HR analytics: the human factor

We know how important numbers are in identifying trends and patterns. The human resources world is starting to embrace the idea of artificial intelligence (AI), and how it can help talent acquisition and retention rather than hinder, and HR specialists are increasingly becoming technology specialists. Data-driven insights are powerful, helping employers better know their staff and customers, but also assisting staff to navigate their own needs and benefits.

However, when it comes to analysing a workforce, the qualitative results can be just as important as the quantitative ones. While numbers can be used to identify patterns and trends, sometimes it’s the story told by non-numerical workforce data that can help businesses understand and improve the working experience in to reap the benefits associated with happy and engaged employees. In fact, recent CBI research suggests the UK could add £110 billion to the economy by improving people management practices within firms.

Crucially, any algorithm is only as good as the data it works with

Razvan Creanga, co-founder and chief executive of tech recruitment platform hackajob, which aims to eliminate bias in the recruitment process, says: “Analytics tools can give huge insight into an employee’s performance, productivity and happiness at work. While these tools are usually associated with workforce and performance management, how about in the recruitment process where it all begins?

“For a happy, productive workplace, the hiring and onboarding of a new employee should perhaps be just as analytical as the processes used by HR departments to measure that person’s future success.”

Mr Creanga believes that too many companies assess potential employees for roles through a traditional CV, a face-to-face meeting and sometimes a task. “But the standardisation and rigour of this assessment process is rarely appropriate to give a reliable benchmark, particularly when placing employees in more technical roles, which requires specialist knowledge,” he says, adding this is an increasing problem especially in the tech industry.

Recruitment is a holistic process, often with many stages, with many HR directors and managers on a mission to eliminate bias. “In support of this, the first stage of recruiting should be for HR leaders to gather robust performance data from skills-testing to see if someone has the right skillset or potential for a role before they progress,” says Mr Creanga.

Standardisation can improve the initial recruitment process and a solid analytical basis and benchmark from which to monitor and support an employee’s professional development has proven to be crucial. Charles Hipps, founder and chief executive of Oleeo, says: “Crucially, any algorithm is only as good as the data it works with. It is important to be self-critical when testing big data to ensure you do not inherit the prejudices of prior decision-makers or reflect the widespread biases that persist in society at large.

“Algorithmic systems that turn data into information are not infallible on their own; they rely on imperfect inputs, logic, probability and people who design them. This inevitably means that predictors of success might become barriers to entry. Without careful attention, innovations can easily hardwire discrimination and reinforce bias.”

So, how can a recruiter go about making big data a viable alternative and fairer way of accelerating a recruitment programme?

The key, says Mr Hipps, is for HR and data or information officers to work together to devise a predictive system that works best for the organisation. “This is all the more essential in the age of GDPR [General Data Protection Regulation], which grants applicants rights related to automated decision-making,” he says.

Younger generations are looking for more trust, more training and more autonomy

“For decisions that are made about a candidate and which are based on their personal data, the candidate must be able to ask for and obtain human intervention, for example to review an automated decision, express their point of view and obtain an explanation for any decision made using their personal data.”

Colin Dulson, managing director of HR and leadership consultancy Berrison, says: ”Data currently does not, and cannot, tell the full story. If we rely on it too much, we lose our inclination to debate and discuss. Increasingly flexible working environments add pressure and reduce the opportunities for ongoing engagement, so instead of using qualitative, organic feedback, businesses tend to use data as a speedy safety net. The ability to reflect is left out of any appraisal or decision-making process, career paths are mapped out by numbers and thinkspace is killed.

“Quality data is, of course, needed as part of the process to arrive at clear decisions, but a decision, by its very nature, comes from options. The importance of the decision and discussion in a working environment is not just the outcome, it is the process that is taken to reach the decision.”

Thomas Davies, founder and chief executive of Temporall, and former global partnerships lead at Google, believes HR needs to be measuring more human traits. “Companies can use tech to gather more complex data about every element of company culture, which relies heavily on human sentiment,” he says. “Previously, assets such as behaviours, leadership qualities and overall human sentiment were intangible, and therefore not recognised as being able to create strategic value.”

It’s now more possible than ever to measure how employees feel about the business, their colleagues and themselves: “How do your employees affect the rest of the business? Do they motivate or demotivate the rest of the workforce? Now, with the support of AI, these assets are tangible, allowing them to be brought to life to help predict future trends in the business. By being able to possess this information, business leaders can discover the value of their employees like never before,” says Mr Davies. “We need to capitalise on the possibilities of human sentiment analysis.”

We are entering a period when previously intangible assets are being highly valued. “Factors that were not recognised as creating strategic value in the past are now at the forefront of business leaders’ agendas,” he adds.

While nobody would argue that data enables us to make better educated decisions, Mr Dulson concludes by asking: “Does this actually provide the right format for appraisals and HR? Younger generations are looking for more trust, more training and more autonomy.” The fact is that the simple process of engaging with employees and really listening to what they have to say is tried and tested. It not only engenders trust between the employer and employee but it forms a fantastic basis from which to make better, more informed decisions.

Find out more by downloading the ‘Power of People Analytics’ report at perceptyx.com now.