People Analytics will outperform HR professionals very soon! Come on, as an HR professional, you cannot deny this any longer… this unstoppable machine is impacting your work to such a degree, that it is showing all of the signs of rendering the HR profession a quaint historic memory. So, are algorithms, Machine Learning and Artificial Intelligence writing HR’s obituary? Threat or not, avoidance itself is a life-shortening choice to make.
Article by: Jouko van Aggelen, Executive Director, Head of People Analytics – Cubiks
These new cutting-edge algorithms find the right person for the job and are diminishing the added value that a Talent Acquisition specialist brings. Retention and employee engagement are not researched via exit interviews or annual engagement surveys anymore, but in real-time by analysing email behaviour. Questions from applicants or employees are handled by chat-bots powered by Artificial Intelligence. Human input and guidance for talent reviews is often not needed anymore; the algorithms have already selected the highest-performing candidates and filled-in your 9-box. So, are algorithms, Machine Learning and Artificial Intelligence a threat or a blessing for our profession? And even if you’re sceptical, could you actually stop their use? So, if you haven’t started leveraging them yet, how and where should you start?
“Will AI take over completely, leaving us in the dust? Not entirely, but many aspects will certainly be automated, and this is a positive. Beyond doubt, HR professionals that choose to implement People Analytics will significantly outperform those who don’t”
It is firstly important to define the foundations of this article’s discussion. People Analytics – or more precisely – using data in an HR context, can refer to two different aspects: 1) Automating operational HR tasks and processes, or 2) Analysing People Data to create more and new so-called actionable insights. Automation creates smoother, efficient and more user-friendly processes. Chat-bots can answer applicant questions 24/7 using multiple languages. And as we know, many tasks undertaken by HR professionals are repetitive and operational by nature – and in a way, ideal for automation. It’s clear that A.I. and People Analytics methods will soon outperform us in this respect; in reality, this is already happening. Research from Oxford University indicates that administrative HR jobs have a 90% chance of being automated by 2035. It has happened in many other disciplines – think call centres, customer services, etc. – so why should HR be any different? Should teams restructure in response? Will those who fail to adapt fall behind?
Analysing People Data creates more insights. These insights could in turn lead to automation of more complex people decisions – which gives more food for thought. Historically, predicting human behaviour at the workplace involved applying general insights or theories about what made people effective in their roles, what constituted high potential (or not), and what created high performing teams with an engaged workforce. These theories, underpinned by insights from the psychology surrounding people’s key motivators, inhibitors, or derailers, are still applied, but are not always proven to be valid. In today’s world, (people) data now adds a new element to the equation, one which is more objective and scientifically grounded. So, is People Analytics HR’s panacea? Data is supposed to make HR practices and people decisions more objective, fact-based, less biased and ultimately fairer. If we apply best practices and academic insights, we all know that that everyone carries their own biases and that truly unbiased human decision making is non-existent. Therefore, contemplate what A.I. would mean for our ambition to create a more diverse workforce? Or how it could help eliminate unconscious biases that exist across recruitment tunnels and promotion schemes. Using data enables us to finally give equal opportunities to all, and to more effectively source the right candidates to fill our critical positions. Working in a data-driven way improves the quality of our people decisions, but it will also do more; it brings us additional insights, which enable us to look ahead, to predict and to anticipate the future. By adding more data and more sophisticated data science methods into the equation, we can evaluate full scenarios and be confident that we understand the impact of our actions and decisions. In a way, applying Machine Learning and A.I. is nothing more – and nothing less – than closing the feedback loop, continuously monitoring and measuring the outcome of our people decisions, learning from these, and continuously improving the quality of our predictions.
So, will data outperform and replace theory, experience and human judgement? Can HR be outdone by smart machines? As briefly discussed in the beginning of this article, it should be anticipated that lots of HR work can and will be automated. However, this will not be a bad transformation; automation will help us to become more effective, more efficient, and more objective. There are great opportunities that People Analytics bring to the automation of many administrative and repetitive operational HR activities. Using more advanced technologies, such as Machine Learning and AI, will increase speed and efficiency even further. Nevertheless, the real added value of People Analytics is its tremendous amount of new insights, knowledge and learnings that will help tackle your organisation’s people challenges, and above all, will empower a workforce to fully realise its potential. Having a more in-depth and objective understanding of people’s key drivers, on what will improve performance and productivity, helps us to take better decisions on our employees’ futures. It is, or should be, HR’s responsibility to enable their employees (and maybe even candidates) to make full use of these insights, to truly empower them to use them for their benefit and to realise their potential.
Good and effective People Analytics teams must have the highest moral standards and must always put people first. It is human nature for people to reject an impenetrable ‘black box’ making decisions about them. To win trust, it is essential that we understand and explain the algorithms that power these decisions. It is key for People Analytics to be a transparent ‘white box’, providing the ability to look inside and modify the algorithms (even partially). HR professionals must feel like they have a certain level of control, being able to look inside the ‘white box’ otherwise People Analytics will feel intimidating and untrustworthy. But of course, security is a top priority. Considering the complexity of people data and People Analytics, it must still embrace the high standards of GDPR which give the user important control over their information. Privacy, data security, data ownership, ethics, and the reliability or integrity of algorithms must be maximally-assured to guarantee the success of using this kind of data to enhance what we do.
Can People Analytics Derail? It’s not undisputed that data science in itself, can already take over the more complicated decision-making process. Currently, though, we see too many pitfalls or flaws to be confidently in-favour of fully automated people decisions. These algorithms can – and sometimes do – inherit the developer’s biases. Any algorithm created by a human ultimately carries that person’s individual differences and biases. In the news, there is huge publicity surrounding Google’s biased search algorithms and Facebook’s biased news feeds. Although People Analytics projects may not currently be immune to the influence of human biases, with awareness and conscious effort in the future, it will eliminate these biases in our processes, enabling best practice and forging better, more-informed, objective decisions.
People Analytics is the vehicle which will drive our transition into 21st Century HR practice. It’s a powerful instrument which enhances the complex aspects of our roles – analysis and decision-making. It allows data to change our minds, it enables meaningful changes in processes that we may not have considered, and it identifies our human biases and assumptions. It integrates science and objectivity into our human decision making and will soon make processes immeasurably more efficient. However, it is crucial to communicate that People Analytics does not aim to automate the human, but to humanise the data – an enhancement, not a replacement, of key HR processes. The outlook is positive; indeed Marlina Kinnersley, Co-founder and CEO of AI-based Fortay.co. is quoted as saying: “AI will not displace HR; however, HR that leverages AI will definitely displace HR that doesn’t”. Recent insights from McKinsey also support this view, emphasising that human work enhanced by machines is the ideal dynamic for the future of effective human-machine collaboration.
People Analytics is an increasingly crucial aspect of our HR decision-making; it fundamentally uses data to force us to be more open-minded, to look outside of our gut-feeling and to make decisions with insights miles beyond our current knowledge and experience. It augments the most crucial aspect of our roles, but its future will be determined by how much it is trusted, and a huge part of this is ensuring it is accessible, understood and has impenetrable security. In an era where technology is omnipresent, an organisation’s people (and their tools) will be the competitive differentiators. So will AI take over completely, leaving us in the dust? Not entirely, but many aspects will certainly be automated, and this is a positive. Beyond doubt, HR professionals that choose to implement People Analytics will significantly outperform those who don’t.
