Adolfo Kunz
Adolfo Kunz

In the previous article, we presented a high-level overview of the importance of using an IT service level agreement (SLA) to measure performance, allocate shared risk, and serve as a continuous improvement tool. We also identified key characteristics of service levels, and offered guidance about how to differentiate critical service levels (which have a high impact or deliver incremental value to the business) from key performance indicators (which focus more on the delivery of data to support continuous improvement efforts).

In this article, we will explain how to build value-driven service levels by establishing service level targets and aligning them with business needs, target audiences, the scope of services, and installed capabilities. We will also explore how to use data (historic performance, market standards, etc.) to set specific target values, and how to prevent value erosion in your IT service level agreement.

Defining Value-Driven Service Levels

Value can be defined as the worth in economic terms or other tangible benefits that an organization receives from an investment or execution of an activity. It’s a given that investments and/or activities with no value attached to them are worthless and should be avoided. The same is true for instances where cost of the activity exceeds the benefit—and service levels are no exception.

Service level management has a cost attached to it; in a theoretical world, it can be seen almost as a non-value-added activity because value is only derived if the organization has the ability to act on the information it provides. For that reason, it is important to implement service levels and service level management processes that enable the organization to achieve its objectives and avoid any unnecessary burden. Organizations need to focus on establishing service levels that matter, along with tools and processes to manage them effectively.

There are several concepts that organizations need to understand when defining the tenets of their IT outsourcing service level agreement. Think about them in terms of the business value they deliver:

1. Service levels must be linked to outcomes that are aligned with the scope of services being measured.
Business users experience the result of an ecosystem of underlying technology, which ideally should be a “black box” to them. Case in point: measuring and reporting end-to-end availability and response times of a critical application makes sense to end users because that is what they experience from IT. Reporting only service level measurements on the availability of an application’s various technology components (i.e., its server, network, database, etc.) would be a waste of time, effort, and money because that does not translate to the real world outcome.

2. Service levels must be defined as the “minimum acceptable performance” and not as aspirational metrics.
Effective service levels do not deliver data—they drive behavior. Unrealistic expectations can lead to conflicts between the customer and the delivery groups. We all wish our services were up and running 100% of the time, but bad things happen. We need to define how much we can accept and how and when we need to act, and consider that acting will also be linked to costs for diagnostics and resolution.

Think about your home cable TV service: It may go down once or twice a year, however, would anyone really expect (or want to pay for) their cable provider to install and operate a city-wide redundant cable infrastructure, just as fail-over for very intermittent outages?

3. Service levels can be expensive to set up and maintain, so do not measure something simply because you can.
In other words, the value achieved through the behavior or action triggered by a service level in your IT outsourcing service level agreement should always be greater than the cost of measuring and reporting (costs can be measured in “hard dollars” or “opportunity costs”). For example, measuring the availability of non-production environments may be important during specific time periods and for specific instances (large developments, critical solutions, etc.), but not for all applications.

Types of Service Levels in an SLA Agreement

To define the right levels in their IT outsourcing service level agreement, organizations need to understand their purpose and audience.

A. To manage performance

These service levels are aligned with end users. The end users can be internal or external stakeholders that are the recipient of a given product or service. The focus is placed on the outcomes from an operational perspective, by making sure end users receive the services they need to perform their activities. Examples include: application availability, timely delivery, incident response, end user satisfaction, etc.

B. To manage risk

Service levels are aligned with areas of exposure or risk for the organization. The user is usually an internal stakeholder responsible for risk, compliance, and security. The focus is placed on mitigating risk exposure and its associated consequences by preventing financial and reputational damages that can impact the organization. Examples include: security incidents, compliance audit results, business continuity, disaster recovery, etc.

C. To manage transformation (next-gen)

Service levels in your IT outsourcing service level agreement need to be aligned with the business impact of transformational goals. The users are usually internal stakeholders responsible for developing and implementing the initiatives that support the organization’s strategic objectives. The focus is placed on the realization of the benefits projected through strategic partnerships or the execution of specific strategic initiatives. It is important to distinguish a strategic partnership (where there is risk or incentive being shared) from a pure vendor-client relationship (where the vendor delivers services defined within a scope of the contract for an agreed-upon fee). Examples include: increased productivity, change management success, realized value, market share gains, innovation index, etc.

Defining Service Level Targets

Your service level targets in your SLA agreement should be (i) driven by business needs and (ii) linked to the capacity to deliver. As mentioned previously, service levels have a cost and the higher the expectation, the higher the cost to meet the service level.

business-needs-delivery-capabilities


Business expectations are always high and people dislike anything that disrupts their work. But meeting these expectations may have a cost that is higher than the value received, so the right balance needs to be defined. It is important that this is understood by all stakeholders involved in the SLA agreement from the outset. 

When establishing service level targets with suppliers, organizations can follow different approaches: (i) market standard, (ii) baselining, and (iii) zero-based approach.

i. Market standards: This method consists of using available market data to define service level targets in an IT outsourcing service level agreement. This is the recommended method for commodity services (such as Wintel server management) and is particularly useful when organizations are outsourcing to a third-party vendor who delivers in a standard, repeatable manner. This method can also be used to create meaningful metrics for internal service delivery teams. Research and business advisory firms like Wavestone provide benchmarks or standards for service levels that define “minimum performance levels” based on business needs and the delivery characteristics or components of the service.

ii. Baselining: This method consists of measuring a service through a defined timeframe to establish the service level target, adjusted by a standard deviation factor. This method is usually recommended when outsourcing third-party vendors. This method is primarily used when a vendor takes over a client’s existing services and technology environment (so the vendor’s standard means of delivering the service does not apply) or for a customized service where no market data is available. There are two main risks with baselining: 1) It may not remove inefficiencies from the existing performance, creating low performance targets that deliver no value, and 2) overperformance during the baselining period may create artificially high targets.

When engaging third-party service providers in their IT outsourcing service level agreement, organizations need to keep in mind that the ultimate objective is to improve the quality of services and align to the needs of the business. Baselining, particularly during the knowledge transfer/transition phase or based on historical performance, will deliver service level targets aligned to the current environment. That said, it will not result in enhanced performance unless and until the vendor can implement service improvements.

iii. ZeroBased: This method requires assessing each component of a service individually and aggregating those metrics into a service level target that has not previously been defined. This method is not usually recommended, as it takes more time and effort. This method should be used for new services where no market data exists. The main risk associated with this method is that it may establish a low service level target because of the aggregated values of its individual components.

The following is a summary of pros and cons for each of the above methods:

pros-cons-approach

Preventing the Erosion of Service Levels

The value delivered by a service level in an IT outsourcing service level agreement can be severely eroded by the way organizations define them, particularly in situations where third-party vendors are involved. Third-party vendors naturally have the tendency to protect their interests—seeking to negotiating an SLA agreement that minimizes their risk—but the consequences can be diminished protection and little, if any, value to the recipient of the services.

There are different ways to erode the value of service levels by wasting time, cost, and effort for the service delivery group and the end users. They include:

Below is an example of an SLA agreement erosion through bundling (use of averages), which is a very common method third-party vendors use to mitigate risk:

Looking Ahead

In this article, we discussed how to build value-driven service levels in an IT outsourcing service level agreement by establishing realistic service level targets that are aligned with business needs, target audiences, the scope of services, and installed capabilities, all while leveraging industry market standards. We also provided information about how to prevent value erosion over time in SLA agreements with third-party suppliers. 

In the next article, we will discuss some of the key concepts in service level management, as well as the elements that should be included in your service level management methodology when establishing IT outsourcing service level agreements with third-party vendors. 

Adolfo Kunz
Senior Specialist

Adolfo Kunz has more than 33 years of experience in IT sourcing, strategy, operations, transformation, cost optimization, service management, vendor management, IT due diligence, and planning for mergers and acquisitions, business development, program management, and project management. He has worked with Fortune 500s in a variety of industries, including energy and resources, oil and gas, consumer packaged goods, retail, transportation, financial services, healthcare, and manufacturing.

Key Service Level Management Concepts Explained

Oct 15, 2021

Leverage these key SLM concepts to incentivize positive third-party behavior.

Three Areas Where Most SaaS Agreements Fall Short (and What to Do About It)

Oct 05, 2021

Best practices to maximize value and gain better savings

Have a Question? Just Ask


Whether you're looking for practical advice or just plain curious, our experienced principals are here to help. Check back weekly as we publish the most interesting questions and answers right here.

Ask Wavestone