The Changing Landscape of Service Level Agreements

As we have engaged in more and more WAN RFPs for our enterprise customers, we’ve noted a purchasing trend worth sharing in some detail.  Moving from the RFP process to negotiating SLAs for our customers we’ve been struck by how focused many enterprises are on, and making purchasing decisions around, applications performance.

That’s an issue, however, because for decades the WAN providers’ standard service level agreements (SLAs) are still calculated and defined as packets and frames. The challenge, therefore, is to get everyone on the same page and start having the same conversation. This is how we’ve come to work on that problem:

By way of example, until a few years ago, discussions we were involved in revolved around the network, reach and performance of MPLS and WAN services in support of enterprise applications. Fast forward to today, enterprise networking conversations more frequently start with an application, such as a remote site that needs specific performance and functionality such as transaction latency.  From there the discussion works its way backwards into the supporting networks and other services.

Network providers should be key in all these conversations and we capture these details in our stakeholder Q&A calls and interviews. What we are trying to discover in nearly every case is whether or not the network and applications are already optimized and properly resourced.  Then we go on to assess the performance of the network to ensure the application is fully supported. We have to continue to find out and address which applications have latency. Once we have discovered the latency we can then construct SLAs for the enterprise that are based on guarantees needed to support the business needs rather than just being pulled from the service guide.

The important take away here is that carriers still present challenges. The first is that their boilerplate SLAs still present performance metrics in the service guides around packet loss, latency and jitter all based on various degrees of monthly averaging. With a well-suited carrier questionnaire, carriers will break their boilerplate WAN SLA latency averages further down to regions and between specific city pairs.  This level of granularity is much more beneficial to the specific customer needs.

The disconnect is still significant, however, because carriers are still talking about delivering packets and frames within averaged windows, and the enterprise is looking at applications performing details. So, the conversation is on the same topic, but not the same metric.  We all know it is not easy for carriers to match every application because each one can be different.  The unknowns are complicated to match to delivery guarantees, but that’s where working with a strong, experienced TEM vendor like Network Control can pay off enormously. Doing the deep investigative work, engaging with carrier reps, and customizing SLAs that match your business requirements, not carriers’ boilerplate calls for expertise.