Employee retention is a big issue right now, and understanding what motivates someone to stay or go can make a huge difference for employers. We heard that the Great Resignation was all about compensation, and even our own research seemed to back that up. But a recent study published in the MIT Sloan management review challenges this notion, arguing instead that toxic culture is the number one predictor of turnover.
We crunched the numbers using a data simulation based on this research, and came to the conclusion that even small efforts on the part of L&D (Learning & Development), HR, and leadership teams can have an outsized impact on employee retention over time:
A slight increase in the amount of efforts to improve work culture could lead to dramatic reduction in employee turnover over time.
The dots represent the number of accumulated employees that left over time.
Over time, people get dissatisfied and leave the company
Let’s unpack what this means for L&D teams, and better understand the small steps we can all take to increase employee satisfaction and retention.
What authors Donald Sull, Charles Sull, and Ben Zweig found was that a toxic workplace culture was 10.4 times more likely to be the source of an employee leaving than their compensation (even when adjusted for industry). They reached their conclusions by analyzing 34 million online employee profiles from Revelio Labs, as well as the free text of more than 1.4 million Glassdoor reviews from April through September 2021. Using this data, they estimated company-level attrition rates for Culture 500 companies.
What they were able to do was paint a picture of why employees left (or lost) their jobs during the Great Resignation:
As shown above, toxic corporate culture was the most likely reason for employee attrition. By ‘toxic corporate culture’, the authors mean a culture that is disrespectful, non-inclusive, unethical, cutthroat, and/or abusive.
A toxic workplace culture was 10.4 times more likely to be the source of an employee leaving than their compensation.
Job insecurity and reorganization also topped the list of factors contributing to a high attrition rate, as did high levels of innovation (which can be explained by the stressful work environment they entail), failure to recognize employee performance, and a poor response to COVID-19.
We also polled American employees about why they were considering quitting their jobs (or had already left) from November - December 2021. At first glance, our survey would seem to indicate that, contrary to the above findings, compensation is the main factor in employee turnover:
But, if we were to see burnout, inability to work remotely, limited opportunities for career progression, lack of learning, and a values/culture mismatch as falling into the categories of toxic culture, failure to recognize employee performance, or poor response to COVID-19, we start to draw a similar conclusion.
So, how can we address this churn and encourage talented people to stay with us?
Through the team’s analysis of the free text left on Glassdoor reviews, they were also able to identify factors that limited attrition. In other words, what attributes encourage employee retention? The most important was providing lateral career opportunities, followed by remote work arrangements, company-sponsored social events, and offering predictable schedules.
Comparing these results with our Great Resignation survey provides interesting additional context. For instance, we also asked employees who recently quit (or were thinking about it) if they had asked their employer about internal mobility (which would include ‘lateral career opportunities’). Sadly, almost half said they were turned down when they inquired:
In terms of remote working, 14% of our respondents who quit or want to quit their jobs, it was because their employer wouldn’t let them work remotely. While many had COVID-19 at the top of their minds, they also cited the convenience and flexibility of working from home:
So, what role can L&D teams play in shaping a company culture that retains employees?
Let’s take a look again at the factors that increase employee retention:
The number one factor—providing lateral career opportunities—is one in which learning and development teams play a critical role. Through working with employees to identify learning needs from the bottom up to initiating effective upskilling and reskilling programs, L&D professionals can work with management and leadership teams to enable more employees to turn to internal mobility if that’s where their interest lies.
While company-sponsored social events might fall more on HR or Culture colleagues, L&D still has a role to play in remote work arrangements and offering predictable schedules. Working remotely may require training regarding processes, tools, and communication.
For instance, at 360Learning we’re a remote-first company that relies on asynchronous communication, and we incorporate courses on how to communicate and use our internal tools set up for remote working as part of our onboarding program. As for offering predictable schedules, this might fall into wider manager training.
The factors that lead to attrition can also be limited through savvy L&D efforts, too. Better leadership and manager training can address many of the toxic workplace attributes, along with DEI programs that address LGBTQ, disability, racial, age, and gender inequities, among others.
Training around managing high workloads and stress (as we’re also rolling out at 360Learning) could help address the added pressure of working for a highly innovative company. As for failing to recognize employee performance, this is another area where effective manager or leadership training could be useful.
But how much time do L&D teams really have to roll out new programs and add new priorities? As budgets shrink and teams turn to heightened efficiency during an economic downturn, how can L&D teams make the argument that these added efforts will pay off?
To give learning and development teams a helping hand, we simulated (as shown above) the scenario outlined by Donald Sull, Charles Sull, and Ben Zweig. Our model (shown above) represents the same findings as their study did, showing the positive and negative effects of each of the factors they identified (except how companies responded to COVID-19) among three fictitious organizations: Company A, B, and C.
Company A represents companies that continue with outdated work practices—the types of (toxic) workplaces employees in the Sull and Zweig study ended up leaving. Company B, however, made small improvements to their culture. This might be through granting more remote working flexibility, or approving more lateral internal mobility. Their efforts remain modest, and wouldn't require excessive time or budget from the L&D, HR, or leadership teams.
Finally, Company C goes a step further, taking moderate steps to address their company culture—they increase their L&D activities by 20%.
According to the simulation, after just one year, Company A is at almost 20% attrition rate, whereas Company B has brought this down to around 11%, and Company C, 7%.
Over time, this adds up to a huge disparity in employee turnover for the three companies:
This tells us that even through small efforts, L&D teams can have a huge impact on employee attrition, whether through their direct intervention or through their influence of other teams. Not to mention the sunk costs associated with constant onboarding and training brought on by frequent turnover.
In other words, you don’t need to solve company culture overnight–even small improvements can add up over time to higher rates of employee satisfaction and lower churn.
If reading this article has left you worried about where to start, don’t stress. So, where can L&D teams turn to help them put in place the kinds of training and initiatives that improve workplace culture and limit employee attrition?
Here are a few good places to start:
We used an agent-based model to simulate the employee actions and dynamics inside an enterprise. Agent-Based Models (ABMs) are computer models that attempt to capture the behavior of individuals within an environment. They have become a standard technique in the toolbox of scholars when real-world data is unavailable.
If you’d like to learn more about how we run the simulation, here is a link to our study methods. The simulation was coded in Julia Language using the Agents.jl toolbox. We also uploaded the source code used in this analysis to GitHub.