People Analytics – Are we Getting it Wrong?

The function of human resources has changed dramatically due to the introduction of technology and big data designed to make the role of hiring, retaining and motivating staff easier and more consistent. Nowadays, the sophisticated analyses of big data are helping companies identify, recruit, and reward the best personnel. However, when it comes to human resources, it cannot only come down to big data for the best results. People Analytics is what is making the real difference.

It’s not only in human resources that big data analytics has been making its mark. Retail businesses have been using these tools to identify customer’s purchasing behaviour, which would in turn assist with marketing plans and development of product lines. Almost every other industry, including manufacturing, agriculture and health care have also turned to these big data strategies for better processes. 

So it wasn’t too long before human resources directors decided it would be worth determining whether or not they could use these predictive models to identify, recruit, train and motivate the right people for their team. Historically, human resources was very much based on factors with little substantial backing. These included intuition, how well someone presented themselves in an interview and how persuasive they could be. It could also come down to referrals and word of mouth, which did not always result in the best employees being hired for the job, rewarded or trained.

However, when it comes to people, big data can be quite challenging. People analytics, on the other hand, is credited with bringing back human data to the decision-making process, as opposed to solely relying on machine data for making business decisions in simple terms. People analytics is the analysis of data that relates to people within an organisation or industry. This people-related data is used to optimise the decision-making process and business outcomes.

Still, getting people analytics right is tricky as well. Humans cannot be programmed by machines. There are no standardised metrics and methods when it comes to processing data from people. Analysing data from people should take into account qualitative and quantitative variables. On the other hand, reverting to the more traditional methods for gathering data on people is not practical either, especially for large organisations. Collecting the data, analysing it and then deciphering it is a mammoth task. Surveys, interviews and the like can cost a great deal of time and money, and again, may not be consistent or accurate either.

When collecting data, it usually arrives in two different formats – structured and unstructured. Structured data is information, usually text files, displayed in titled columns and rows which can easily be ordered and processed by data mining tools. Structured data is readily searchable by simple, straightforward search engine algorithms or other search operations; whereas unstructured data is essentially the opposite. Most organisations are likely to be familiar with structured data and already using it effectively. On the other hand, unstructured data, usually binary data that is proprietary, is that which has no identifiable internal structure. Unstructured data is a massive unorganised conglomerate of various objects that are worthless until identified and stored in an organised fashion. Unstructured data can include email, PDF files, video, spreadsheets and social media posts among others.

It is important for human resources departments not to become confused between structured and unstructured data and make the wrong decisions. Even HR analytics that seem to make sense may send the wrong message and result in misplaced decisions. For example, analytics may show HR that employees who have a spouse and family are more likely to leave a position that involves frequent travel, than those who are single. But it would be a bad decision for a company to pass up on top talent just because they are married, even if they are aware of travel obligations for work. Or alternatively, if data points toward women in their 30s taking maternity leave, assumptions can’t be made that every new female employee in their 30s would like to (or indeed can have) children.

Using specific data in this way is legal, but is it ethical? Can people be fairly evaluated on actions they haven’t yet done, but data predicts they may do? Will using data in this way lead to the best decisions for the business?  When collecting data, an important message is always to be clear with employees about why the data is being collected and what it’s being used for.

When analysing data, always try to look at the big picture, rather than zeroing in on unstructured data or other details that may not seem relevant in the human scheme of things - such as the fact that one employee took more days off than another, for example. Always focus on making the best decisions for your employees and the business as whole.

The term human resources may in fact be misplaced. Humans are not “resources”, we are people. Data cannot determine everything about a person based on how others may have acted in similar situations. The best business decisions are made with both viewpoints in mind.

About the Author

Liam McNeill is a Managing Partner at Presence of IT. He is proud to introduce new clients to their progressive range HCM services and software solutions. Presence of IT is Australia’s leading consultants for the world’s foremost HR/Payroll and Workforce Management solutions. They are experienced in partnering with organisations to provide end to end service in HR processes; from software selection advice, all the way through to ongoing support after solution deployment.