When developing your robotic automation, the temptation is to look at the existing human process, document all the current manual steps and tasks, and programme the robot to replicate them. Unsurprisingly, you will get the same outcome – but typically around three times quicker. Job done … right?
Well, not necessarily……… if you slavishly copy the human process you may not be getting the best out of the software. Also, this approach pre-supposes that the current human process is optimised. Look at the process critically first to determine whether improvements can be made. Take time to understand the capabilities of the automation software and how it can do things differently (as well as quicker). For example:
- Think about how a person accesses the system – they will usually log in and use the system via user menus. Quite often a robot can access the system via an API and bypass the menus. This is much more efficient and helps future proof the automation.
- If an API is not possible, think about how to use the menu option. A person will often save their regular choices as favourites – never programme the bot to use favourites – these can often be changed and then you are left wondering why your bot stopped working!
- Often, humans need interim steps in a process, such as saving data to another file and sending to someone else for input. A robot probably doesn’t need to do this and can work from significant amounts of data in memory.
- We often use standard forms when inputting data. Many automation software packages will support smarter forms where you can pre-populate data from other systems, perform validation checks or propose the next best action. This starts to move you towards much more intelligent automation.
A final word from Dilbert……..
If you would like a conversation with our in-house expert in RPA and AI – Rod Chadwick, about how he and the team at DKA can support you on your journey of process transformation, then please reach out via firstname.lastname@example.org.