Human resources boils down to data — the larger the number of employees, the more data. Still more employee data is kept for compliance reasons. Gone are the days sometime far back in the last century when the employee roster was just name, last name, address, job position and date of hire. Now it is Jane Doe, accounting, female, African-American, graduate of XYZ business school, last quarterly evaluation, last job satisfaction survey, flex hours request granted to take morning computer science course, etc. etc. Multiply that by hundreds or thousands, and the data fields around employees get very big.
AI is changing the way firms screen, hire and manage their talent
The data surrounding each employee allows the company to do the right thing at the right time (pay monthly salary staff, grant vacations and other leaves), to tell government and other agencies the right required information — we have X % women and Y % Hispanic and to do some planning ahead — Jack Doe turns 65 in six months, should we prepare for his retirement and if he does so, start looking for a replacement to promote or hire? HR plays a major role in all of these processes. Also, there is probably more value and insight for HR that can be extracted from all the data one has about employees.
Take Jane Doe, with an excellent five-year track with the company. What contacts does she have at XYZ business school in case the company wants to recruit there? Is she a member of a state or regional Afro-American female alumni association — another possible talent resource? And what about this interest in computer science? What if she gets her second degree and quits to join a fintech startup? Or can that new qualification be used to move her to another position inside the company. And how many other employees are developing second degrees or skills on the side?
HR faces data flows that need AI to be manageable
When dealing with huge amounts of data, HR is turning to big data analysis techniques and artificial intelligence (AI). AI is using computer software not only to gather, sort and sift data, but also to make decisions based on these calculations the way a person would make decisions, at least on a focused range of matters. The decisions can be made by including a set of rules in the software, a complex “tree”of “if X, then Y” choices, or more advanced AI can use so-called “machine learning” to modify its own actions in order to reach a desired outcome. The more data it processes, the software “rewrites” itself to eliminate choice-sequences that do not produce defined results.
Recruitment is just one example of the technological disruption that AI will bring to the workforce
The first big flow of data that HR departments can use AI to process is job applicants — the hundreds, perhaps many thousands of CVs and letters of application. Applicant tracking software can sift through electronic CVs (more than 90% come in with the e-mail, not the snailmail) and eliminate up to 75%, leaving the rest for closer study by live HR staff. A rule-based AI will look for key word and phrases in applications and CVs. If a sales opening has to be filled, it would eliminate applicants who mistakenly apply for a position in product field service. A more advanced HR system with AI features would modify the applicant screening criteria based on existing employee track records — if past applicants who were members of sports teams or interest clubs at university did better in some team-intensive jobs than “loners”, this CV item would be given greater weight by the screening software.
Chatbots — a familiar way for young applicants to talk about themselves
Screening can also be done interactively, using online chatbots. Chatbots or at least the idea of interactive chatting is familiar to young applicants who exchange SMS or Whatsapp messages in their daily lives. The chatbot can be tweaked to elicit certain information from applicants by asking about sports teams or clubs in the hypothetical example.
Regarding chatbots, Josh Bersin, founder and principal of Bersin by Deloitte Consulting writes:
“There is a wide range of new AI tools used in recruiting. For example, there are AI-based chat systems that can communicate with candidates and quickly screen people. These systems already enable candidates to select the right job or shift, and they can dramatically reduce the time recruiters spend on candidate screening. This frees up recruiters to focus their energy on assessment and selling.”
Whatever applicant screening solution one uses, machine learning can hone and improve it by feedback from further down the hiring pipeline. Face-to-face interviews or a chatbot session may reveal that some applicants may not have the specific training to start immediately in a technical job. The company can either hire and provide the additional training, politely turn them down or have machine learning adjust the screening process so that the system will scan all initial applicants and weed out those lacking the necessary specific training and skill.
Chatbots will replace humans on routine, structured, labor-intensive HR processes
Onboarding is another HR process that can benefit from AI, allowing it to customize the process for different new hires by “learning” from the previous experience with many similar new hires. Chatbots can also play an important role, as they are always available whenever a new employee has a question and they can be asked “dumb” questions. Past chatbot interactions will also “train” the system to anticipate these matters — “Sara Jones is the deputy IT systems administrator but Sarah Jones is on sales team B, they are used to new people mixing them up.”
Predictive analytics — AI that promotes employee retention and loyalty
Once hired and working, AI in HR systems can help with employee retention and promotion issues. By looking at data across the company, AI can predict which employees may get dissatisfied and pro-actively address the issue — a performance dip for lack of training, a workplace environment issue (poor lighting or noise that seems to be driving away workers from Building A) or other matters. Like the predictive analytics systems that have a machine tool repaired before it, these HR solutions will detect and remedy employee issues before they lead to someone quitting or losing productivity.
Eva Wislow, writing on the Big Data Made Simple website explains:
“As much as it is difficult to hire talented employees, it is as difficult to keep them in your team. This is why almost 60% of organizations consider employee retention their biggest problem. However, AI has the ability to analyze and predict the needs of staff members.
It can determine individual affinities and reveal who should get a raise or who might be dissatisfied with the life-work balance. Such analysis gives room to HR professionals to be proactive and solve the problem even before it actually occurs.”
Or as Bill Carmody, the founder and CEO of the US marketing company Trepoint wrote about AI in HR for Inc.com:
“When you have thousands of employees, you have a wealth of data flowing inside your company. Predictive analytics decodes and deciphers that data in order to provide the kinds of insights you need to say predict when you’re going to need to hire more people in different regions of the world.
Or perhaps you’d like to be alerted when a key strategic hire may be looking to quit his or her job. With the right machine learning, you can start to capitalize on all the historical data and the predictive nature of how individuals come in, grow and eventually leave your company.”
AI will be an increasing part of HR IT solutions, across the full spectrum of HR functions
Josh Bersin sums it up:
“Artificial intelligence in HR is likely to transform HR operations in three profound ways. First is the emergence of the conversational interface, where we can talk to systems, ask questions and interact through chat. This can be supplemented by augmented and virtual reality, which is developing even faster than we thought. Second is machine learning, where software analyzes people-related data and offers smart recommendations and decisions. The third is the growth of predictive models, which are systems that can identify patterns and quickly find areas of risk, fraud and other possible performance problems. “
Should AI be looking at student stunts and personal convictions?
While that sounds fine, the ability of sophisticated AI systems to “scrape” and use data about anyone, including employees (both yours and even those of competing companies) has a downside and a dark side. Social media posts by potential applicants can reveal their hidden talents, but also cause problems. Do photos of a nude fraternity stunt five years ago really address the person’s ability to be part of a programming team? What about political or religious views? Should we be having an “extremist” on animal rights in marketing (the company makes stationery, no animal products)? Or maybe we need a practicing Muslim for diversity and because of a growing number of customers of that religion? An employee is criticizing the company on Twitter? Is the person breaking the protocols for reporting a grievance or issue, or maybe those protocols are not working? Where to draw the line? What about compliance with new privacy standards affecting employees in European Union countries?
Do photos of a nude fraternity stunt five years ago really address the person’s ability to be part of a programming team?
HR management software incorporating AI techniques will give enterprises deeper and more extensive insight into employees than ever before — but the red lines on what data to use and how to use it will still have to be drawn by the most important HR component — the top humans in management that set policies.
Juris 👽
