Now that companies have finally settled their core systems into the cloud, HR leaders need to get ready for a deluge of innovation.
The agility of the cloud means technology teams can deliver new features and interactions quickly and seamlessly. Cloud-based HR systems also mean vendors can implement new iterations faster and with a lot less hassle.
That is good news for clients, said Dan Staley, principal HR technology leader for PwC in Atlanta. “Vendors used to roll out upgrades every one to two years, now they are coming out quarterly.” That adds value for users, who get access to the latest features as soon as they are ready, and allows vendors to increase the functionality of their products.
This is allowing them to speed road map timelines, and making it easier for larger vendors to acquire best-of-breed smaller firms and integrate them into their suite of tools. “We expect to see vendors taking their products’ capabilities further, faster,” he predicted. That includes embedding more social and collaboration capabilities and adding new reports and dashboards. It will also allow them to integrate data from multiple sources, to support workforce analytics — which is where the real business value will be generated.
HR management systems vendors have been promising predictive analytics for a long time, without much significant progress. Though that could soon change, said Christa Manning, vice president of Solution Provider Research at Bersin, Deloitte Consulting LLP. “Most platforms are experimenting with machine learning to derive meaningful insights from the masses of employee data they have.”
A Big Year for Big Data
While true predictive analytics for workforce management is still something of a pipe dream, several vendors, including Workday, Visier, Vista, IBM Watson and SAP Successfactors now offer some data analytics capabilities. These tools promise to provide a range of insights into things like whether companies are meeting diversity goals, where they face turnover risks, and training advice for career development.
Many of them are taking advantage of the vast databases stored in the public cloud to hone these systems. The public cloud holds masses of workforce data, which is critical for creating useful algorithms, which in turn are a set of rules the computer uses to analyze the data. “Algorithms need to be trained on large data sets to understand what information is relevant,” Manning pointed out. “They learn from every exchange and get better over time.”
As these algorithms are able to tap more data sets they will be able to offer more targeted insights, Staley predicted. For example, imagine a single system that can review employees’ overtime log sheets, travel spending and their LinkedIn behavior to determine which overworked employees are most likely to quit — then offer HR advice on what they can do to get them to stay. “There are a lot of possibilities for using predictive analytics for making sure your best talent doesn’t leave,” he said.
Analytics tools in the HRMS will also play a role in managing gig workers, according to Cristina Goldt, vice president of HCM products for Workday in Pleasanton, California. Being able to review data regarding all types of workers and projects in a central location will help companies better analyze where and when to hire contractors versus full-timers, who to choose and what to pay them. “They can match skills to different roles, and make their hiring systems more efficient,” she said.
Some vendors, including Workday, are also offering customers the ability to compare their data insights to industry standards to see where they stand. “It makes it possible to benchmark themselves against their peers,” Goldt said.
Are We There Yet?
All of these scenarios are enticing, though the days when business leaders can predict workforce trends through a cursory glance at an analytics dashboard are still well into the future. Unlike other software that is rolled out and ready to use, machine learning takes time and training, and requires access to linked databases with relevant data, Goldt said. “It’s called machine learning for a reason.”
Customers are also still somewhat uncertain about how they will apply analytics in their own organizations. This is partly due to the lack of meaningful case studies, Manning said. “Every vendor is talking about machine learning for HR, but there aren’t a lot of examples yet.”
For companies hearing pitches from their vendors about the magic of workforce analytics, she urged them to “demand live customer references” and real world examples that prove what other companies are doing, how they did it and what results they saw. “Training algorithms requires strong partnerships with vendors who understand the technology as well as how it can deliver actionable information,” she said. This transformation will take time so choosing a vendor you can trust is important.