Split HR function in two to master technology, says data strategist

Bernard Marr tells AI conference current working models unintentionally sideline HR professionals with analytics skills

HR needs to be viewed as two separate functions if it wants to truly master analytics and data, as this requires a skillset the profession inherently lacks, an influential thinker has said.

Speaking at the Cognition X conference in London earlier this week, Bernard Marr, author of Data Strategy and keynote speaker, warned that “people people” were sometimes in want of the mathematical know-how needed to properly take advantage of HR analytics. 

By splitting the function in two, Marr argued, those with stronger people support skills could concentrate on creating a positive employee experience, while those who were more analytical could focus on capturing data and insight. “The two functions don’t always blend well,” he said.

The author is far from the only voice to sound the alarm recently over the use of people analytics. Also this week, a report from the CIPD and software provider Workday warned that only half (52 per cent) of organisations took people data into account when making business decisions, while nearly two-fifths (39 per cent) had no access to people data when considering business issues.

A report published last month by NGA Human Resources drew similar conclusions, finding that many HR professionals did not collect data on even the most straightforward of metrics. Of those surveyed, 52 per cent said they did not collect data on retention, 50 per cent did not track career progression at their organisation and 43 per cent did not gather figures on diversity and inclusion. 

Marr is also not the first to advocate splitting up HR. In a controversial 2014 article for the Harvard Business Review, author and leadership adviser Ram Charan wrote that the function should be divided into administration – which would focus on remuneration and report to the finance chief – and leadership and organisation, which would concentrate on people capabilities and report to the CEO. The idea provoked huge debate, and proved deeply unpopular among many HR commentators.

Marr also warned that HR was currently failing to appreciate the true workplace potential of artificial intelligence (AI) and was “just jumping on the bandwagon”, rather than taking time to assess the strategic points and identify genuine organisational issues that AI could solve.

Elsewhere at the AI industry conference, Dr Karen Croxson, head of research and deputy chief economist at the Financial Conduct Authority (FCA), outlined some of the risks posed by AI and the opportunities it offered financial markets and the broader economy.

Algorithmic technologies, said Croxson, were already being deployed in everything from credit scoring and risk management to compliance systems. The FCA, which regulates financial services firms and lenders, has a significant task on its hands to maintain stability given the penetration of machines into markets, she said.

But Croxson said she remained optimistic that technology would not mean significant displacement of jobs: "Fundamentally, I'm optimistic – [AI is] an incredible tool for us, which we can harness to flourish. We can liberate ourselves from many of the routine tasks that are painful for humans to do and which we haven't enjoyed. There is an opportunity to spend more time to focus on what it really means to be human. 

"At the same time, we haven't seen persistent drops in employment [through AI deployment] but there could be short-term drops and dislocations that could be very painful for the people involved. We have to be thinking about those people. In some cases, that might mean retraining or perhaps increased social security to help them transition to a different way of living."

Delegates also heard about the latest developments in AI tools for the HR market, including chatbot coaches for managers and leaders that were being independently verified for their efficacy, and new automated performance management systems that collated feedback from multiple colleagues, clients and channels to give up-to-the-minute data on individual employees.