On Service Automation and The Future of Work To what extent are service automation tools driving improvements in the businesses and how they operate? We have been studying robotic process automation (RPA) through 2015-16 and see this as having massive take up in 2017/18. RPA involves configuring software robots to carry out standardized, routine tasks using structured data. The software is now mature enough to be easily and cheaply adopted. It needs no great special technical skill to configure, run and manage. RPA can scale to become a virtual workforce. One company we studied runs 35% of its back office processes with RPA. We have also been finding multiple business benefits including significant cost savings, much faster processing (in one example in the London insurance market, what previously took 2 days to process needed 30 minutes using RPA); higher quality, lower error rates, higher regulatory compliance, more satisfied customers and employees. Our studies suggest that more advanced forms of service automation through software moving into more cognitive non-routine work are less advanced than the hype suggest and will be mostly be small scale, discrete projects within businesses until the back end of 2018. But physical robots, into which a lot of big companies have invested heavily, are seeing some very interesting developments, especially in Japan, Germany, USA and China. How is this impacting the workforce itself? Does it mean job losses, or is it the case that talent can be redeployed to other roles or have more time to focus on work playing to human strengths? There are optimists and pessimists on this. The most alarming studies (e.g. Frey and Osborne, 2014) suggest that on a 10-20 year horizon, 30-35% of current UK occupations are under high threat from automation. The converse point is that 65% are not, and 25% of UK occupations have a creativity component too high to be automated. But studies that look at work activities as a better unit of analysis than whole jobs suggest job restructuring will be the more normal pattern. Mckinsey in 2017 estimated that only 5% of jobs could be completely displaced by automation tools currently available. Certainly our own organizational-level research suggests every person’s job is likely to be changed by at least 25% on a 5-10 year time horizon, as technology increasingly permeates task performance. However, all too few studies focus on job creation from new technology, though this has invariably happened in the past. The pattern has been that process innovation enabled by technology has seen jobs lost, while product and business innovation has seen jobs gained. The studies we have done with corporations seeking to digitize suggest that for every 20 jobs lost from the combinatorial impact of new technologies, 13 will be gained. Our own research finds that all studies have a black hole in their analysis when it comes to not allowing for two obvious developments with massive implications for increases in the amount of work. The first is the exponential data explosion. On one estimate 90% of data that has existed was created in the last two years. That trend can only increase and will create a lot of work as well as need much heavier technology deployment. The second is the huge growth in audit, regulation and bureaucracy. When you factor in how far new technologies can create problems as well as solutions - think cybersecurity for example - then a huge, if under-analysed, work creation scheme may be underway. Two other dimensions are worth mentioning. Ageing populations in the G19 (plus Nigeria) suggest significant global shortfalls in labour and skills over the next thirty years. Second, major economies are going to experience large productivity shortfalls even to attain their present projected economic growth targets over the next twenty years. Automation and its productivity contribution may turn out to be a coping, rather than a massively displacing phenomenon. How can technology be used positively to enhance the workforce and its role in the business? Is there an ideal mix of the two? By adopting a strategy that sees technology augmenting, complementing and amplifying human skills rather than being seen as a replacement technology. The ideal mix depends on the job level and type. Technology will need skilled technical people to work on current technologies and make them function, but also technologists focused on designing tomorrow’s technologies. Automation technologies will enable jobs to be assembled that play to the strengths of humans supported by machines. These may be at higher levels – big picture analysis and judgmental work – or involve knowledge specialization, or may well involve doing tasks requiring a combination of skills that really only humans have. In our current studies of robotic process automation we are finding software robots across sectors taking over the processing of structured, standardized routine data, enabling the workers to take on all these kinds of different roles, and also experiencing more satisfying and challenging work. We keep hearing and seeing examples where human capability is being eroded by automation but human capabilities like empathy, creativity, intuition, judgement, tacit knowing and social interaction are not all that easy to replicate in specific contexts. Humans also have a facility to combine any or all of these in ways that machines are unlikely to master. In studying service automation we found these skills were frequently vital – in health care, insurance, utilities, service providers, and legal services, as just some examples. Should governments and businesses be pushing for long term adjustment measures on automation? Businesses should certainly be much more willing to invest in their workforces and in the skills needed to work in the coming digital businesses that most organizations will increasingly become. Governments are alarmingly behind the curve on the societal impacts and work implications. As an insurance policy, therefore, today’s potential employees needs to identify the relevant skills bases and make their own investments, if they are to ensure they are always employable. There will be work. One major issue is the speed of technology deployment. Our research suggests that even major organizations struggle to deploy new technologies quickly, and that the technology is rarely perfectible and hardly plug and play, especially when you move into cognitive automation. The second major issue is the readiness of the workforce. The UK’s past record is not good on this, but here is an opportunity to get it right this time. Leslie Willcocks is Professor of Technology Work and Globalisation at the London School of Economics and Political Science. He is co-author of Service Automation, Robots and The Future of Work and Robotic Process Automation. Available from April 2017 is Robotic Process Automation and Risk Mitigation: The Definitive Guide. Both are available from www.sbpublishing.org.