Automation: The game changer & the involved challenges
Implementation of automation in application support and maintenance helps tackle issues like manual error-prone processes, high cost, and limited resources, and enables enhanced efficiency and productivity. However, automation in application support comes with its own set of challenges:
- Inability to identify use cases – Identification of use cases for automation can be a time-consuming exercise that needs involvement from multiple teams.
- Dependency on individuals due to lack of documentation – In the absence of proper documentation, knowledge typically resides with couple of individuals. Overlooking some critical aspects will lead to the entire use case to fail in the production environment.
- Lack of involvement – People consider automation as a yet-another organization initiative. Identification of use cases, doing POC, testing, implementing it, change management, etc. can be time-consuming activities. Motivating everyone to contribute towards automation can be a challenge but people involvement is critical to the success of automation.
- Lack of unified processes across the organization – Basic processes have to be implemented before embarking on an automation journey.
- Business Complexity – Every business process may have multiple variants. Automating each variant might be difficult and may not give the desired ROI.
- Wrong approach taken for automation – The most common approach taken for automation is to look for plug and play options or look through the partner ecosystem for automation solution. If something is not available or doesn’t completely solve the problem, it’s written off.
- Absence of central governance – Automation is not the job of just the support team. There has to be a central team governing and encouraging automation across multiple teams. Multiple teams might have to come together to make it possible.
- Ever-changing IT landscape – Multiple releases happening every now and then, new applications getting added, addition of SaaS applications, etc. can make automation challenging.
- High degree of customization – Customizations, especially in ERP environments, always increase the complexity when adding anything new in the application.
- Considering automation as one-time activity – Once automation is done, people become complacent. It’s important to remember that automation is a journey and not a one-time activity. Continuously training and retraining the models, looking at value delivered and identifying new automation opportunities is the key.
The apt approach to automation
So, how do we approach automation? First and foremost, we need to understand that automation is a journey which has to start from within. The starting point lies in existing tools and infrastructure. It is essential to start with simple but high-impact use cases. The focus should not be on automating end-to-end process, but breaking it down to pieces and beginning automation with small steps. For example, before a support engineer deep dives into the system to investigate the issue, automation can help him/her gather the basic information by traversing through multiple related transactions and applications. Even better would be to update the ticket with all the information gathered, even before he/she starts working on the ticket.
Figure 2 explores how automation can be enabled at each level of support ecosystem.