Current Filter: Cloud>>>>>Opinion>

   

Service Level Agreements (SLAs) are part of a contract that defines a formal level of service between the provider and end user. In the context of the cloud service provider, most SLAs are centred on defining the level of availability and uptime that the customer can expect to receive, specifying the compensation in place if this service level is not met. Here, uptime and availability are seen as the most important principle in the contract and are often the core metrics against which services are evaluated.

So far, so good. But as with all best-laid plans, life intervenes. Service, more specifically, good service, is not black and white; its definition is changeable and subject to circumstance. Service requirements are unique to each business and will change over the course of a contract; as the business evolves, so too must the SLA. And although agreeing to an expected level of availability is obviously fundamental to any cloud service, SLAs need to look beyond this scope and include performance objectives that are meaningful to the customer with respect to their changing business needs.

Currently, the vast majority of SLAs are unable to get to the heart of what's important to customers or - at the very least - fall short of guaranteeing what's really needed and expected. As businesses have come to rely more heavily on third parties to deliver their IT, and as solutions have become more complex, 'good service' can't simply be reduced to five nines of service availability.

For example, that a service provider is meeting the levels of availability stipulated by their SLA will be of little solace to the CEO or FD that can't access their emails fast enough, or the online retailer is missing out on sales because of slow page loading times. These performance-based issues have proven to be something of a bugbear for the service provider industry, and have remained a grey area for some time now - they fall beyond the remit of the traditional SLA but remain key to the overall customer experience.

If the service provider industry wants to create more value for their end users, which is what the overall goal should be, they must look beyond simple measures of uptime and availability to measures that are meaningful to end users and contract against them. These measures, or Service Level Objectives (SLOs), address issues that cloud providers have historically resisted making meaningful contractual commitments on; service performance, data back-up arrangements, encryption specifications, and incident response/resolution and data transfer times. All are vitally important to the running of a business, but all-too-often neglected by the SLA.

The problem is that standard SLAs do not reflect the true dynamic nature of the relationship that now prevails between customer and cloud provider. The best providers understand this and are able to assess and contract their services specifically in the context of their customers.

It's important to remember that it's the end user's actual experience that counts and SLAs need to reflect this. If end users have a situation where their SLA is being met and exceeded by their provider, but are still not seeing the levels of service that their business requires, it is often a sign that the SLA is too generic and objectives have not been set that reflect the specific performance needs of the end user.

Ultimately, an SLA is a good baseline contractual agreement but, as the relationship evolves, so too must the level of 'measurable' engagement to suit the performance and optimisation criteria of each customer. CSPs need to look at those things that are actually meaningful to businesses when constructing their contracts - in essence, bringing customer service into the SLA.
More info: www.claranet.co.uk