A digitally transformed company can only succeed if it has reliable connectivity and a high-performing network. However, service providers can’t always keep their commitments made in their service-level agreements (SLAs) because downtime and outages are a fact of life.
An SLA is meant to support a service provider’s commitment to live up to customers’ expectations for network performance. The majority of providers work to ensure they do so, but sadly, they aren’t always successful. This can result in overprovisioning and increased expenditures.
You need to ensure your service provider is meeting their SLA – but how do you do that?
Why assurance matters – and is increasing in significance
As digital transformation has accelerated, the assurance problem has become more obvious. At the same time, many enterprises that used to have their own resources and data on-premises have moved workloads en masse to the cloud. They’re using SaaS applications from multiple vendors, which means basically giving away control to third parties.
Concurrently, because applications have evolved to such a significant extent – and are very sensitive to latency and availability of bandwidth – more issues may arise. In the past, you might have had one or two sensitive applications of this type. As long as you had those under control, you were fine. But now, regardless of what industry you are in, your enterprise has likely deployed applications that require high reliability, high availability, and end-to-end persistent performance. That has made the need to drive SLAs even more important. And the way it was traditionally solved would have been with overprovisioning – just add more bandwidth.
That “solution,” of course, increases cost. Enterprises can’t afford to keep adding bandwidth, which is really just a patch to the problem, anyway. You’re not getting to the root of the problem, and that becomes increasingly expensive.
Ensuring your provider delivers
One of the biggest challenges is that you have to take what your service provider is telling you at face value; you don’t really have a way to police or monitor what you’re receiving as a service, other than what they show you in your customer portal or in their reporting tools (if they have them.). And when you do experience issues, you usually find out because your customers are complaining, and that’s when you turn back to your service providers.
The solution? It’s all about finding ways to take back control of that end-user experience. The way you do this is to ensure you have visibility. Even though your infrastructure is running somewhere else, you as the enterprise can still have eyes on the traffic that’s impacting your customer. If you have eyes on the traffic and you have the means to provide performance assurance on that traffic and that infrastructure, then you regain control over the end-user experience.
Holding SLAs to the fire – Three things to do
When expectations are satisfied, a performance-based SLA helps maintain service providers’ accountability and fosters customer trust. It also spells out the consequences for the provider if the SLA assurance is not met. An SLA is simple to sign and then ignore, especially when inboxes are already overflowing with papers. However, IT cannot afford to disregard the SLA guarantee they have been given.
To help you keep your provider accountable, take these three actions.
- Check the numbers
As opposed to being constant or instant, network polling cycles typically occur at intervals of one to two minutes for ping checks and five minutes for data gathering. Accordingly, network degradations may escape the notice of monitoring tools. That’s why SLAs should specify the granularity needed to measure downtime, which should be in seconds rather than minutes. If not, a provider could claim that the service was operating satisfactorily despite disruptions of up to 35 seconds every minute. (These disruptions are time-consuming; research found that for 1 in 5 organizations that experienced network brownouts daily, IT teams spent up to 12.5 hours a week troubleshooting issues that could have been avoided.)
For the one in five organizations that experience network brownouts daily, IT teams spend up to a staggering 12.5 hours a week troubleshooting issues that could have been avoided.
In some cases, service providers advertise high SLA numbers while giving customers an SLA that includes a wide range of numbers that offer little to no practical benefit.
- Ask for real-time reports
The majority of service providers don’t want the responsibility of providing real-time measurements of uptime themselves, thus they demand that customers assess network performance on their own and produce data to demonstrate an SLA breach has happened. Demanding real-time reports for verification is a great way to make sure you’re receiving the service promised in the SLA.
- Retain professional counsel
Like most legal documents, SLAs can be difficult to decipher. Legalese is tedious and jam-packed with jargon. But because IT teams are increasingly depending on their service provider to maintain network performance at a high enough level that apps function properly for users, it’s critical to have visibility into your provider’s claims. To ensure that you understand the conditions and to get guidance on how to ensure that your SLAs are satisfied or that you can recover costs, talk with your corporate legal team or a lawyer.
Don’t set and forget your SLAs
Proposing service level agreements (SLAs) based on systems availability is easy enough to achieve – but guaranteeing SLAs at a service performance level to deliver seamless end user experience is a whole other matter. If you sign the contract and then ignore it, you’re liable to hear from your customers when your provider’s promises are falling short. Penalties in the SLA should be significant if the SLA assurance terms aren’t met, and you should ensure that it contains a clause allowing you to break your contract easily if the benchmarks are repeatedly missed. The best approach is to focus on preventing an SLA breach in the first place, of course. While you can’t always control what your service provider does, the above guidance and best practices will go a long way in helping you hold their feet to the fire.
