Read this blog to learn what a service-based SLA is, the metrics associated with SLAs, and how to react if a breach occurs. Also, find tips on how to prevent breaches.
The most important determinant of customer satisfaction and quality is whether all promises made by the service provider have been fulfilled. The Service Level Agreement or service-based SLA lists out the promises and how to measure if they have been met. If they have not been met, then it is a breach of the service level agreement.
Metrics for Service Level Agreement
Generally, service providers will provide statistics for the service levels. These are made available mostly on an online portal for the customers to verify whether or not the terms of the SLAs are being satisfied. They can even check if they have entitlement to service credits and penalties based on the terms of the SLA.
Generally, the outsourcing company identifies these methodologies and processes. They are identified keeping in mind that they are capable of supporting the SLA. Nevertheless, it is better for the outsourcing company and the client together negotiate the SLA contract so that there remains no misunderstanding regarding the methodologies and processes of support, management, and the methods of reporting.
In the case of critical services, it is a good idea for the customer to use third-party tools that will automatically capture data with respect to SLA performance, thus getting a measure of unbiased and objective performance.
Which Metrics to Monitor?
Which type of metrics will be required for the service level SLA will be decided based on the services that are being provided. Several aspects lend themselves to monitoring as an aspect of the SLA, and this monitoring scheme should be absolutely simple to prevent high costs and confusion for both parties. While selecting metrics, see what is most important in operation. If the monitoring scheme and remedy are made complex, the effectiveness will decline due to a lack of time for analysing generated data. Better to go for ease of metric data collection. The best option is an automated system, as metrics collected manually are both costly and not always reliable.
Based on the service being provided, the metric type to be monitored might comprise:
Availability of service: time duration for which the service is available to the client
Defect rate: The percentage or count of problems in key deliverables, such as errors in coding, delay in backups and restores, and deadline slippage
Technical quality: commercial tools are used to measure the technical quality of such factors as defects in coding and the size of the program
Security: measuring such security procedures as anti-virus patching and updates
Business results: making business process metrics part of the SLAs where the best approach would be to employ the existing KPIs considering that the contribution of the vendor to them is measurable.
Points to Consider When Selecting Metrics
Work with the objective of incorporating those requirements and best practices that help to maintain service performance while preventing further expenditure.
- Chosen measurements must motivate correct behavior
- Metrics must reflect factors that lie within the control of the service provider
- Chosen measurements must lend themselves to easy collection
- Chosen metrics must not be too few (some measures will be missing) or too many (excessive data will be generated)
- Readjust settings when required using a process defined in the SLA for it.
Furthermore, ensure that the SLA lays down how monitoring of the services will be done, including the manner of data capturing, reporting, frequency of review, and the stakeholders involved with the review.
How an SLA Breach Happens
An SLA is a legal contract among two parties (the service provider and the client) – the services to be provided and the performance level of the services are listed along with how they will be monitored and measured. Here the organisation acts as a supplier, and the SLA will usually be part of the contract for the provision of services.
SLAs will mostly be ideal in nature. Also, the provider of the service will commit that they will put in their best to meet all the commitments based on factors that are in their control. So, when the service provider does not meet the promised targets, the agreement is violated, and an SLA breach is said to have occurred.
Let us take an example. The SLA of an internet service provider promises 99% availability of the service in a month. The provider has a leeway of down tome upto a total of 7.31 hours in a month. If it is more than 7.31 hours, it will be considered a breach of the SLA, so long as it was within the service provider’s control and not some force majeure (a natural disaster/ faults caused by a third party).
Check the details of the contract along with its fine print. The fine print could be, for example, that the providers will accept it as an SLA breach only if it has been reported to them along with logging of the complaint on their service management system to enable the measuring of the metrics on a common platform.
The onus of proving that the breach has happened is generally on the customer. The customer could have their own monitoring system to have data as proof.
Reacting to an SLA Breach
Here is what needs to be done if an SLA breach is found.
The customer must use the provider’s channels to log the issue, which could be the service management system, social media, e-mail, etc. This needs to be lodged will depend on the terms listed in the SLA.
- If it is immediate, then the time starts then itself for the service provider to provide a response and tackle the reported problem within the time specified in the SLA. Else it will be considered a breach of the contract
- If the reporting was done after the event occurred, the issue’s record is valid, and an SLA breach has occurred. Yet, it must be kept in mind that the lodging of the issue must be done within the time period agreed to in the SLA.
After the breach of the SLA, it is for the provider to have clear and open communication with the customer regarding the status of the issue and how to move ahead.
Often, service providers use the ‘stop the clock’ method when they respond to a breach in an SLA. This functionality found in tools used for service management lets the provider suspend the agreement, generally in cases where the metric is timeline-based. As an example, they could suspend the SLA in the event that:
- They are waiting for customer feedback
- The issue is escalated to some third-party supplier
- The breach was caused due to circumstances beyond the control of the service provider.
While the provider will find the ‘stop the clock’ method to be correct, the customer will see it as cheating since it is not reflecting the breach’s reality in the SLA’s measurement. It has given rise to ‘watermelon SLAs’ in which the provider reports that they have met all the SLAs (green outside), and the customer has had a bad experience (red inside).
It is not right to leave the dandling of SLA breaches to an automated system or service desk agent. There must be a system to escalate breaches to the supervisors for review and redressal.
The provider should share a plan of action for preventing and solving issues. In the absence of follow-up and clear communication, customer dissatisfaction and even lawsuits will occur based on what the customer sees as the breach’s impact on it.
Tips to Prevent SLA Breach Circumstances
It might not be impossible to limit SLA breaches from occurring or even preventing them together. Here are some actions that can be taken to smoothen the path.
- Factor in the SLA while designing the service and not at a later stage
- Build-in continual improvement and strong problem management. Checking for root cause and how to tackle the root cause and add improvements will decrease recurrent breaches, even if they involve:
- Redesign/ upgrade of technology
- Reengineering of processes
- Training of employees
- Another action that needs to be taken is a regular review of the SLA to keep it relevant and realistic at all times.
- Build a framework that will monitor and alert. It could be an automated workflow that notifies or routes breaches that might occur to those skilled at tackling them.
Conclusion
An SLA breach points to poor service performance or delayed product delivery. With knowledge of the same in your arsenal, you will be better equipped to deal with such instances of a breach.
Read more,
