’ve helped companies through this process and a few key best practices have emerged:
1. Demand performance visibility
If uptime is your primary cloud performance metric — as it is for one-third of our surveyed companies — know that you’re running almost blind. Simplistic dashboards or server response time averages are not good enough. Today’s complex cloud environments require detailed monitoring metrics on individual transactions.
Your provider may claim a 99.9% uptime guarantee, but that means you can expect almost nine hours of downtime yearly. Are you prepared for an outage during your holiday rush campaign? Any penalty reimbursement is only a fraction of the total loss generated by an outage. So demand deep visibility into the true performance of your Cloud application and be ready to address issues on the fly.
2. Measure from the end user and analyze the business impact
Another big gap is lack of end-user perspective. Dashboards, like statistics, are only as good as the data behind them. Even the best server-monitoring tools leave you without a sense of what your customers are actually experiencing. So it is critical to monitor “from the edge of the Internet” and in context of the customer’s device and location, as well as performing ongoing analysis to show the business impact of varying performance levels.
When you consider that e-commerce platforms are among the most-used cloud services, you can understand how slowdowns or outages directly affect revenue. However, it’s not just the major problems that will cost you. Recent studies show customers are increasingly impatient and will abandon slow-loading assets, some within two or three seconds.
That’s why it’s important to correlate business impact metrics, such as conversion or abandonment, with end user performance. Actionable patterns will emerge from this data and have a direct beneficial impact on revenue. So engage your own end user performance monitoring and business impact analysis or work with an external provider.
3. Troubleshoot cloud problems early
CIOs also realize that in today’s complex cloud environments, troubleshooting problems is a time-consuming and expensive process when the infrastructure is not in your own data center. This is confirmed by our survey finding: 79 percent are concerned about the hidden cost of cloud services.
This requires support or development teams to gather detailed information quickly to resolve a problem the first time it occurs or before it goes live. If you can remove the guesswork and eliminate the need to recreate problem areas, teams can move fast and know that any bugs will be addressed quickly.
So the cloud is being asked to do more than ever before, and it is evolving to keep pace. We’re seeing this in the explosion of services offered and how some providers are now tailoring cloud servers for the needs of vertical industries.
As more companies learn to take back the control over their cloud-based infrastructure, it will allow them to make full use of its efficiency and scalability as they build the next generation of profitable business applications.
VB's research team is studying web-personalization... Chime in here, and we’ll share the results.