The power of people analytics in crafting a personalised workplace

4 minute read

Systems made for command-and-control leadership don’t work in a Gen Z world, argues chief scientist and author Jim Webber. He examines how people analytics, and in particular graph-based models, are key to personalising and enhancing the employee experience in the modern workplace

Graph based model graphic designed on ChatGPT

The landscape of work has undergone significant transformations since 2019. Rapid innovation is forcing us to reskill, and faster than we would have foreseen even a few years ago. Basic work demographics have also changed radically in the wake of the pandemic. These changes present challenges that we have yet to fully address with the appropriate tools and strategies. 

One valuable tool to help manage the new heterogeneity in the way we work now is graphs. Graph databases, unlike traditional business databases, are specifically designed to store and query data that is inherently connected. In the context of people analytics, a graph database can be employed to represent various relationships and connections within a workforce.

Graph-based models can therefore analyse large volumes of interconnected data and uncover valuable connections among employees and their skills, ambitions and career paths. 

I believe graphs will play a crucial role in enabling HR leaders to develop highly effective and responsive systems that surpass the capabilities of today’s tools. After all, businesses rely on clear structures to manage tasks with well-understood reporting processes, which HR supports by setting up an organisation chart and recruiting into clearly defined roles. But beyond these formal structures, individual employees contribute skills and work with colleagues across functions over time. 

The emerging field of people analytics aims to provide HR professionals with a comprehensive understanding of the intricate relationships and roles within the workforce. However, as of now it has not fully delivered on that promise because its underlying data models belong to a previous era.

The good news is that by adopting graph-based systems we have the potential to fulfil people analytics’ promise.

Encoding of rich employee data: DXC and NASA examples 

IT services and consulting firm DXC is a notable example of the transformative potential of graph-powered people analytics. DXC works in a highly competitive marketplace. In order to excel the firm has to attract, retain and upskill its team so that it can provide top-tier consultancy services and delivery expertise to clients. In response to this challenge, DXC implemented a cutting-edge system empowered by graph technology where employees can easily get recommended career development advice based on the roles and job pathways that would be most valuable to the organisation. 

The tool makes recommendations based on skill progressions of similar employees, eg, it recommends skills for a software engineer based on the gaps between their skill sets and those of similar software engineers – all based on the way the graph now models all the structural information about the employees and their skills. The graph algorithms also identify similarly skilled employees, which can be used as a prototype for others’ career progression.

Another example is NASA, which is building new skills, programmes and projects to achieve its Artemis and beyond aspirations (its mission to land the first woman and first person of colour on the moon). Using a graph database as the engine of a new people analytics skills tool, the agency has captured core and adjacent skills, cross-functional skills, training certifications, educational credentials and career path information; it also records where skills are located geographically and in which teams and projects. 

There’s no reason other organisations couldn’t exploit the same idea and use graphs to manage projects and have a super-detailed map of all the skills in the team. A CHRO or project lead could keep up a dialogue with such a system to see what they need today but also run a machine learning analysis on this dataset to see what they might need tomorrow.

That will be both for the project pipeline and to keep experts engaged and constantly learning or on call, which helps everyone, as happy people are more productive people. A huge additional benefit is that I’ve talked to other customers who use their graph-based people analytics database to identify team bonds – keeping links in their graphs about who likes to work with whom, so they maintain social cohesion and collaborations that work. Again, happy employees work harder, and so you get happier clients; if you can connect up your staff in the right way, you'll get better outcomes from them and so, ultimately, for the bottom line. 

A tech for tomorrow’s world?

Not all leaders see value in this approach just yet. Perhaps they’re stuck in a ‘command and control’ world where people are still cogs in a machine.

There’s no shortage of older-style HR software that can operationalise such traditional thinking. However, this approach to human capital management doesn’t have much longer to run, especially in a jobs market where Gen Z will constitute 30% of the US civilian labour force by 2030.  Increasingly, smart and talented people will have options, and if they don’t like what you have to offer, they will exercise them.

To provide fulfilling and flexible employment graphs have an excellent data structure for turning us all from line items in a spreadsheet to a data representation that, by default, records our social graphs, skills and contextual work history within the organisation. 

This approach forms the basis for much easier automation of our emerging work patterns like remote or hybrid arrangements. It also enables better (e.g. data) informed promotion pathways and ways to identify when you as an organisation may need to invest in cross-skilling and upskilling. It could also identify groups of people who are underperforming or overperforming, or whose skills need transitioning. Aggregating up, it can even provide predictive analytics, powered by graph-native AI, about which projects have a higher likelihood of failure based on their staffing profiles.

A new dimension to people analytics

CHROs would end up with so many more degrees of freedom to think about people management using a graph structure for more personalised and sensitive people analytics versus using a limited old-fashioned database or HR management software. As Josh Bersin, a leading HR authority, has said: “Graph databases are vastly more powerful [than other approaches] for modelling how people work in networks, search for data, communicate and build different types of relationships [and] storing this information into a graph of the company which can also evolve over time.”

Just like modern work is evolving too?

Jim Webber is ​​chief scientist at graph database and analytics leader Neo4j, co-author of Graph Databases (1st and 2nd editions, O’Reilly) and Graph Databases for Dummies (Wiley) 

 

Published 31 January 2024
Enjoyed this story?
Sign up for our newsletter here.