For all the talk about HR as an experiential position, one that favors emotional intelligence and people skills, the job has always involved a lot of work that could be described as robotic. Though each employee is unique and requires their own approach, each also requires the completion of a number of extremely self-same tasks, from employee profiles to custom investment packages.
With modern HR machine learning solutions, however, business leaders can finally do something to change that, and greatly improve outcomes in the process.
HR leaders all over the industry are clamoring for any technology that can change the day-to-day experience of working in HR, making it more enjoyable and more effective at keeping the business running smoothly. According to the Society for Human Resource Management, managers believe that in an average week they waste 14 hours on tasks that could be automated through machine learning.
In modern parlance, machine learning solutions are often called AI because they can do things that were previously confined to human employees alone. Modern AI can automate simple, repetitive processes and apply them on an inhumanly large scale. Machine learning can let HR professionals deemphasize these tedious, often frustrating aspects of the job while also allowing them to pull a greater amount of insight from the results. This has been the major failing of “big data” for business — it’s great to have all that information, but past a certain size threshold, the human brain is no longer the ideal tool for analysis.
The reason is that in many ways human beings are too intelligent for these sorts of tasks; a human being may be able to apply a series of simple operations to employee data and accurately produce a ranking between the results, but no human worker could ever maintain their focus on such a task hundreds or even thousands of times in a row. AI products never become bored, however, and they can remember larger volumes of past results to inform future decisions. For those parts of the job that do recur over and over again, AI is unquestionably the future of business.
What machine learning can do — no coding required
What truly separates modern machine learning from previous attempts to computerize the tedious side of HR is the ability to reliably find and exploit patterns in data. That might not sound like much, but it’s the difference between being able to send a series of emails to a provided list of recipients, and being able to build that list based on its own reading of the intent of the email in question. It’s the difference between having a computer calculate the relative returns from a variety of investment packages and having it customize a new package for each employee.
AI can really show its value in the ability to monitor a business’s most constant concerns, however, those that require someone to keep a vigilant watch over some aspect of the business for long periods of time. Most directly, the IT department has been turning to AI to automate network monitoring duties for defense against omnipresent criminal hackers, but the ability to keep an obsessive watch over developing issues is a boon to any business.
For example, effective retention of top talent requires the ability to identify those workers who are at the greatest danger of leaving, and address their most likely concerns. A properly customized machine learning solution could provide HR workers with a list of most likely employee departures based on factors like time spent in their current position and direct interaction with competitor organizations. Given enough freedom, an AI can also recommend appropriate responses, whether that be promotions, non-mandatory raises or modified schedules.
From applicant screening to onboarding to hours management, modern HR machine learning solutions have the potential to truly revolutionize the industry — but potential is only meaningful if the products that are actually on the market can deliver on it.
Which HR machine learning platform should you choose?
The variety of machine learning products is becoming as vast as the variety of problems they could solve and, in its own way, just as maddeningly complex. Major players like Amazon, Microsoft and IBM have suites called Alexa, Azure and Watson, respectively, but these are still mainly meant for use with some measure of technical know-how. The abilities are there — but to get them, many organizations are finding they need to involve the IT department, which undermines much of the goal of making HR more agile.
Instead, machine learning’s real potential for HR lies in using further machine learning technologies like speech recognition to allow employees and managers alike to easily interface with advanced AI. Amazon, in particular, has been racing to incorporate the ability for employees and managers to express their wishes in so-called “natural language,” and receive the result they’re looking for in return.
This could allow employees to intuitively request time off from an AI, or HR managers to simply tell their AI assistant to compile a list of possible applicants for a position. With the ability to send communications of its own, an HR machine learning suite could even proactively gather some types of information, to help complete its report. Modern machine learning suites are capable of all of these tasks, individually; the wait will be for a business making them available through other side-technologies that unlock that potential for everyone.
The right time for adoption
All this might seem as though it implies modern managers should hold off on looking into machine learning for HR, and wait for the perfect solution to materialize — but that would be a mistake. The potential gains in productivity and employee satisfaction are already great enough to justify investment on their own, but more to the point, such a fundamental shift in processes takes time. No business is going to want to be scrambling to adapt to the most crucial new business technology of the day, while their more forward-thinking competitors thrive and race ahead.
In all, machine learning could free the HR world to focus on the people problems that made HR necessary in the first place. It might seem ironic that it takes an artificial intelligence technology to make human resources more human than ever, but with an understanding of the potential gains and just a bit of preparation, that’s what tomorrow’s managers could accomplish.
This column originally ran on the Spark blog site.