7 Expert Tips to a Successful RPA Implementation

RPA is taking the spotlight as the latest innovation in business process management. Our RPA experts at Robotic Process Automation (RPA) is the latest innovation revolutionising business process management. According to Gartner, by 2020, the RPA market will top $1 billion, with 40 percent of large enterprises having adopted an RPA software tool, up from less than 10 percent in the market today. Researchers at Hadoop estimate that the potential savings that companies will experience with RPA by 2025 is between $5 trillion to $7 trillion, with tasks performed equalling the output of 140 million FTEs. Statista believes that the RPA industry will be worth $3.1 billion by 2019 and $4.9 billion by 2020. According to Forrester, this figure is more likely to be around $2.9 billion by 2021. The number of major companies implementing RPA technology into their business processes is rapidly increasing, further proving the age of automation is upon us.

What about your enterprise? Has your organisation already started to adopt RPA, or is it considering the idea of automating processes? Here are 7 tips from our experts providing a better understanding of what types of RPA exist, and ways to successfully implement it into your business processes.

1.Understand and choose the right RPA type to match your problem

There are several variations of RPA to consider before selecting a vendor. To get value out of RPA, it is important to know which variation is best suited for your automation and AI needs.

Types of RPA

Some projects fail just because the chosen RPA solution was not the right fit for the job! Therefore it is important to have a good understanding of what types of RPA are available. Let’s take a look at the four use case types of RPA – Runners, Repeaters, Integrators and Rogues:

Runners and repeaters for simple and repetitive tasks

RPA “runners” and “repeaters” use cases employ fixed business rules. A runner is meant to automate daily activities covering simple tasks like processing input data in a certain way. For example pre-sorting applicant CVs based on rule-based criteria to support HR. When your processes are less regular and have higher complexity in terms of information required to complete the task, repeaters are your best friend. Repeaters can carry out larger and more time-consuming tasks like closing the books at the end of the month. Both runners and repeaters help save your backoffice FTEs for more valuable and effective purposes. Runner RPAs are often a good starting point for organisations as it is relatively easy to implement, and quite tactical in assisting or replacing employee activity. Cost reduction is one of the main reasons to do this, but don’t underestimate the value of improved quality and reduced time to deliver.

Repeater example: DSM nutrition health and sustainable living company

DSM was looking for a faster way to complete monthly and quarterly closings. They therefore implemented a ‘Repeater’ RPA type. By automating 489,000 tasks using 44 processes and 190 company codes, RPA turned 15 days of work done by employees into 3 days of work requiring no human interference.

Runner example: Aegon life insurance

Aegon needed a faster way to implement changes in a customer’s insurance plan (i.e. marriage, divorce, buying a house), and implemented a Runner type RPA to complete this function. From using 17 applications across 195 different screens, this process was streamlined from 141 minutes to 21 minutes of work.

Integrators to replace humans as solution for IT gaps

While the concept is quite simple – keep input data similar between systems – in reality this helps employees by considerably reducing their manual labor. Many business processes are often not seamlessly matched by supporting IT solutions. Employees often have to fill this gap between systems by acting as ‘integrator’, manually copying information over from one system to another to continue the process. An integrator software robot solves this instantly and without mistakes, all the while allowing for consistency among all departments, and keeping all records up to date as information is received over several different systems.

Rogue RPA for intelligent and self-learning automation

A more advanced type of RPA are the ‘Rogue’ solutions. When processes require multi-variable inputs and considerations for decision making, artificial intelligence and machine learning can be combined with automation. The combination of AI and RPA means that the automation software learns from historical data and human interaction. This type of RPA is useful for customer relations. For example, it can provide recommendations for follow up actions, pricing, discounts and other retention tactics.

2. Start implementing RPA in the departments with a large volume of standard processes

We recommend starting with RPA in departments and functions that typically involve a large amount of standard processes using fixed business rules. Often this already allows to automate up to 60% of the back office processes. These opportunities should form the basis for your enterprise automation roadmap. It also provides a relatively non-intrusive start in automation, which brings the largest chance of success. Capture quick wins first and establish the organisational foundations for RPA (governance, experience) before moving on to more advanced types of RPA/IRA.

Where we often see RPA being used:

3. Check the economic viability of automation opportunities and prioritise accordingly

The next step is to look at the economic viability of the specific opportunities within these departments, as well as their feasibility and desirability from an organisational perspective. What is the effort and investment vs. the return? Does it make economical sense for your organisation to invest into RPA projects? After shaping the general needs RPA can satisfy, the first stage of an RPA implementation initiative should always involve a process and opportunity scan together with the relevant stakeholders. This is done to create a more specific business case; sanity checking if the identified opportunity and solution still makes sense – before moving into the design and implementation stage.

4. Get buy-in for RPA initiatives from management and stakeholders

Without awareness and understanding amongst stakeholders on what RPA is and how it will benefit the business, it is hard to run an effective project. To fully realise the benefits of RPA, the department responsible for the processes affected should be involved and consulted. Leadership in your company should help people understand how RPA can benefit them, provide opportunities, and make their lives easier. Not only does this build confidence around the implementation of RPA, but it also prevents the activation of the organisational immune system.

5. Pace RPA scaling to match speed of business adoption

As required implementation time can be very short for RPA, it is easy to move too quickly from a programme and technical perspective. In other words, it is easy to add more robots, but if the organisation cannot handle the implications, it will do more harm than good. Start small, get the organisation used to the process and add more as you see fit. A successful implementation often includes a post-implementation adoption process to help manage this.

6. Involve IT in RPA from the beginning

RPA is often seen as a business initiative rather than additional software. Executives therefore can make the mistake of thinking IT does not need to be part of the implementation process. However, technical expertise is often required to smooth over unforeseen problems caused by RPA. IT can also help in the RPA vendor selection process, making sure the organisation does not pick the wrong RPA solution for the job. Next to that, there are often already existing tools and solutions in place, which can have a significant part of the functionality required. IT can help assess this alongside RPA tools to get to the best solution. Last but not least, IT can even share some best practices in reusability that are relevant in the RPA world as well.

Vendor solutions mapped to capability focus

7. Setup a clear RPA governance model for implementation and BAU

Be clear about the governance model. There are several models that can be used with different implications – centralised, hybrid, federated – but always ensure clarity around who is managing what, the material and processes RPA will handle, and how each employee will be involved with RPA before, during, and after implementation. Lack of communication and awareness can cause significant damage. For example, in one situation, when an employee changed the company’s password policy but no one programmed the bots to adjust, it resulted in significant data loss. Keeping bots running efficiently requires all personnel to be fully aware of what changes can impact the programmed business logic and adjust accordingly. Maintenance is key when it comes to keeping RPA processes moving productively.

High level view on types of RPA governance models

Closing thoughts

RPA is taking the spotlight as the latest innovation in business process management. Implementing RPA into your organisation provides huge opportunity for cost saving as well as revenue, helping to digitize labor using machine intelligence. Careful selection on the RPA type that matches your processes requirements is essential. But the real success lies into the business adoption and governance model behind it. Ensuring RPA is used and maintained properly – allows the continued deliverance of value that RPA was designed to bring.


Want to hear more about our experience, or looking for support on your RPA initiatives? Get in touch with our RPA lead, Mathieu Jonker for more information or check out our RPA solution page for more insights into how we can help your organisation embrace automation.