Service Reliability Index: A new measure for the digital banking services
Banking in today’s world offers a host of digital services like Internet banking, mobile applications, instant payments, and many more. These services rely on a range of software applications and extensive IT infrastructure to execute the intended task. The key metric that is measured and reported consistently across all organizations in this aspect is Service Availability.
Service Availability is measured as a percentage of uptime of the underlying systems. This metric has been immensely useful in diagnosing issues and offering a service continuously to the users. The major issue with this metric is from a system’s point of view and is completely devoid of the user’s experience with the service. For example, consider a service like a mobile app — a common issue is the time taken for a page to load is long enough such that the user abandons the journey without achieving their purpose. Now technically speaking, the mobile app is available, but given its “slowness”, it's not useful. To capture this gap, IT services need to adopt a probability measure, which is predictive and prescriptive. Reliability, a metric widely used in engineering domains can be adapted to achieve a more comprehensive metric.
Service Reliability can be defined as the consistency and dependability of a service in meeting its expectations.
Broadly speaking in IT services, reliability can be defined on four dimensions —
- Availability
- Usability
- Functionality
- Quality
To explain in the simplest terms possible — For a service to be reliable, it should be available when the user wants to avail it, it should be usable in a practical sense, it should have the functionality that the user desires, and the output given to the user should be of quality. When it comes to selecting metrics for these dimensions, the service owners have to carefully study the user journey to identify metrics that fulfill the following criteria:
a) Relevance in the customer journey (for eg: response time is a relevant metric when it comes to the usability of a mobile app)
b) Measurable repeatedly with minimal error
c) Actionable — methods are available to analyze and improve the metric
Service Reliability Index is the probability that a service will perform its intended function as expected.
Mathematically, it is the product of probabilities of each of the metrics selected in the above four dimensions. An additional criterion of normalization is added here so that the index can be expressed in simple terms, i.e., all metrics need to be expressed on a scale of 0–100%, where a higher number is better.
Service Reliability Index (SRI)= Availability x Usability x Functionality x Quality
The below formulae are some examples of how SRI can be calculated for a few generic services like Internet banking, payments, and customer services (CRM application to employees).
As you can see some fields are omitted on purpose- the SRI can be customized as per our requirement and dimensions which are not relevant or difficult to measure can be dropped. While these formulae are indicative, any organization can use this as a template to develop a comprehensive set of metrics and thereby a Service Reliability Index for their diagnosis and governance.
(Disclaimer: The author is a Business Consultant in the Banking, Financial Services, and Insurance domain at Tata Consultancy Services Ltd.)
References:
The Service Reliability Index described in this article is defined by the author, while the concept is broadly derived from Google’s Site Reliability Engineering and widely known DevOps principles.
