Scaling Your Intelligent Automation Program: Challenges


Throughout this series on scaling your RPA program, we have shared insights about best practices and important program considerations in the areas of strategy, governance, people, and operations with the hope that they will help to increase your odds of success and avoid any pitfalls. With some effort and thoughtful planning, you can successfully scale an RPA program across your enterprise. However, even with the best effort and planning, it will be difficult to prevent every issue and avoid all obstacles. 

In this final installment of our examination of key considerations when scaling your RPA program, we focus on challenges you may face.

Scaling Challenges

1. Maintaining quality as your program scales can be a challenge.

As more people and departments get involved with RPA, attempting to validate the quality of everything going on in the enterprise program will not be possible. For example:

  • Reviewing every use case for automation suitability.
  • Code reviews to ensure the bot is not introducing risks to the production environment.
  • Documentation of the designs.

This is where tools, governance, and compliance will help. The goal is to put guardrails and guidance in place that ensures people understand what is expected and cannot veer off course. 

Your operating model should include specific milestones and artifacts that address important areas. This structure will drive an environment of compliance and give you the ability to pause on a bot if things such as a process area leader signs off or a code review was not completed. In order to do this, your operating model provides the rules of the road, the tools to help ensure people stay on the road, and that they can easily comply with the rules. Each of these governance aspects support each other and enhance your ability to maintain quality from afar.


2. Your best advocates can also be your biggest critics.

Early RPA users in your program will feel the growing pains. As you figure out what works for your organization, there will be friction in your processes, resistance from different internal stakeholders, and necessary changes you need to make to the processes in order to ensure your program satisfies company policies, expectations, and risk tolerances. Striving to provide high levels of stakeholder engagement and the most responsive experience possible for these early adopters is critical to maintaining support and confidence of the teams who have embraced automation. They are your best advocates and can also be your biggest critics when their experience with the program is subpar.


3. Hidden problems will be revealed at scale.

As the number of bots, developers, and different departments in your program grows larger, you may observe problems you failed to see when it was small. You will start to observe issues with polices, infrastructure, procedures, governance, tools, and processes, both in how they operate and where you have gaps in your capabilities. Be prepared to adapt and reinforce anything that is lacking in order to support the broader organization.


4. Process optimization will create conflict.

The promise of RPA is to transform the organization, but the temptation will be to automate processes as they are. Start by documenting the current process, but then take the opportunity to reimagine how RPA can be leveraged to work in a way not suited to people. 

For example, when processing vendor invoices, have the bot track metrics on the volume, amounts and top vendors, and then compare to trends looking at payment history and other factors impacting your cashflow. Armed with this information and the capabilities of a bot, you could then:

  • Have the bot send an email to the customer thanking them for making a payment early.
  • Use the bot to periodically reward good payment behavior with a loyalty credit.

Examine the process and see where tasks and information traverse people or departments and ask why. Does the process cross domains because of specialized knowledge required for the next step, or is there an intentional segregation to satisfy separation of duty requirements? Process velocity slows and creates an opportunity for error every time a business process changes hands. Because bots are not bound by the same limitations of people, consider how RPA can provide a better experience for your employees and customers.


5. Existing manual processes may not be compliant with company policy.

 While compliance issues might be tolerated with manual processes, they may be a showstopper when applying automation to those same ones. To avoid significant delays in your bot development process, anticipate compliance gaps and incorporate checkpoints in your operating model to look for them early in the build process.


6. Execution and delivery of use cases across the enterprise will be inconsistent.

There will be a bell curve of skills and capabilities that will impact delivery execution by each team. Some teams will excel while others will struggle. Listen to your stakeholders, promote good communication, make adjustments to the program when systemic issues are identified.

Another good practice is to provide support and enablement services to teams that have unique needs. Use case pipeline management will become an important tool in order to see trends in delivery. Develop KPIs focused on program velocity, publish these metrics, and use them to reach out to stakeholders who might not ask for help. While they may not tell you that they are struggling, they may be very vocal to their colleagues about how the program is lacking. Not only will the outreach be appreciated, it may even amplify your “digital” story throughout the organization. 

Word of mouth organic praise can be one of the most powerful tools to build the credibility of your RPA program.


New call-to-action

Navigating a Shared Services Implementation
5 Things to Consider Before Commencing a Shared Services Journey
Related Posts
3 Key Lessons for Family Office Deal Making
3 Key Lessons for Family Office Deal Making
Celebrating Data Privacy Day
Celebrating Data Privacy Day
Deploying Workday: Too Many Applications Result in Too Much Risk and Inefficiency
Deploying Workday: Too Many Applications Result in Too Much Risk and Inefficiency