This is a guest post by Rahul Varshneya, the co-founder and President of Arkenea. Rahul has been featured as a technology thought leader in numerous media channels such as Bloomberg TV, Forbes, HuffPost, Inc, among others.
Several organizations use a wide range of tools to execute and monitor their internal processes. These systems are interfaced by human agents who pass on the data output from one system as input to the next system in the pipeline.
Because of the non-automatic nature of manual execution, manual data entry work makes these processes slow and error-prone.
Wouldn’t it be nice if you had these interfaces taken care of automatically?
That’s right, we are talking about Robotic Process Automation (RPA) – a transformational paradigm that optimizes such processes by automating repetitive, rule-driven tasks.
According to Forrester Research, RPA market revenue is expected to gross $1 Billion USD in 2019.
With RPA and by using RPA tools, your internal processes can be finished faster and more efficiently. It would free your agents from trivial data entry and data exchange tasks so you could deploy them to do more strategic process initiatives.
In this article, we at Process Street will present 5 best practices for RPA. We will then introduce our favorite RPA tools, so you can walk away with actionable robotically automated processes.
This article has been structured as the following:
- How does Robotic Process Automation eliminate tasks from structured environments?
- Software silos and how they affect your teams
- How to break silos with the implementation of Robotic Process Automation
- How to achieve exceptional results with Robotic Process Automation implementation
- RPA tools: Process Street’s top four RPA tools
- How you can use Process Street’s RPA tool capabilities
Let’s get started!
How does Robotic Process Automation eliminate tasks from structured environments?
Every major function of an organization has introduced automation in some way or another to eliminate repetitive processes from structured and predictable environments. Standard Bank of South Africa reduced its customer onboarding time from 20 days down to 5 minutes, all thanks to their investment in smarter processes led by RPA.
From data entry automation to leveraging marketing automation tools, RPA has found its foothold in all segments within the organization.
Let’s take claims processing as an example.
Robotic Process Automation: Claims processing example
Typically, an agent first records the claim data in a spreadsheet. During this data entry, he will quickly verify that all necessary and tertiary data fields are filled to completion. This is followed by a complicated series of processes that include authorization and issuance rules to check how valid the claim is. This second part of the process is where the agent takes time and may have to consult one or more supporting software systems.
An enterprise RPA solution will take over the authorization and issuance checks by deploying a bot process that follows the same steps as the human agent. This means minimal probability of errors and a faster process. The bot can execute tasks without human limitations and without slowing down.
A large healthcare organization experienced 50% savings in the claims processing turnaround time. Over the course of the year following the RPA solution deployment, they earned $2 Million USD courtesy of these RPA-powered investments.
Robotic Process Automation: Marketing example
Marketing also serves as an incredible example of how the automation of robotic rule-driven tasks can benefit your company.
A large database of leads is of no use if you can’t target them at the right stage of the customer journey. It is part of the job for marketers’ to use this list to its full potential. However, it’s not practical to track the activities of thousands of customers in the email list, to send them emails at specific intervals in their customer journey.
Marketing tools are used for a number of routine tasks such as tracking user activity on the brand website and sending emails to large email lists, thus generating a predictable revenue via email marketing.
This allows marketers to track and nurture leads automatically through mass email marketing.
As an example, we can look at Yves Rocher’s marketing automation (MA) case study. As a part of their transformation, they experienced the following benefits:
- Growth in email efficiency by 1200%
- Dynamic emails recorded a 50% growth in open rate and a 15% growth in click-through rate
- Increase in sales when 30% of first-time customers bought targeted products suggested through dynamic banners.
Software silos and how they affect your teams
In the agricultural storage industry, silos are large cylindrical columns that store grains. The term ‘software silos’ or ‘information silos’ describes a similar phenomenon in the IT industry as well.
Silos happen in IT when teams working on shared tasks try to reduce their dependencies by not sharing important information with other teams. This is a way for them to derive power in group dynamics.
Knowledge is not shared, members make assumptions in the absence of crucial data which leads to expensive last-minute changes, and may even cause trust issues among the teams. When more and more people start adopting this train of thought, the ‘silo mentality’ starts to destroy productive company culture from inside.
Typically, siloed teams tend to reject suggestions for improvement and generally seem hesitant to new ideas. When queried, such teams often share information only on a need basis or delay sharing them.
How to break silos with the implementation of Robotic Process Automation
When implementing RPA, you can either hire RPA developers or train in-house resources from within your team. They will create RPA bots, which are simply small pieces of software deployed to automate specific tasks.
During this process, you will not have to create deep system changes as RPA is meant to integrate with your existing systems, without sweeping overhauls. This means minimal disruption and is also one of the reasons why RPA adoption is growing every year.
Implementation is the key to extracting numerous benefits of RPA.
Patrick Lencioni goes further than this in his book titled Silos, Politics, and Turf Wars:
Because the teams are segregated as per their function, it minimizes the opportunities for collaboration. The RPA development team may interpret things if the material they receive is not descriptive or comprehensive. This may also cause misunderstandings if the business team finds a major variance from their expectations in the final result. Subsequent rework may extend the RPA implementation timeline and may even affect future RPA projects.
Instead, RPA implementation must be adapted to fit the needs of a modern team in a way that allows them to predict and solve problems on an ongoing basis. It’s all about collaboration and knowledge sharing to realize business goals.
Faster RPA deployment is key to minimizing roadblocks and maximizing RPA benefits. Without structural changes and support from management, your RPA implementation will only supplement existing silos.
To help break the silos:
- Let teams collaborate and communicate at all important junctures of the RPA lifecycle
- Open communication channels with regular stand-up meetings
- Keep emails focused on a single intent
- Hear out any concerns during meetings and use them to plan for contingencies by focusing on their potential solutions
How to achieve exceptional results with Robotic Process Automation implementation
All transformations start with a clear picture of the process. In this section, we discuss some best practices that will help accelerate and refine your RPA implementation while similarly reducing rework and time/money wastage.
1. Start with detailed Process Design Documents and Solution Design Documents
Before a baby learns to walk, he/she must first learn how to move thier feet. Similarly, an organization using RPA automation for the first time cannot directly start implementing without preparation. Simply put, it would not be wise to skip the planning stage and start implementing RPA bots.
Two must-have documents that help plan your RPA implementation are:
- Process Design Document (PDD)
- Solutions Design Document (SDD)
What is the Process Design Document (PDD)?
PDD is a document outlining the exact steps necessary to complete a particular task or a process as the user performs it and how the system responds to it. It is a user-level document that captures the existing ‘as-is’ process in full step-by-step detail.
A business analyst usually captures this information required for the PDD. An RPA developer may not directly use this document as it captures the problem and not the solution. This document must be translated into a Solution Design Document (SDD) for it be useful to RPA developers.
A PDD is typically a spreadsheet containing the following fields:
- Process name
- Sub-process name
- List of steps in the process/sub-process with a sequentially marked serial number
- Any additional comments/pre-requisites corresponding to each step
- Name of the system being used and the credentials for its main account login
Note that this PDD will change as the requirements of the process get more complex. Talk to your stakeholders and find out from them what attributes of the process are important to them.
What is the Solution Design Document (SDD)?
SDD is all about how the RPA bot will process the task. So it will capture all the scenarios that a bot might face during each execution, and what steps it will take during those scenarios.
The goal of SDD is to capture everything an RPA bot is supposed to, which is currently done by a human agent. Besides the core computation task, this includes data scanning, notifications, validations, and error handling.
The RPA developer uses this sheet directly as a reference while writing the code for an RPA bot.
The fields in an SDD are the same as the fields of a PDD, but with a few modifications.
First, instead of process steps, you will capture bot steps to describe the actions the bot will take at each step.
Secondly, you have to mandatorily include an error-handling sub-process to program error handling and recovery steps into the bot’s SDD. This will prevent any embarrassing crashes and system outages.
Data validation can also be included as a separate sub-process in the SDD.
Careful planning and solutions can help companies identify risks early and be ready with a solution when those risks are realized.
2. Log errors and provide detailed error descriptions
Needless to say, when your bot is handling errors, it must provide a readable description of the error to its (human) admins.
Let the bot capture the complete trace of steps it took which led to the error. With each step, the bot will also log its state (input, output, and temporary values). This is called the stack trace and can be sent via email to help the admin debug an issue faster.
Apart from email, having a log file or a log server capturing all error outputs of RPA bots makes debugging easier. Ensure that your RPA bot follows different logging levels to help developers monitor its performance.
3. Keep variables and arguments configurable
Sometimes, RPA developers use static values for common variables such as an email address for sending error logs or a website URL to fetch the data.
They would assume that the value will never change and therefore does not need to be configurable. So, they add static values for these variables in the RPA code. These are called ‘hard-coded values’.
Hard-coded, static values are discouraged in any programming language for any kind of program. Instead, set guidelines for your RPA developers to avoid hard-coded values.
Ask them to put important values in a file in a name/value pair. They are called configurable variables.
Configurable values can be stored and changed from configuration files. They are an easy way to supply fixed values to a bot.
In this process, the bot reads the value by using the name of the variable, which stays the same in the file and the code as a link between them. The code remains abstracted from any changes in the file because this name won’t change, only its value will change. The bot reads this new value dynamically.
Want to change the value supplied to the bot?
Change the value of the relevant property name in the file, the code of the bot remains untouched.
This has two advantages:
- As all bots needing that value will read from the single file, the amount of rework to change the value is minimized to a single text file
- Business users, who don’t understand or deal with code on a regular basis, find it easy to edit the text file. It makes the bot configurable by business users, besides the IT team
4. Explore workflow reuse and process modularization
Your programming workflows are costly. They cost your developers’ time and effort directly and they cost you (i.e. the organization) money indirectly.
Therefore, small pieces of repeatedly used code must be kept in publicly named blocks called ‘functions’. Functions are simply names assigned to frequently reused pieces of code. So, instead of rewriting the same piece of code again, you can refer to the function using its name. Your RPA tool will execute it for you.
This block-based execution is called modularization of code. It saves time and effort for the RPA developer and ensures your code stays nimble.
For example, login functionality is a commonly reused piece of code. Before using any tool, the RPA bot must log in first into that tool. Thus, the ‘login’ function is an easy reuse candidate.
Let’s assume that the bot uses URL A to authenticate all the credentials it receives. If you decide to change the authentication service that your RPA bots use, you don’t have to go to each place in the code where the bot logs into the tool. Simply update the URL from A to B in the ‘login’ function. The rest of your application stays unaffected by this change.
If you have several of these reusable functions, create a library of reusable resources and components to manage them easily.
Do note that ‘over-modularization’ is also a thing, where developers create physical boundaries in the code (i.e. functions) to simulate logical boundaries between the substeps.
If the function is being referred to by only one block, it is not a candidate for modularization. The RPA tools may even face a performance lag if too many such single-use functions exist in the application or if lengthy functions are called only once.
You can also grab one or more of the several types of RPA plugins, which are basically pre-designed and tested repeating RPA components.
5. Consider custom app development if existing RPA solutions don’t fit the bill
Finding an existing RPA solution to fit your organization’s needs is a lot like finding a needle in a haystack, there are many solutions for each different type of task.
If you want to automate many tasks across multiple departments, you are less likely to get a single solution across all the functionalities. Therefore, talking to a custom app development company can help you analyze your needs better and define what exactly you are looking for in an RPA solution.
You don’t have to know all the steps for building apps, the company you consult with will lay down the development process for you. Let them take care of developing your RPA app, whilst you define your processes to enable them to help you better.
RPA tools: Process Street’s top four RPA tools
If you are ready to implement RPA into your business today, then you should check out Process Street’s favorite RPA tools, detailed below.
RPA tool #1: Another Monday
Another Monday offers complete automation of your processes from end to end. Quick scaling is possible with increased efficiency through the standardization of process parts.
Automation is provided through Another Monday’s new tool, AM Muse.
RPA tool#2: Blue Prism
Coding knowledge is required with Blue Prism, however, the platform is user-friendly for developers.
A multi-environmental deployment model is supported, that can work for any application and can be used on any platform.
RPA tool#3: Automation Anywhere
Automation Anywhere provides core capabilities on-premise and cloud services. The tool is user-friendly providing bank-grade security, real-time reports, analytics and platform independence.
RPA tool#4: Nice Systems
Nice Systems is named as NEVA – Nice Employee Virtual Attendant. This tool is a smart tool for helping employees with repetitive tasks. Attended and unattended server automation is provided that is cloud-based and gives an on-premise solution.
RPA tool #5: Zapier
Zapier moves work between everyday apps for automation purposes. With no coding knowledge required, you have the potential to connect 1500 apps with this integration tool.
How you can use Process Street’s RPA tool capabilities
Process Street is superpowered checklists.
With Process Street, you can document your business workflows in a checklist format.
By implementing checklists in your business, you will increase workplace productivity, efficiency, and effectiveness. Checklists have also been proven to save you time. By implementing both checklists and RPA in your business, you will save heaps of time whilst bringing together the benefits from both.
At Process Street we have worked to make our checklists compatible with RPA. How have we done this?
Despite the fact that Process Street is not a dedicated automation software per se, your business processes are documented in a format that makes RPA simple and easy to implement. That is, your documented business processes are digitized.
Once you have recorded your business processes in a digitized checklist format, you can create custom links to run checklists from inside other apps.
To learn more, check out our article on creating checklist run links.
Have a browse through common Process Street integrations by visiting our Help site’s Knowledge Base, Integration page.
For more information on process automation within Process Street, refer the following articles:
- What is Digital Process Automation? The essential BPM Evolution
- The Ultimate Guide to Business Process Automation is Here!
- What is Zapier? The Essential (Unknown) Time Saver
If you are new to Process Street, you might be wondering why our checklists are considered superpowered.
Aside from the fact that our checklists are compatible with RPA, Process Street’s checklists include the following features:
- Stop tasks to ensure task order
- Dynamic due dates, so no deadline is missed
- Conditional logic, creating a dynamic template that caters to your needs
- Role assignment, to ease task delegation within your team
- Approvals, allowing decision-makers to give the go-ahead (or rejection) on important items. In addition, necessary comments can be provided
Find out more about Process Street and our offerings by watching our Monthly Webinar: An Introduction to Process Street.
Go on an RPA-based transformation today: Plan it well and implement RPA tools
Simply deciding to go on an RPA-based transformation is not a predictor of its success. The transformation must be planned well lest it becomes unmanageable.
Effective RPA implementation begins by breaking down silos in your organization and enabling open communication.
As they say, well begun is half done. So, start by describing your as-is solution in a PDD and document the actions of your bot in an SDD. You may also need to document a to-be state in your PDD if you are using RPA as an opportunity to overhaul your current processes.
Do you have any questions regarding Robotic Process Automation not covered in this article? If so, please comment below. We would love to hear from you, and who knows, you may even get featured in one of our upcoming articles!