Robotic Process Automation is buzzing right now, and companies went from “What is RPA and how can I reap its benefits?” to “RPA works great for me, what are the next steps I have to take in order to be able to scale RPA drastically?”. At Roboyo, we focus on delivering the best solution for every client designed according to their specific needs. We have implemented different software solutions while finding new ways of combining technologies to enhance the capabilities of RPA. Additionally, we found out which aspects to consider ensuring a smooth and fast implementation of automations. At Roboyo, we never stop exploring new strategies and technologies to optimize the RPA journey of each client. Let’s take a closer look at 7 aspects that we filtered out over the last years on how to bring RPA to the next level:
I. Get Everyone Involved
First of all, RPA is a team sport. A fluent and efficient Implementation starts with a well informed and communicating team that is made of:
The IT Department needs to provide all necessary systems, access rights and check for possible restrictions based on the IT landscape that need to be considered
The final implementation is done by the RPA developer. They need a well written PDD (Process Definition Document) to be able to build a robot according to the business’ needs.
It is always important to stick to the automation schedule, coordinate the tasks and responsibilities between everyone involved in the automation process.
The Business Analyst writes the PDD together with the SME. Before that, an analysis of which processes are suitable for an automation with RPA and which changes in a process need to be made (e.g input format) have to be made. The more effort you put into writing a precise PDD, the faster and better the final implementation will be – do not rush it, although it might seem unnecessary sometimes.
SME (Subject Matter Expert)
There are always exceptions to every process. The SME makes sure all known exceptions are described in the PDD and consequently can be taken into account during the automation.
Often, the Process owner is the one who initiates a use case for RPA. It is important for all Process Owners (and SMEs) to look out for Automation opportunities and communicate them to the CoE (or a Business Analyst), hence they can check if an automation is reasonable.
II. Establish a Center of Excellence (CoE)
Investing into a CoE is really worth the effort. It is a place for new ideas, the process pipeline and for collecting learnt lessons as well as an accumulation of reusable RPA components to drive and speed up the RPA journey across the whole organisation. As UiPath puts it on their website: “Regardless of its task, a robot’s essential job is to make the company vision come true. In our approach to automation, we think of RPA as a long-range capability meant to empower organizations to evolve in terms of strategy and increase business value. […] A Center of Excellence (CoE) is essentially the way to embed RPA deeply and effectively into the organization, and to redistribute accumulated knowledge and resources across future deployments.“
Redistributing knowledge is key in the ever growing, fast rising RPA world. Keeping this in mind we will move on to the next aspect.
III: Take Advantage of Reusable Components
With every implementation, each developer will create new workflows and find easier or more stable ways of automating a specific sub-task (e.g. a login into a platform, transferring files between several applications etc.). Take advantage of these learnings and share them with your peer developers! Doing this will enable the development team to save time and improve their automations. The learning curve, together with the RPA implementation across the whole company, will grow exponentially. Remember to always break down a complex process into smaller and more manageable components. These components can be saved to a shared folder and invoked into another workflow. Since October 2018, developers using UiPath have the possibility to even benefit from the work of developers from other companies by using “UiPath Go!”. It is a marketplace like an “app store” for plug&play automation kits. Just sign up and search for a component that solves your problem.
IV. Maintain a Pipeline of Processes to be Automated
Everyone hates the following situation: After weeks of experimenting and struggling with a problem, the initial boost and motivation within the team is gone. Even if you created a great solution, the momentum of the project is slowed down. Prevent this by always keeping the pipeline of processes to be automated filled with projects for the future and communicate realistic time schedules. If you do this correctly, you will have colleagues approach you with new project suggestions regularly. This effect facilitates change management across the whole company and helps scaling RPA without putting too much effort into convincing stakeholders.
V. Write Clean Code and Stick to Best Practices
Writing clean code takes more time and often is annoying when you want to get things done quickly. However, this extra work will pay off in the long run. Remember: automated processes are built for the long term and are usually carried out thousands of times. Processes can have up to thousands of steps and you could easily lose understanding of what your robot does. You should implement Best Practices that developers can abide. (If you aren’t a developer you can learn from this:
Don’t put your development team under too much pressure
Let them do their job and let them do it correctly
They know what they are doing and it will be worth it
VI. Combine Technologies and Scale RPA Drastically
This is where the magic happens. RPA alone depends on structured input, human help when making decisions and is ergo limited in its own capabilities. Adding the right tools can do the trick and be very effective. For example, by using ABBY, an advanced OCR tool, the required input can be much more variable. Adding ABBY into your automation lets the robot easily read and “understand” written data. With Natural Language Processing (NLP), also language input can be processed. Think of it as when you are using Amazon’s Alexa. A customer call could automatically start a robot which processes all necessary data. Then it sends out an Email to the customer with the needed documents attached. We even created our own tool to boost and improve our implementations. “Snitch” automatically checks an UiPath project for Best Practices. Our developers appreciate this feature as it prevents mistakes early on.
VII. Provide Suitable RPA Training
It’s all to do with the training: you can do a lot if you’re properly trained. By Queen Elizabeth II.
Help your RPA Team perform at their best by providing the training they need. There are online (video) courses for UiPath, and tailored classroom trainings available. We offer both, on site and online courses tailored to suit your needs. Learn more about our training offerings here.
Do you still have any questions?