From theory to practice: Insights from Cole Napper, VP of people analytics at Orgnostic

Meet Cole Napper, a VP people analytics & Chief evangelist at Orgnostic. Beyond authoring and podcasting, Cole’s contributed to various sectors, from venture-backed startups like Orgnostic to giants like PepsiCo, showcasing his expertise in people analytics, HR technology, and talent management and teaching in top MBA programs.

In this interview, we delved into the nuances of people analytics, discussing its significance in modern businesses, its intersections with HR functions, and its potential benefits. We also explored the challenges and implications of AI in people analytics and the future of the field in a rapidly digitalizing world.

People analytics is just like any other function in a company focused on adding value to a business through data-driven decision-making but based on its employees rather than its customers or its technology or whatever other analytics functions may be focused on. Because I take a comprehensive definition of anything related to its employees as people, that’s the sphere related to people analytics.

A somewhat extreme viewpoint suggests that people analytics should be a distinct business unit reporting directly to the CEO. I find this perspective a bit excessive. More fittingly, people analytics should be housed under the HR umbrella but maintain significant influence on business matters, not just HR matters. Ideally, the function would be overseen by a CHRO or a Chief People Officer, although not all companies have such positions. People analytics is a horizontal, boundary-spanning function, which makes it distinct from traditional HR functions in some ways.

At its core, the primary purpose of people analytics is to be self-sustaining by generating value. In essence, it should not only justify its costs but exceed them.

That could be by increasing innovation, increasing sales, decreasing defects, decreasing churn turnover, improving talent acquisition and hiring, etc. It could be a variety of different tasks that the functions is doing to make a 10x return. But whatever is going on in that business or industry, people analytics is responsible for adding value. And so, I believe if you’re looking to try to pay for yourself ten times over, you’ve got to have the equation right for both the numerator and the denominator, right? So, you want to keep the denominator as small as possible to create as much value for as little cost to the business as possible. Many folks don’t believe in this type of rule, and they create these massive people analytics teams. And it’s almost impossible to pay for yourself multiple times over if your cost structure is too high.

Let’s review these three use cases below.

Talent Acquisition. When discussing talent acquisition, it isn’t just about measuring output like the number of people hired or the quality of the hire, although these are crucial metrics. There’s also a need to analyze throughput measures: the applicant funnel, the D&I (diversity and inclusion) progress at various recruitment stages, and identifying at which steps candidates typically exit the process and candidate experiences.

As for Retention. It can be seen as a bookend of the overall employee experience. Retention often involves predictive measures, forecasting turnover, and understanding why employees depart. Economic context matters, too. For instance, during the “great resignation” period, retention was a dominant focus for many companies. Due to economic instability and organizational upheavals of late, such as mergers and downsizing, turnover is much less of an immediate concern in some industries.

Employee engagement. While engagement might not have been as prominent a concern a few years ago (when perks and benefits were plentiful), it’s now at the forefront. The current focus on engagement stems from its relationship with productivity. Companies are tasked with doing more with fewer resources. However, this push for increased output is counterbalanced by rising employee burnout and feelings of job insecurity. It’s not just about gauging employees’ engagement but also strategizing how to maintain or improve engagement in a competitive and sometimes challenging business landscape.

So, trying to create measures and understanding of how engaged employees are, which is important, but including passive and active methods for collecting data to understand the situation better, using data and analytics. But also, what are the requisite initiatives that companies can tackle from an employee value proposition standpoint? How can they keep pace with their competitors to increase productivity without burning out their employees?

Well, this is an interesting topic to discuss. There are a few factors at play here. One we talked about on my podcast recently, called Directionally Correct, where we discussed managers’ regret over hastily deciding to revert to office work without sufficient data. Their remorse wasn’t based on the belief that returning to the office was inconsequential. Instead, they felt they lacked the data to back their decision.

The crux of the issue is that while many organizations could’ve gathered data to inform their decision to RTO, they opted not to. HR teams at many organizations have suggested collecting this data to avoid abrupt decisions. Ironically, some business leaders, now regretful, were initially against data collection in the first place. They felt it might suggest to employees that they had a choice.

Another twist to this narrative is the potential misuse of people analytics. Some companies did gather data, understanding the employees’ preferences precisely. Yet, they intentionally formulated policies that could drive certain employees to resign.

For example

If a company wanted to downsize without the negativity associated with layoffs, it could force conditions that lead employees to quit. They might promise an employee the flexibility to work remotely from Kansas but later mandate relocating to the New York office, indirectly compelling the employee to resign — a subtle approach to layoffs. Very Machiavellian.

Yes, there are various perspectives on this matter:

Manager’s remorse. There are numerous companies where managers need more data to back their decisions regarding office returns. Many of these companies needed a dedicated people analytics team and thus needed to be more informed in their choices.

Relocation. Some companies urged employees to relocate. However, not all such requests were a masked attempt at layoffs. Some legitimately wanted to increase collaboration by having employees in the same place.

The past six months to a year have seen a flurry of questionable actions, especially following the onset of tech layoffs, which then influenced other sectors. Previously, there was a steady, moderate rise in pay rates. However, the pandemic brought about a significant surge in talent costs. Companies now aim to revert to the pre-pandemic pay trajectory, which means they’re looking to cut wages.

Companies are grappling with increased labor costs, which they believe may be unsustainable. While PA professionals might approach these issues from a research perspective, it’s essential to understand the business viewpoint. Businesses will leverage data for their decisions, but whether they’ll utilize the data provided by their analytics teams remains uncertain.

A specific term in academia currently eludes me and describes the phenomenon where the most talented individuals in your workforce are also the most mobile (talent fluidity). Conversely, those less talented are usually less mobile. Many businesses might assume that adverse decisions would predominantly affect the less talented segment, expecting them to be the ones to leave.

However, the paradox is that these individuals are often most inclined to stay unless they are pushed out. Conversely, the top talent, due to their mobility, are most equipped to leave if they wish when unfavorable conditions arise.

I have a lot of complex thoughts and feelings about the topics we’ve been discussing. But we haven’t discussed the root cause of what we see occurring.

They might point fingers at the economy, shareholders, the board of directors, or investors, claiming they have no choice but to make layoffs. However, digging deeper it often concerns me as to whether a business is innovative or stagnant.

Businesses that are innovative and growing consistently push the boundaries, disrupt their operations to stay ahead, and usually avoid such cyclical challenges. However, those companies that have ceased to innovate and are merely managing their decline find themselves in these predicaments. They tend to adopt a strategy of continuously slashing resources, aiming for efficiency even as their revenue and employee numbers dwindle.

We can preempt many of these challenges by emphasizing creating an innovative culture and retaining top talent while promoting collaboration and engagement. Businesses thriving in such an environment empower their employees with autonomy, mastery, and purpose — key elements for optimal performance. Many of these strategies aren’t even capital-intensive. They usually have very low costs, usually. Like treating people with respect and dignity.

On the contrary, bloated businesses with excess middle management and cumbersome processes stifle innovation. By having multiple layers of approvals and endless meetings, you’re not only adding overhead but also curbing the spirit of creativity. These factors are measurable. A proficient people analytics team should be delving into such matters, ensuring the longevity and success of their businesses.

What sets a people analytics manager apart from a regular business manager? For the regular manager, breaking the norm often means taking risks.

A people analytics leader is responsible for being innovative, gathering intelligence, and staying ahead of trends. It’s easy to stick to what’s familiar, but the challenge lies in actively engaging across business units to foster change. Effective people analytics functions must drive this change, positioning themselves as leaders rather than mere participants.

We often hear, “Employees are our greatest asset,” yet actions frequently suggest otherwise. If a business plan lacks commitment and action, it’s merely an idea, not a strategy.

I approach this from the 80-20 perspective: what is the 20% of data that will address 80% of our challenges? This is why I’m firmly against a one-size-fits-all strategy in people analytics. Such an approach demands 100% data collection on every possible issue, which is unrealistic unless you have unlimited resources.

Instead, I focus on identifying the four or five key challenges that might face the business in the future. By proactively collecting data on those challenges, we’ll be equipped with a year’s worth of insights when these issues escalate. This foresight ensures that when business leaders approach people analytics for solutions in the future, they are already prepared with data.

Sadly, many analytics leaders claim to grasp this concept but fail to implement it. To be effective, aligning closely with the business’s core concerns is crucial. For instance, collaborate with operational analytics teams if you need operational data. For sales data, partner with sales teams, etc.

Too often, people analytics relies solely on basic metrics like talent acquisition and retention. But when a unique challenge arises, they’re left scrambling to gather relevant data over an extended period. This reactive approach makes them less valuable to the business.

Your primary data sources in people analytics include the HCM, ATS, and increasingly, what’s termed as “digital exhaust” – data from meeting schedules, calendar invites, and communications platforms like Slack. Additionally, employee surveys are invaluable. While often underrated, they can provide insightful data, especially if kept confidential but not anonymous. However, augmenting this HR data with relevant business data is crucial.

For example

A trucking company might want to analyze the efficiency metrics of its drivers. In contrast, a software company would be interested in sales, engineering, and product data. A manufacturing firm might prioritize data on production throughput across different sites or even data from badge swiping systems.

The goal isn’t just to operate within the confines of HR but to broaden the scope, be business-centric, and span boundaries.

In my career journey, I realized that despite my Ph.D. in industrial-organizational psychology, there were two crucial areas in which I needed to gain knowledge. Though I had a strong foundation in social sciences, I needed to delve into two additional pillars: understanding businesses and data science.

Additionally, I deepened my knowledge of data science, data engineering, and HR technology. To be truly effective in people analytics, one must be proficient in three areas: understanding human behaviors through social sciences, knowing how businesses operate, and grasping data systems. Balancing expertise in all three makes for a potent people analytics professional. Those who only emphasize one or two of these areas often face challenges.

I work for a vendor specializing in generative AI exclusively for people analytics. Interestingly, while other vendors hint at developments in this area, we are among the few truly delving deep into the topic and shipping products that work. I’ve also explored offerings from platforms like Midjourney and ChatGPT.

ChatGPT, for instance, operates on broad AI principles. It can help craft a job description or summarize performance reviews. But the novelty of this type of work quickly fades. While I frequently write articles, I’ve observed that ChatGPT is a handy tool for generic content. However, for unique insights, its AI origin becomes quite evident. It’s useful for literature review sections but not-so-useful for introducing novel concepts.

I categorize the AI goals in people analytics into two: those prioritizing AI for decision-making and those emphasizing AI for automation. I firmly believe in the latter.

Despite the challenges, there are still advocates for such approaches, which I find perplexing.

There are numerous issues with the current approach, yet one stands out that isn’t being widely discussed. I’ve delved into this topic multiple times on my podcast. In the U.S., if an entity commits an error, there needs to be a means for redress.

It reminds me of an old IBM slide from 1960 that stated, “Computers can’t make decisions because they can’t be held accountable.” Replace ‘computers’ with ‘AI’ and the sentiment remains. If AI can’t be held accountable, it shouldn’t be making decisions. The mechanism for AI accountability doesn’t exist now and seems unlikely in the foreseeable future. So why is it a recurrent topic of discussion?

For example, what if we no longer needed to design another dashboard because an AI could do it more efficiently, attractively, and accurately?

That’s the direction my company is heading in. Our goal is to leverage AI to help us understand what’s happening through textual or visual interpretations, so you pose a question and get an immediate response. But when it comes to decision-making, it’s not about the AI dictating our actions. Instead, we envision AI analyzing the best available research and providing recommendations based on top-tier studies. It would empower us to make informed decisions rather than solely relying on human intuition.

I truly believe this is the future trajectory, not just for people analytics but for various sectors. It promises to free us from time-consuming chores that we dread, allowing us to focus on more meaningful endeavors.