Making your first steps with RPA

After careful consideration, you’ve decided that automation is definitely a priority for you and your business – but where and how to start? It’s important to reap the benefits of automation as early as possible but you want to balance this with progress at a pace which is manageable for everyone involved. This article introduces you to the automation lifecycle and the critical elements you should consider at the very beginning of your RPA journey.

It’s not always easy knowing exactly where to begin, and ultimately, this always depends on you and your company, for example, here are some key questions you need to consider:

  1. How committed are you and your company to RPA?
  2. What do you want to achieve with RPA and by when?
  3. How quickly do you want to progress?
  4. What resources do you have available to dedicate and what skills and experience do those resources have?
  5. Do you have buy-in and support from major stakeholders in your business, from the top down?
  6. Do you have budget and timelines defined?

If you’ve got firm answers for all of these, then you’re ready for the next steps, but if not, then you may just need a little help and time to get there, so read on.


The RPA Lifecycle

This is simply the series and sequence of correct steps and practices involved in creating and running a safe, secure and scalable automation programme. If designed and implemented properly, it will save you time and money, avoid mistakes and sustain confidence.

The RPA lifecycle typically covers four major areas, not that dissimilar to other business frameworks, and these are Technology, Process, Delivery and Operations.


RPA Technology

There are really only two major considerations when it comes to RPA technology: Firstly, your own IT infrastructure and environment. Second, your RPA technology.

For RPA purposes, your IT infrastructure is generally already mandated by company IT strategy and will be either on-premises, cloud or a hybrid of these. Regardless of this, your IT team will need to be involved to professionally set-up, maintain and manage architecture and applications for your RPA programme.

For RPA technology, you have a bewildering number of options to choose from in both tools and vendors. This is really a much bigger topic for another article and another day, but the main question to answer now is, do you want to be locked-in to a service provider and tool, or do you want greater flexibility and control, at least to begin with?


RPA Skills & Competencies

RPA is still a relatively new and fast-growing industry that requires a balanced blend of skills and expertise, so finding or developing those skills is your next major toll gate.

Again, keep it simple to begin with, ask yourself this one question: do you want to go it alone, or hire an expert to help?

It’s another much larger topic for another day, but a really good service provider will offer solutions to fit both options. When you’re getting started, it can and will save time and money and avoid mistakes to get advice and support from experts, even if you plan to build your own RPA team in the long run.

If you want to develop your own in-house RPA skills, then finding suitable candidates and providing them with training is essential, BUT this will depend on which RPA technology you’re going to choose. Many of the top RPA vendors provide free online training courses, but these alone may not be enough. Many RPA service providers offer a managed service based on outcomes, and so it doesn’t matter which RPA technology you use.

Remember one thing, RPA is a business-led strategic initiative and tactical technology that involves process change management, requires the support and collaboration of IT professionals and clear leadership from the business in all other areas.

The primary skills you’ll need to support your RPA programme are:

  1. Process analysis, modelling and documenting
  2. RPA design, development and testing
  3. IT infrastructure and architecture
  4. Change management
  5. Project or programme management
  6. Business process ownership


RPA Pipeline

Building an RPA pipeline can actually begin in parallel with the two above. It’s probably going to be the most fun and interactive part of the lifecycle too. This is also a critical communication phase, and a major opportunity to gather buy-in and support from your colleagues and co-workers and generate some enthusiasm about this new and exciting new initiative.

Building a pipeline simply involves creating a list of possible processes, tasks or activities that could be automated. Initially, look for activities that:

  1. Are manual, mundane, repetitive
  2. Follow logical, sequential and rule-bound steps
  3. Use data that is all digital and well structured
  4. Use applications and systems that are relatively stable and robust

You may be tempted to consider ideas that only generate a saving or benefit greater than “X”. At this early stage, there are a few reasons to not restrict yourself too much:

Firstly, you’re better off defining a clear “WHY” and overall goals for your RPA programme, which will naturally influence candidate selection, as objectives vary for different business units and departments. Second, it will take you time to properly quantify every candidate sufficiently for decision-making and not all benefits are easily quantified – for example, staff morale and customer experience. Lastly, the idea generation phase is a fantastic way to engage people and think of ways to improve their work, and you don’t want to miss out on this opportunity.

Once you’ve created a list of potential automation candidates, you can evaluate and prioritise according to their suitability, feasibility and value to the business. If you’re just getting started, it’s advisable to find your top 3 and select your pilot project. If you’re going it alone, keep it simple and basic. If you’re getting the help of experts, you can probably afford to be a little more ambitious.


RPA Governance

Your first process automation should follow a well-defined and proven methodology for the deployment to run smoothly, quickly and deliver the desired benefits.

Your “RPA Playbook” defines the best practices and governance for all phases from this point forward. If you’re just getting started, you’ll need to create one. It’s not difficult and doesn’t need to be the ultimate and final guide to everything RPA, but you’ll need to consider some essential points at a minimum:

  1. The RPA charter – who, why, what, when, how
  2. Phases of the automation lifecycle
  3. Tools and applications
  4. Roles and responsibilities
  5. Process discovery, definition and design
  6. Automation build and test
  7. Development best practices (code structure, naming conventions, exception handling, assets and files, logging and audit, selectors, security, code review)
  8. Deployment readiness and operational handover
  9. Support and maintenance


RPA Process Discovery, Definition & Design

There are three parts to understanding your process candidate before rushing into building your robot, these are:

  1. Defining the existing process As-Is
  2. Designing the future process To-Be
  3. Documenting the robotic solution architecture design

The best place to start is a meeting with the actual process performers for an end-to-end walkthrough of the activities. Remember to record a video to refer to later.

The existing and future processes are both from the business user’s perspective to confirm you’ve understood the activities fully and show what will be different in the future once automated, including error or exception handling and hand-off points. The solution design is a technical perspective and needs some understanding of developing RPA. However, if you’ve picked a simple process to start, it shouldn’t take long and will ensure a reliable disciplined approach is adopted.

Definition and design can be completed using any number of tools and techniques. The best advice is to use what you’re most familiar with for now.

Another top tip is to begin collecting test data now, in readiness for the next phase.


Build & Test your Robot

All process automations will ideally be developed following established methodologies or industry best practices relevant to the RPA technology being used. At a minimum, be sure to consider:

  1. Coding according to development standards
  2. Peer review of essential coding, connections, business and workflow logic
  3. Code commenting with development specifications and source control
  4. Error and exception handling
  5. Recovery, back-up, logging, reporting and reuse requirements

Your process automations will undergo a series of testing phases throughout their development and prior to go-live to ensure a smooth deployment and prevent critical bugs reaching production, those will ideally include:

  1. Unit testing during build to validate the operation of coding, workflows, connections and business logic
  2. Integration testing to verify end-to-end functional operation, validate business process scenarios and exceptions
  3. User testing to validate automated process outcomes, operational scenarios, processing times and user interactions

Each testing phase will ideally be defined during the design phase to establish the approach and phases, success criteria, which environments to use (production & non-production), test data, test scenarios and who the testers are.


Deployment Readiness & Go-Live

Deployment control procedures will ensure you have a safe transition to go-live and may need to be reconciled with your existing IT release control procedures.

Practices may vary, but once your users have successfully signed-off testing, the key considerations are:

  1. Deployment Readiness to agree continuity plans and check technology, people and process are all ready
  2. Decide support and maintenance
  3. Confirm go-live schedule dates, times, participants and scope
  4. Handover plan and business process user manuals
  5. Production release control


Business as Usual & Ongoing Operation

At this point the automated process will be live and running under the supervision of the business process teams. If the robot was built well, it should operate effectively with little intervention or support, but there are a few new activities that the process team will have to adopt:

  1. Daily admin of the process to check workload and queues
  2. Handling exceptions
  3. Requesting any technical support as needed
  4. Manage any possible change or optimisation requests

All that’s needed next is to repeat, repeat, repeat, working through the pipeline at your desired pace.


Lastly, here’s your checklist with a few extra tips as you embark on your own RPA journey. Good luck!



All successful journeys start with good preparation. Begin with a clear idea of the end destination and the pathway to get there.


The RPA journey must be synchronised with your organisation’s strategic direction to avoid being side-tracked by lower priority processes which wouldn’t deliver the required results.

Top-Down Leadership

Executive buy-in and support will maximise your potential for success, as they promote adoption and advocate budget.

Business & IT Allegiance

A strong, clearly defined and coordinated relationship between the business and IT is essential. The business must lead and  focus on the primary goals and the IT organisation provides enablement by delivering future-proof architecture.

Programme Structure

Introducing a robust governance framework from the start will help align expectations, avoid confusion and ensure appropriate resources are identified and made available.

In Conclusion

This article was written to guide those who are new to RPA or at the beginning their journey, so hope you find it invaluable. However, if RPA isn’t new for you and you are further along the journey by the time you read this, how many of these points did you implement in the beginning and how many did you get right? Please do get in touch and let us know, we love to hear about others experience with RPA, what worked, what didn’t and why, and we’re always happy to offer our insights if it helps.