Fueling Transformation with Automation
By: John Locke
As a topic, digital transformation has been fodder for many articles, especially the aspects of how important it is and will be for organizations that seek to keep pace with the rising demand for greater service quality and faster delivery at manageable cost. Digital transformation projects cover a wide range of initiatives, from basic and small-step process improvements to wholesale technology changes. The common factor in all of these projects is the need to drive business value—whether by improving operational efficiency through automation or implementing new intelligent integrations between applications.
The recent focus on Machine Learning as part of AIOps is the latest incarnation of how data analysis is used to make sense of massive amounts of collected data. It’s really about answering an essential question: now that we can collect all of this data, what do we do with it? Understanding how to apply intelligence to automatically raise a ‘root cause’ alert to your operations team or create a trouble ticket for someone to investigate is obviously of massive benefit in improving operational efficiency. The additional business value of performing more advanced automation rooted in deeper data analysis, however, is still not being implemented widely by service providers, meaning that they are missing a big chunk of the machine learning and AI benefit.
There are certainly a number of reasons preventing adoption of more advanced automation: among them are concerns about losing control over processes and the risk associated with change. If the implementation is carefully planned and tested, however, risk can be minimized and significant benefits achieved. These benefits include:
- Decreased Human Time to Resolution (HTTR)
- Maximizing the use of skilled and valuable human resources to focus on other value-add tasks
- Implementation of agility and speed 24x7x365
- Improvement of operational efficiency to enhance quality and service levels
- Ability to run increasingly complex operations with the same or reduced resources
The ability to improve operational excellence across the areas that matter most—for example customer-focused Level 1, 2 and 3 support teams—with increased efficiencies and reduced triage and resolution times means your team can focus on other priorities that, in turn, allow the team to scale. This shift has obvious business value: faster incident resolution directly links to improved service levels and customer experience.
Of course, questions arise: What happens if an automated action does something that we don’t want it to do? If the automated process goes wrong, will I be able to correct it before it’s too late? To address them, let’s look at them logically. the process is only going to do what you program it to do, so the answer is to design and implement a process that you trust. Using business process modelling, human intervention and manual steps can be added into any automated process—but doing so reduces efficiency and therefore lessens the business value that is unlocked, as well as your goal of actually automating anything end-to-end.
To be able to implement automation and drive business value effectively, it is helpful to understand the steps involved in defining a solution. It’s important to start small. After all, you are trying to automate multiple individual tasks, which may all be part of the same business process, but we find that splitting each step of the process out, analyzing it, automating it and moving onto the next step with the same approach is by far the most efficient way. At the highest level, each process has three distinct steps: an input, a processing step and an output. The output from one process can then be used as the input into the next and so on.
