Have you ever made a hiring mistake? If you are like most firms, hiring errors are among the most costly mistakes firms regularly make. Biases, human error, and unsophisticated tools to understand candidate-company fit are among some of the reasons people—and companies—continually make hiring mistakes.
Recently, I came across a firm called Pegged Software that is using “People Analytics” to significantly improve hiring outcomes. Applying sophisticated tools to very large databases aggregating candidate and employee information, Pegged Software has helped over 130 facilities in the healthcare space achieve an average decrease in turnover of 38% and improve organizational outcomes, such as employee satisfaction. To learn more about how people analytics can transform an important part of the firm (especially in labor-intensive industries such as retail, healthcare, etc.), I talked with the CEO of Pegged Software, Michael Rosenbaum. What follows are his thoughts on how people analytics is disrupting talent management.
Kimberly Whitler: What is “people analytics”?
Michael Rosenbaum: People analytics is based on the idea that there is a lot more information available on everyone in the world–and specifically employees and potential employees–than what we currently use to make hiring decisions. Employees interact with technology all of the time, and these interactions generate a lot of data that can be used in a variety of ways.
Marketers, for example, have really pioneered the use of this data and analytics to figure out a number of consumer-related issues, such as the likelihood that somebody would be interested in a product. Those same technologies and analytical tools can be applied to predicting whether somebody will be an exceptional performer in their job, helping predict his or her career trajectory. At Pegged, we collect data from the 3 million job applications a year we process, from the approximately 135 healthcare facilities where we are deployed. We have a lot of outcome data across institutions and can apply that to the new hire process. Our goal is to identify the right person for the job by using all of that data to predict who is most likely to perform exceptionally in each job. There are other companies which are focusing on predicting turnover. Our goal is to prevent turnover by identifying the right candidate up front.
Whitler: Is this a new or mature area of analytics?
Rosenbaum: It’s a pretty new space. We’ve been working on this awhile but there are just the beginning signs of a number of companies entering the field. The reason that the field is just emerging is because: 1) the technologies you need to have to run sophisticated analyses at high success rates have matured over the past couple of years, and 2) while the idea that you can use data to make decisions more effectively sometimes has been around, there is now a growing realization that you can apply this to a broader set of questions.
Whitler: Why does this matter for C-Level executives and more specifically, for CMOs?
Rosenbaum: All C-level executives should care about how to hire more effectively. Having the right talent in the right place at the right time impacts morale, individual and group performance, efficiency, and performance outcomes. Labor is often a significant cost for any company, and perhaps more significantly having exceptional talent critical for driving any organization forward. I have yet to find a CEO who says that s/he would not like to reduce turnover and improve performance. In fact, at the top of most lists that rank CEO challenges—you’ll find talent and human capital.
CMOs should care because they actually are the enterprise leaders whose area of expertise pioneered much of this technology, and as a result they are closest to understanding the technology and its potential. Marketers originally innovated and spawned this technology and so they are arguably in the best position to help a firm understand the value. As a result, CMOs can be an important source of knowledge and expertise for the entire executive team, when that team is discussing ways to apply these same processes to talent. This is an opportunity for the CMO to be a thought leader in the C-suite around about how these technologies work and why they are effective.
Whitler: Is it useful at all levels of the firm, or more useful for junior levels where you have more data?
Rosenbaum: Interestingly, what we’ve found is that what predicts success in each role varies from firm to firm and even within firms. For example, we have a client that is a hospital system and it has two facilities across the street from each other. The organization is hiring nurses’ aides at both facilities, with one being an acute care facility and the other being a long term care facility. The organization had been hiring the employees in the same way, assuming that all nurses’ aides jobs were the same. However, it found that in some cases, attributes that predicted the success of a candidate for one role actually predicted the opposite for the same role across the street. What that tells me is that every organization is unique, and that the idea that all CMO jobs or all nursing jobs are the same is wrong. The company, the responsibility, the culture, the style, the type and degree of experience can all impact whether somebody is a good fit, which then turn into predictors regarding whether somebody is likely to succeed or not.
Whitler: What happens if somebody is deemed a non-fit for a specific position?
Rosenbaum: One of the strengths of people analytics is that a candidate can apply for a job and when we analyze the fit, we analyze it across several positions and not just the one for which the candidate applies. If the candidate would be a good fit for a position for which she did not apply, we route the candidate to that recruiter or hiring manager. As a result, we are able to allow candidates to get the jobs for which they would be a good fit, and we allow organizations to draw on the strength of their entire applicant pool for each position it needs to fill.
Whitler: How is it disrupting talent management and executive recruiting?
Rosenbaum: We have a tendency to hire people like ourselves, who have a common set of experiences and outlook. If you think about it, it is in many ways an irrational way to make a hiring decision, and it results in more homogenous teams that perform overall at a lower level. Yet, as human beings, we often are not even consciously doing this. In contrast, data and analytics allow us to find the best candidate in a big pool of people rather than filtering decisions through biases that are innate to each of us. If you can make better hiring decisions, you can transform the way a company operates.
This matters whether it occurs at the entry level or in executive recruiting. Those firms which incorporate more analytical tools into their candidate assessment approaches should yield better outcomes and client satisfaction. The industry is in the middle of disruption, with some staff who have done things one way for a long time leaning into a human-centered (and often distorted or flawed) decision-making approach and others more open to disrupting a process they recognize as deeply flawed leveraging data and analytics to search for a way to create differentiation in a relatively commoditized industry. As an example of the power that this can provide, our median improvement of employee turnover is 38%, and worst deployment ever in any department or any role or any organization in which we have deployed was a reduction in employee turnover of just 13%. When you actually measure our cost (to figure this out) relative to the cost of having higher turnover, the ROI occurs within 5 days.
Whitler: Has this technology been used to help young people, candidates, and those in career transitions better identify the jobs and type of firms where they be a good fit?
Rosenbaum: There are a couple of startups starting to look at the benefits for individuals. One important factor is that we can see anecdotally in the data that people get discouraged by a “no”. They may apply for two nursing jobs, get rejected, and you can almost see that they believe that it means they aren’t a good nurse. What our data suggests is that this isn’t true. They actually may a great fit for a nursing role in a different type of organization.
This is powerful. Explicitly or implicitly, the labor market generally assume that all nursing jobs are the same and all CMO jobs are the same. But your boss, your responsibilities, the company culture, etc. aren’t. While you might be successful in one CMO job, you might very well fail in another. And this is where many talent management and executive recruiting firms have failed. A firing does not necessarily mean the candidate is incompetent. The answer is much more nuanced than this. People analytics provides this nuanced understanding.
Furthermore, we have a lot of data that demonstrate that hiring managers are often bad at identifying what they need or picking the right candidate, and people are often bad at finding the right job for themselves. Both of these decisions and actions are exceptionally challenging for an individual human being to do well. The good news in our work is that: 1) the fact you were not hired into a position does not mean you are incompetent—it may mean either that you were not a good fit for that particular role or that the hiring manager or recruiter could not tell how to determine you would be a good fit, and 2) an ability to bring greater sophisticated insight into the hiring and firing process delivers dollars to any company’s bottom line both in terms of improved strategic progress and revenue and in terms of operational savings.
On top of it, there is a silver lining to all of this insight. Since we have such massive amounts of data about an organization, we are able to generate broader learning and information about cultural dynamics in an organization. For example, we can tell which departments are doing things differently and which are innovating? A side benefit is that firms actually get tremendous insight on their own company operations that they otherwise may not have been aware of.
Interested in more articles about healthcare and analytics: A New Marketing Framework for Healthcare, Start-up Marketing Challenges for a New Healthcare Company, Managing Rapid Growth: Perspective from a HealthCare Executive, How the Best Firms are Using Analytics to Create Competitive Advantage, The Marketing Analytics Silver Bullet, The Big Data Challenge: Generating Actionable Insight.
Join the Discussion: @KimWhitler
This article was written by Kimberly A. Whitler from Forbes and was legally licensed through the NewsCred publisher network.