Alemba - 7 Tips for Better Managing SLAs

7 Tips for Better Managing SLAs

Alemba • 30 May 2017

Service level management is an important piece of the overall IT service management (ITSM) jigsaw. However, it can often be a struggle for ITSM practitioners after the hard work is done in creating, and agreeing, the required service level agreements (SLAs). This blog offers up seven tips to help you ensure that your SLAs are effective as one of the necessary steps in improving your service level management.

1. Have an Agreed “Escape Plan”

It’s not a nice thought, granted, but make sure that you include a breakpoint in your SLAs. This is only ever invoked in extreme cases but it protects both you and the customer from unacceptable levels of performance.

This “get-out” clause makes it possible to terminate SLAs under certain conditions such as exceptionally poor performance, major problems, or significant changes to services. Some examples of unacceptable performance could include:

  • SLA target(s) missed three months in a row
  • Major increase in major incidents

Having defined break points means that an SLA can provide increased certainty and security for both parties by defining particular events that, without argument, allow for the agreement to be stopped.

One-way of doing this is to set minimum service levels at the level that allows termination. You can also define an amount of accumulated service level failures that allows an SLA to be disbanded for consistently poor performance.


2. Build in Some Flexibility

Escape plans are very much at the extreme end of the scale and sometimes a little compromise on both sides can do wonders for goodwill and the overall relationship. Building some flexibility into your SLAs means that both parties know where they stand and it avoids the need to constantly adapt SLAs every time something out of the ordinary happens.

When agreeing exceptions within the SLA, ensure that you understand them and their potential impact on the overall service levels. Some examples of SLA exceptions include:

  • Planned releases, upgrades, patching, and emergency maintenance
  • In the case of a cloud service, what the customer does with their own servers
  • Third party services and suppliers
  • Any customizations carried out by the customer
  • Busy periods and seasonal considerations
  • Relaxation of availability targets if the business has requested a release with known errors and defects.

3. Manage Your Measurements

Build performance management into your SLAs such that both sides can understand what should be reported on. To aid this, SLAs should be SMART:

  • Specific: The target is clear, unambiguous, and understood by all.
  • Measurable: It’s possible to easily quantify or report on available metrics for the service being measured.
  • Achievable: It’s possible for the goal to be reliably and regularly achieved; if you are unsure of what this looks like, work with your support teams to agree a reasonable target.
  • Relevant: Achieving the target is relevant to the mission or objectives of the organization, linking in to the overall business goals.
  • Time based: The target includes the timescale over which it should be tracked, measured, and reported

Ensure that you agree how your SLAs will be measured and reported on, as well as agreeing the timings of service review meetings.

Tailor your metrics to your specific SLA requirements and needs rather than using default data or generic SLA reports. We’ve all seen organizations where pages upon pages of metrics and reports are generated every month only to sit unread in an already overcrowded inbox. Instead, base your metrics on what really matters, e.g. availability, performance, and overall customer satisfaction.

If you have an ITSM tool that allows it, set up automatic monitoring and set thresholds for well before the SLA is breached. If your service is approaching amber status, according to the SLA, don’t wait for it to go red and breach – instead get proactive and agree an action plan to restore service to acceptable levels.

Some examples of things to do when your SLA is threatened could include raising a problem record and pulling together a technical observation post to look at ways to improve the overall service. Or temporally giving incidents related to the amber service a higher priority when being logged by the IT service desk.


4. Train Your People

Your teams must be thoroughly trained on what SLAs are in place as well as any exceptions, or service-specific protocols, to ensure that: everything works the way it should, there are no misunderstandings, and nothing is lost, ignored, or forgotten about. They also need to know exactly which services are in scope as opposed to those that are looked after by other teams or third-party suppliers.

Everyone should be familiar with the different priority levels as well as what to do in the event of an emergency, for example, a major incident or an event that leads to the IT service continuity or disaster recovery (DR) plan being invoked.


5. Don't Forget Your Suppliers

SLAs are only part of the story. For your SLAs to be truly effective, they will need to be underpinned by the appropriate operational level agreements (OLAs) and underpinning contracts (UCs). Where:

  • OLAs are agreements that define the interdependent (internal) relationships in support of a SLA. The agreement describes the responsibilities of each internal support team towards other support teams, including the process and timeframe for delivery of their services.
  • UCs are agreements between your organization and third-party suppliers such as software companies or network providers.

A common service level management mistake is to have a SLA that’s unachievable because the fix-time agreed in the OLA or UC is longer than what has been agreed for the SLA.

So when drafting your SLA, talk to your service delivery managers and supplier management teams to ensure that your SLA has enough time to deliver the agreed services taking into account the related OLAs and UCs. Also bear in mind that you’ll need to allow for extra time for hand-offs and for the service desk to manage communications.


6. Improve, Improve, Improve

Building targets into your SLAs is a great way to encourage continual service improvement (CSI) in small, manageable chunks. Targets are also a great way to ensure that any existing improvement plans don’t lose momentum.

So, use SLAs to drive the right behaviors, such that you can incentivize everyone to improve the overall quality of service over time. There are lots of options for rewarding your teams, including positive feedback, annual bonuses for hitting or exceeding targets, or employee of the month award.


7. View SLAs as part of a Bigger ITSM Picture

Service level management shouldn’t be treated as an ITSM island. For instance, Business relationship management (BRM) can help to manage and improve the existing relationship between the service provider and the customer.

Whereas availability and capacity management can provide information about uptime and performance. This can then enable a service level manager to carry out a gap analysis and progress improvements through a service improvement plan.


Having fit-for-purpose SLAs in place not only protects your business from unacceptable levels of performance but also empowers support teams to deliver service in an actionable and measurable way. Great SLAs can provide a strong foundation on which to build a strong relationship with the client and to better manage their expectations in a transparent and fair way.