9-Step Roadmap for RPA Implementation in Healthcare 30 May 2024

RPA Implementation

Robotic Process Automation (RPA) needs no introduction. You are probably already familiar with RPA and why it is regarded as one of the most desirable ‘developing’ technologies. The benefits and business impact of automation (improved productivity, reduced errors, cost savings, etc.) make the investment worthwhile. When incorporating emerging technologies in your healthcare organization, having a clear and precise action plan makes the process go smoothly. For example, invoice processing is one of the most common automated operations. Detailed research and project strategy are required to construct a well-oiled AI engine (with machine learning and retraining capabilities) that processes invoices quickly and accurately. The sample below depicts the final process flow for such an RPA project.

Once you’ve decided what you want, you’ll want to know how to achieve it. So, we’ve created a nine-step roadmap for RPA deployment and implementation.

  • Identify processes that have the potential for automation.
  • Prioritize processes.
  • Obtain stakeholder alignment.
  • Delve deeply into the process.
  • Choose your technology colleagues.
  • Develop your solution.
  • Test, test, test! (and retrain the solution.)
  • Run a pilot.
  • Go live and plan for the future.

Identify processes that have the potential for automation:

You want to get the most out of your RPA investment, just like you would with any other large investment. This requires identifying processes that are clearly in need of automation. Such processes typically include:

Cost and/or revenue drivers:

This is a given, but it’s equally crucial. Consider the most expensive procedures that impact external stakeholders (customers, suppliers, etc.). For example, while it is critical to process supplier invoices accurately and on time, manual invoice processing may fall short in terms of speed and efficiency. This is why invoice processing is an ideal choice for RPA.


Since we’re talking about efficiency, processes that are prone to human error can result in customer discontent and regulatory issues. The more manual errors there are in a process, the higher the return you can receive from automating it.


This one appears to be contradictory, but it is not. We hope to eliminate manual errors through automation, but if the process cannot tolerate any inaccuracies, you may not want to automate it just yet (or at least implement a rigorous quality control mechanism). This is because RPA bots employ established rules and user interfaces to complete jobs. If these rules/UI change, the bots’ output may contain mistakes if they are not set to shift in response to the process modifications. For example, invoice processing is often automated. However, invoices with payments exceeding a certain amount must be reviewed by humans because the bots are not designed to handle such payments. Similarly, many firms continue to use manual teams to evaluate the bots’ output and spot any problems.

Volumetric and speed-sensitive:

When you imagine RPA, you think of simpler, faster procedures. Any processes that involve a significant amount of human work and cause a delay in the delivery of products/services to end users are suitable candidates for automation.

Prioritize processes that are easily automated:

Now you want to prioritize the procedures depending on their ease and feasibility of implementation. Such procedures tend to be:


RPA’s core concept is that bots are trained to follow programmable instructions and accomplish specified jobs. It is assumed, then, that the process you choose for automation has a set of rules that can be defined, mapped, and coded. It is not enough to choose processes with simple rules; it is also necessary to choose processes with well-stated rules.

With minimum deviations:

After being educated with sophisticated rules, all bots can delve further into the process and discover rules that their human managers may consider ‘a given’. Although this is a benefit of RPA, if a process has too many exception instances and unrecorded rules, it is up to the RPA implementation team to identify and record these rules through interviews with process managers. This increases the implementation complexity of RPA, but it can be circumvented by choosing a process with fewer anomalies.


Consider where the process is in its lifespan. If it was just launched and is still in its early stages of development, it may not be a suitable fit for automation because it evolves frequently and requires ongoing tweaking. To that end, legacy systems/processes are an excellent fit for RPA because replacing them with an automated solution not only allows you to pioneer RPA in your organization but also reinvents a legacy system that you were planning to replace anyway.


This criterion focuses solely on cost and effort estimations. Determine whether you’re looking at a process unique to your healthcare organization or an industry-wide, recognized process. If it’s a procedure that many businesses of comparable sizes perform (in similar ways), there are probably well-tested, well-developed RPA solutions on the market that you can employ. This way, you don’t have to spend more time, effort, and money on a cost-effective, bespoke solution that you can readily investigate and purchase.

Obtain stakeholder alignment:

This directly follows the final criterion in the cost matrix above. You can’t move forward unless all of your RPA implementation’s internal stakeholders (senior management, process team, and other business units) agree.

Senior Management:

If you run a small or medium-sized healthcare organization, top management is most certainly already actively participating in RPA conversations. In larger firms, however, it is sometimes necessary to actively seek management support by developing a compelling business case. The secret sauce for each business case is to do a thorough, rational examination of costs, benefits, and return on investment.

Process Team:

This is the team that currently runs and oversees the process you want to automate. If the process is managed in-house and takes years to develop and nurture, you may face some pushback. People’s main concerns about automation are redundancy and a lack of recognition for their efforts. So it is critical to bring this team on board by proving the importance of automation and providing proactive assistance for the post-automation ‘new normal’ (upskilling, finding positions for them in the same team or other teams that require their specialized abilities, etc.).

Other Business Units:

It is critical to include your company’s IT/technology professionals in the process from the start. Having their opinion sooner rather than later can have a positive impact on RPA research and business case preparation. They may provide information about whether the company has previously adopted RPA, whether vendors were employed, and their experience. Not to mention that your RPA project should align with the company’s future technical strategy, and technical experts are the go-to for this.

Delve deeply into the process:

After obtaining stakeholder alignment and before beginning development, you may wish to properly understand the process you intend to automate. This is where we understand and define the process rules that we identified in Step 2. Again (as in step 2), identifying and recording these rules through interviews with process managers may not provide an entirely accurate picture because some rules may be regarded as ‘a given’ by these managers who have utilized the process for many years. In such circumstances, process mining can reveal norms that are not obvious to human teams. For this aim, process mining software searches transaction records for actual, precise process flows that explicitly demonstrate each rule that the process incorporates.

Not to mention, this effort in understanding the process may yield process improvement suggestions that you can adopt prior to RPA. A simple example is to simplify process rules so that RPA bots can be simply developed and trained to follow them.

Choose your technology colleagues

Although utilizing a single RPA partner for an RPA deployment may appear quick and easy, you should consider a four-party strategy for a successful long-term deployment and good ROI:

Process mining software:

As previously noted, process mining can assist you in understanding the process in its totality (by specific process flows and rules), assessing the viability of automation, and monitoring the impact of automation on the process in the coming years.

RPA Technology Provider:

It goes without saying that when selecting a technology vendor, you must conduct due diligence; perhaps you might additionally examine variables to consider while considering your selections.

AI/ML Provider:

As seen in the cost matrix above, the data required for RPA deployment may have to be retrieved and evaluated from many sources. Once the data is in place, RPA is perfect for automating processes; however, to prepare the data for automation, AI/ML/data extraction providers should be considered. This is because RPA solution vendors focus on automation, and their solutions for phases before automation may not meet your specific requirements.

RPA Implementation Partner:

Of course, if you’ve gotten this far with an internal team, they may decide to implement the project themselves. If you want to focus on your core operations and rely on industry experience, consider hiring a consulting firm or a BPO provider.

Develop your solution:

Rather than automating the entire process, you may want to focus on automating specific aspects. Regardless of the size of the project, building a detailed process map is strongly advised. This is where your company’s process and subject matter expertise come in. After the plan is in place, RPA bots can be programmed. Also, keeping an eye on your RPA platform’s marketplace will help you find readily accessible code and prevent having to start from scratch.

Test, test, test!

The value of testing your RPA system cannot be overemphasized. Seemingly innocuous changes in settings (for example, running the solution on desktop vs. mobile, on various operating systems, etc.) can result in unexpected abnormalities. Testing the solution (and then testing it again) is the only way to iron out the bugs and prepare it for the pilot. You may also use this chance to retrain your solution to ensure that it consistently produces high-quality results.

Run a pilot:

As the name implies, you will be checking the actual output of the RPA bots you’ve created. Just because it’s a pilot doesn’t imply you should omit any of the processes you would take when creating a full-fledged project.

Set your aims and objectives:

While a goal is the ultimate goal of what you want to achieve as a pilot, goals are the steps that will get you there. Make sure your goals and objectives are SMART (specific, measurable, attainable, realistic, and time-bound).


Get the RPA bots to do predefined tasks in the process, and have the process team analyze the output regularly to assess quality and efficiency.

Monitor and course-correct as needed:

Evaluate the entire pilot’s outcomes, taking into account outliers and exceptions.

Go live, and plan for the future.

Once the pilot has met the predetermined goals, you are ready to design and execute the full-fledged rollout. Begin by creating a project strategy; this is where your organization’s project/implementation expertise may help. Below is our tried-and-true project management framework, which we’ve been using for over 20 years!

The following tools/concepts are excellent for developing a complete, unambiguous project plan:

  • Responsibility matrix (RACI).
  • Communication Matrix
  • Project tracking and monitoring matrix
  • Risk and change management contingencies
  • Quality assurance process

Once the plan is in place, all that remains is to keep all stakeholders updated and go live! Follow the Agile and Scrum frameworks with two-week sprints; this way, you’ll gain frequent iterative input from the target user, which you can use right away to improve product development.

Track, measure, and analyze results (both during and after RPA deployment) about business objectives (cost savings, productivity increase, revenue growth, etc.). As stated in your project strategy, it may be beneficial to build different frameworks for routine solution maintenance, user support, and future process team training.

If you’d rather focus on your core activities and leave the technical stuff to us, we’d be delighted to collaborate with you as your decision support system and create an RPA solution that is tailored to your specific needs. Our extensive experience serving healthcare clients, as well as our connections with RPA experts, enable us to provide the finest service possible!


Peter Paul

Technology Consultant

About the Author:

Peter has over 20+ years of experience in managing and delivering enterprise applications and IT infrastructure. He served several IT companies in the US and Canada before joining Velan. He is instrumental in deploying, managing and delivering latest technologies at Velan. He can be reached at peter.paul@velaninfo.com


Quick Connect With Us