The hosting biz is easy. Right?
As anyone who has worked at a hosting company knows, its primary business and financial directives are to increase monthly recurring revenue (MRR), reduce churn, and maximize EBITDA. After all, most hosting companies are provided external capital by investors or shareholders who keep a close eye on how their money is being spent and expect a high return on that investment.
Analysis and management of your support organization become a critical part of this profitability equation. Addressing support tickets takes time, and the more tickets a customer submits, the more that customer erodes margin out of their contract. Clearly, keeping your support processes efficient can lower overall operating cost, increase margin, and at the same time improve customer satisfaction and decrease churn.
To do this, it’s imperative to work smarter, not harder. Instead of building large, over-staffed, multi-tiered support teams to deliver that exceptional service, you can invest time into understanding the most frequently occurring issues and top reasons for the support calls, and then create new processes, procedures, or implement new tools or technologies to specifically address them head on.
Troubleshooting performance is expensive.
In a managed hosting environment, trouble tickets can come from almost anywhere, including application support, updating or changing firewall rules, and adding IP addresses to new servers. 451 Research confirms that the #1 customer issues reported in a service provider’s support center are performance-related.
Source: The 451 Group. “Hosting and Cloud Study 2014. Hosting and Cloud Go Mainstream.” 2014
Unfortunately, some of the hardest tickets to resolve are those around application performance. Whether perceived or real, they take a lot of support hours to troubleshoot and resolve.
In the multi-tenant world of cloud or shared hosting where hundreds, maybe thousands, of users or applications are sharing the same underlying infrastructure, plenty of potential culprits can cause performance problems. From experience however, we know that in most cases the issues are caused by bottlenecks in the underlying storage subsystems. Many multi-tenant hosting environments use controller-based SANs to provide the primary block storage to the virtual servers, and when your performance is limited to the amount of processing power in the controllers — you see bottlenecks and the service provider sees trouble tickets.
What does an all-flash array have to do with trouble tickets?
All-flash arrays, specifically those with Quality of Service (QoS) controls like SolidFire, can dramatically reduce the number of performance-related trouble tickets you receive. With QoS controls, you can specify the IOPS for every provisioned volume in the system and adjust them on the fly.
For example, if you had QoS controls in your storage platform and you receive a support call about poor performance, your support agent could instantly increase the IOPS on the volume with the customer on the phone and ask the customer if the performance issue went away.
- If YES, then the customer needs more performance resources for their application. The support technician then charges the customer more money for more performance.
- If NO, you know that storage is not the issue (for a change) and that you’ve removed from the equation the most complicated piece of the puzzle. Troubleshooting can now continue with better focus.
Without QoS, when a customer submits a ticket claiming poor performance on a server, a support tech could spend hours trying to troubleshoot the issue. If that tech earns $15.00/hr. and takes three hours to find the problem, move the assigned LUN to a faster tier, and then close the ticket, your margin can erode very quickly by dozens of expensive $45+ support tickets.
At the end of the day, trouble tickets created by poor-performing storage systems are expensive and chip away the bottom line profits of every hosting business. With an abundance of IOPS combined with storage QoS, you can reduce the volume of trouble tickets received and more effectively deliver a better hosting experience with better margins.