Why RPA cognitive technologies have not seen smooth journeys so far
Extracts from a journalist interview about the stop/start unpredictability of the intelligent automation market with Professor Leslie Willcocks, who with two other KCP authors has been working on the book Becoming Strategic With RPA, due to be published later this year.
Do you think that cognitive technologies are in a very early stage right now- earlier than some of the hype might suggest? And when might we see a critical mass in deployment?
Leslie Wilcocks: Yes. This is a very strange market. I have to say that I sort of thought it would be onwards and upwards very quickly. And so we wrote a book in 2016 that said, well yes, there is something very important going on here, it’s great stuff, it’s a bit of a no-brainer. Here are 25 action principles to get it running and good luck. Then I thought, well, we’re going to have to look at artificial intelligence right now, because it’s going to move fast. But then a lot of the early users ran into problems, not all of them, but 80% of them. So we then wrote a book called RPA and Risk Mitigation, the Definitive Guide which basically said hey, you’re going to encounter these risks, you have to do a lot of management, more management than you think. So that was 2017. In 2018 we returned to it thinking they would have solved all those problems, they will now be in the next phase of cognitive technologies. But they weren’t really up to speed on the RPA let alone cognitive. And the cognitive stuff hadn’t matured enough nor had the suppliers. So that book was really ambitious and way ahead of its time. So our next book, which I have been writing with John Hindle and Mary Lacity, is called Becoming Strategic With RPA to be published in October 2019. This takes us back a step. Basically it is saying, well you didn’t do RPA properly in the first place so until you do that well, you’re going to have to wait for the technology to come through. Do RPA well and then you can move into this integrated automation platform with the cognitive technologies, perhaps within the next 18 months.
But RPA has been around for several decades already. Why has it still not been perfected and integrated?
Leslie Willcocks: It’s difficulty is on the point of integration. It was very small scale automation of processes up to about 2010. Then the problem we describe in our books started to kick in - the dramatic increase in the amount of work to be done started hitting organisations. After 2008, what you get is work intensification - trying to do a lot more with a lot less people. And you run into a problem eventually. One big one is that you run out of the number of hours in a day people can work. Then you start looking around at how to solve this problem And then they started looking at the technology more seriously and it got rebranded as RPA - it used to be called many things and wasn’t particularly high profile. So between 2010-2012, a couple of organisations said, well what is this and started to discover what it could do. And progress was finally made. Some of them are now doing 35% of their back office with RPA because it just solves the problem. And those exemplars led the field for other people.By the time you get to 2015-16, a lot more people are looking at the same problem and saying, well we’ve run out of levers we can pull on this, what is this RPA stuff -other people seem to getting a lot of benefit. And then they started looking at seriously and finding that it actually is very effective if you manage it properly. Since then a lot of people have been buying into it so it is a bit of a no-brainer as a technology in my view, within the scope of what it can do. And the speed at which you can do it. But they still run into big management problems because they don’t understand that it’s not just software, it’s a sort of organisational change issue as well.
What are some of the practices that will allow organizations to more successfully adopt automation?
Leslie Willcocks: Unfortunately, it’s quite a group of things that needs to be in place. I think that a big part of it is to do with a a mind-set. It’s about becoming more strategic with RPA and the chapter we’ve written on that suggests this has several components. The first one is you see RPA as strategic in terms of three to five-year long-term investment which fits with your cognitive and your digital strategies versus operational quick wins. It’s about recognising that it’s culturally embedded - we’re going to automate, we’re going digital, we’re going to innovate rather than see it as IT as usual. It’s all about planning for the long-term rather than just using it for opportunistic quick wins.
It’s about seeing it as a programme which needs governance over a long period of time rather than just a single project delivery. It’s about seeing the RPA as a platform rather than an individual tool. It’s about change management rather than what we’ve called silo tolerance. A typical organisation could be what we’ve called the seven siloed organisation. Siloed in data, technology, structure, mindset, processes, skillsets, and culture. And automation doesn’t like siloes. It needs to go across organisations.
So it’s about changing those siloes, optimising across them. But the last area which no one’s really gone into, which we’re just doing a lot of work on is having systems in place that measure what you’re trying to achieve, and moving away from traditional return on investment and total cost of ownership kinds of measurements. We recommend it’s better to measure the three Es as we call it, efficiency, effectiveness and enablement. A measurement of how much your RPA platform enables other kinds of technologies, but also other kinds of processes and products and services. So it’s the mindset of having these key attributes and working on them that shifts into getting superior business value into the leading companies, the top 20% that are deploying RPA at the moment.
Leslie Willcocks and John Hindle are co-authors with Mary Lacity of Becoming Strategic With Robotic Process Automation to be published in October 2019 – see www.sbpublishing.org