RPA : The Robots are coming

03 April 2017

Robotic Process Automation, or RPA, has been bubbling around for a few years now. I learnt about it because a friend of mine decided to start a business specialising in RPA a few years ago, and I’ve been following it ever since. I think it has developed sufficiently now to start really taking notice.

I attended a fantastic analyst briefing last week by Everest Group on the shape of the RPA market. It made the concept very easy to understand and I would recommend making the time to watch the playback if you can (email me for a link). If not the summary is below.

RPA is effectively using software to emulate human interactions with a process or programme. It tends to only deal with Structured Data, and delivers a deterministic solution to a problem – i.e. currently there isn’t a great deal of intelligence in the handling of the process. It's designed to relieve users of high volume monotonous processing functions.

Everest described the evolution of RPA in 4 phases:

1. Assisted RPA – effectively running on a user’s desktop to complement the activities of a manual user when they can be automated. An example of this could be a simple cut and paste of information from one screen to another. This level of RPA has been around for a fair while.

2. Unassisted RPA – this is more interesting because it effectively means the “robot” performs a business process on its own, or unassisted. Instead of complementing human interaction, it replaces it with the business process. Users still need to schedule and manage the robot workloads, but it delivers significant efficiency over human processing.

3. Autonomous RPA – this builds on Unassisted RPA by giving the control and scheduling of robot processes to another robot. The process can be scaled up and down in line robotically in line with demand, giving an answer to dynamic / peaky workloads. This is the current “latest” version of RPA.

4. Cognitive RPA – this is where the robots can think. Natural Language Processing, unstructured data, decision making are all performed by the robot such that more complex tasks can be automated. This is coming (“the robots are coming”).

The most interesting factor is the growth rate of this technology. The global market is estimated to be about $217m in 2016, growing to $600m+ in 2018. Whereas this is a relatively small market value, it saw 70% growth rates in 2015/16 and clearly intends to continue. The UK market is 21% of the global value, with about 45% of the revenues being spent in the Financial Services sector.

One could object morally to robots and automation replacing human jobs with robotic processes. This is a valid concern but realistically, a good majority of repetitive dull work currently handled by workers has either been offshored or is being performed by people who probably would rather be doing something else. Interestingly, Everest made the point that in the RPA deployments to date very few people end up laying off workers, but give them more interesting / higher value roles to accelerate business processes. Relying on RPA reduces staff attrition by removing tedious activities.

With the stated efficiency gains being between 10-35% and payback within 12 months, people need to take notice. Technologies such as Automation Anywhere, Blue Prism, UiPath, Nice are in some cases offering free trials to entice clients into this phenomenon. Based on the payback figures, it must be worth at least having a play!

My next predicted issue is lack of skills in the technology sector. If you look at the partner landscapes for some of the main RPA software houses, the main contenders are the big boys (Accenture, KPMG, Deloitte) who will generally only play with major corporates, and charge appropriate premiums for their brands. There will be very few midmarket channel players who develop the skills in both technology and business process re-engineering to implement this properly. My entrepreneur friend clearly had an epiphany and I know is doing very well out of this relatively new technical trend. I can imagine more companies like his will be born in the coming months and years.

If you are a client with significant processing volumes of repetitive tasks, I’d be happy to talk about whether this is something worth pursuing. To do it well, you need to start with simple tasks and grow out from there. You need to be willing to adapt processes to suit the available technology. If you are looking for up to 35% efficiencies in costly business processes, I’m sure I can help you find the right partner.

We use cookies and other technologies to allow us to remember you, improve our service and display relevant ads to you. To accept cookies, continue browsing, or view our Cookie Policy to find out more, including how you may withdraw your consent.