When most organizations begin their Robotic Process Automation (RPA) journey, they’re often laser focused on a couple of nagging, repetitive, and expensive tasks they’ve been looking to automate for some time. Once those tasks have been automated, though, where do they go from there?
Are those tasks the end of their RPA journey or only the beginning? Where else should they be looking to grow their program and create additional automations?
There are actually three types of bots your organization should consider as you continue to mature your RPA practice and expand the way you view automation within your environment. By understanding, exploring, and applying these three bot types, you’ll find new and insightful ways to employ RPA to help meet your business needs.
This is the most obvious category — and probably the reason an organization would bring in Automation Anywhere in the first place. Automating high-volume, repetitive, rule-based tasks in an auditable and reliable way is the goal of most organizations when they bring in an RPA solution — and that's great.
It’s the first type of bot for a reason: It’s the most common. This is the bot that can take care of tasks that weigh down employees, teams, and even entire departments. These bots typically automate “robotic work.”
Robotic work is created when new systems are brought in that don't integrate well with existing applications, and humans are tasked with being the glue to help these disparate systems connect.
When humans are asked to do robotic work — transferring data from a spreadsheet to a web app or copying data from one screen to another, for example — it keeps them from doing things that can help move your business forward.
But, when robotic work can be automated, humans can focus on human tasks. Human tasks are higher-value responsibilities that improve customer experiences and propel business initiatives. Additionally, when robotic work is automated, your organization will benefit from improved business process efficiency, data quality, and employee engagement.
Once teams have identified and automated most of their stuff-we're-doing-today tasks, they have the space to think about the things they as a team should or could be doing instead of only the tasks they have to do. If a team is tasked with transferring data from incoming forms to a web app or completing tasks in one workflow by kicking off a task in another, for example, they don’t necessarily have time to focus on how else that process or they themselves could add value to the organization.
With those tasks automated, however, employees and teams can begin to explore how they can bring additional value to the organization through enhanced reporting or improved task tracking to create a better, more informed customer experience.
What if those long-running reports on the team’s throughput that are run quarterly could be run weekly, or even daily? How could access to near-real-time reports give leadership better visibility into bottlenecks or gaps in the process? How could an organization get a better grasp on where work may have been missed or identify service-level agreements (SLAs) that have been violated?
Scheduled bots running reconciliation reports and SLA reports can give helpful insights into where steps may have been skipped or where work may be held up — which is often the first step in improving and streamlining the process.
From a systems perspective, what if there were bots to proactively perform system checks so issues could be identified BEFORE a failure or outage report comes in from a customer? Even if your “customers” are other systems integrators, proactive issue resolution makes for a much more professional experience.
And what about those deployments — are they validated and automatically tested in all your test, production-fix, and dev environments? Are those environments themselves even up? Are the environments in sync on the agreed-upon dates to stay in line with the release schedule of other dependent applications?
The main point to consider with the stuff-we-should/could-be-doing bots is, “If time and resources (money/people) weren’t a limitation, what could be done to improve the way our team contributes to the company’s objectives and, ultimately, to the best service of our customers?”
These are the bots that enable us to do exactly what their title says: Make better business decisions. If data exists in disconnected systems online, or in some format that doesn't perfectly work for you, use a bot to bring it all together, format it, and help you make more informed, data-driven decisions.
This bot type is a personal favorite of mine. About a year ago, I was in the market to buy a used vehicle. I knew the make and model I was interested in and wanted to make sure I got the best deal — but how could I know I was getting a good deal? With different mileages, model years, and asking prices, it’s tough to compare two vehicles of the same make/model/year — let alone across model years.
I wrote a bot that scraped the prices of ALL the Toyota Priuses across the nation to get their model year, current mileage, asking price, and listing URL. I then had the bot apply a formula I created: asking price divided by the expected life of a vehicle (150,000 miles, for example) minus the current mileage to determine the price per remaining mile of the vehicle’s expected life (see Figure 1).
This isn't a perfect metric by any means, but it gave me an additional data point to consider as I looked at vehicle listings. After the bot had scanned thousands of listings nationwide, I realized it would be about USD 3,000 cheaper for me to buy a one-way plane ticket, fly to Florida, buy a Prius, and drive it back.
Had I not used a bot to assist in that decision-making process, I likely would have overpaid for a local car because I didn’t have enough data to make a sound decision.
My personal example aside, if gathering data and using it in a new way could help you make better decisions, then a bot can surely help with the assembly, gathering, and formatting.
Could your business or team gain an advantage from reading pricing information from different portals to put together a competitive analysis? What about a bot that could check with different vendors as you book travel to make sure you’re getting the best rate? If you had data for every one of your customer’s transactions from intake to delivery, would that help identify bottlenecks/inefficiencies and improve customer experience? Absolutely.
Decision-making backed by hard data is something many organizations struggle with. Be creative and intentional — you’ll be amazed at the insights you can gather just by thinking of new ways to put data together and analyze it.
If you would categorize most of your bots as stuff-we're-doing-today bots, don’t worry, you're not alone. As you continue to grow the RPA footprint within your organization, think through how these other bot types can help you gain new insights into the work that's currently being done — and help you make better, data-driven business decisions from the data that may be spread across multiple systems.
If you’re in need of some inspiration or tools to help in your exploration of new bot types, check out some of the prebuilt free and paid bots in Bot Store. They can be helpful examples in thinking through the art of the possible and accelerating your bot-building process.
Go explore new ways to create bots to meet company objectives, serve your customers better, and make smarter, data-driven decisions.
EXPLORE RPA COURSES
Micah Smith is a developer evangelist and lover of all things automation. His background is in Robotic Process Automation, document imaging, and optical character recognition, with a specialized focus in financial services and government sectors.