Forget About Balance Sheets—Profit Comes From Social and Environmental Data
Leadership diversity and safety records drive profits and stock returns more than some financial metrics
TDLR: Balance sheets, cash flow statements, and profit & loss statements can only tell us so much. Recent data reveals that nonfinancial factors—like leadership diversity and workplace safety—are reshaping the way in which investors evaluate profits and long-term success. In this piece, I share surprising correlations ranging from the beneficial effect of women in management in male-dominated industries, to the negative correlations between injury rates and profits. Many of these drivers are stronger than traditional financial metrics. It seems time to rethink what really drives performance.
What if the secret to predicting a company’s profits was not buried in its balance sheet but sitting quietly in its HR department or safety logs?
For decades, financial analysis has revolved around spreadsheets and ratios, trusting that numbers like EBITDA and free cash flow hold all the answers. But the world is changing, and so is the data that matters.
In male-dominated industries, an unlikely variable—leadership diversity—has emerged as a surprising predictor of profitability. My recent findings show a strong correlation between the percentage of women in management and profits not just today, but years into the future.
Similarly, metrics like workplace injury rates appear to reveal more about revenue potential than many traditional financial indicators.
This is not about throwing out the old tools—it is about understanding the bigger picture. In this article, I will unpack the numbers, share concrete examples from real industries, and explore why these so-called “soft” variables might hold the key to hard financial success.
We will start by quickly examining the status quo before diving into our recent results. We will then go deeper into what this means for financial analysts and the future of the investment industry.
Traditional metrics are severely limited
For decades, financial analysts have clung to a narrow set of metrics—revenue growth, profit margins, return on equity—believing these numbers alone could predict a company's future. This reliance on historical financial data is not just outdated; it is myopic. By focusing solely on balance sheets and income statements, analysts are missing the forest for the trees, ignoring critical factors that drive long-term success.
Traditional financial metrics are inherently backward-looking. They tell us how a company performed in the past, but offer little insight into future potential. Two companies might report identical EBITDA figures. One of them could be led by a diverse, forward-thinking management team, while the other is entrenched in outdated practices. On paper, they appear identical — in reality, their futures could not differ more.
Conventional metrics fail to account for intangible assets—elements like corporate culture, employee satisfaction, and brand reputation—that are increasingly recognized as vital to a company's success. To be fair, analysts do take these into account, but they rely on very few datapoints and often not very advanced analytical techniques to incorporate them into their models.
A study from 2018 highlights that intangible assets now constitute a significant portion of a company's market value. In traditional financial statements, these intangibles do appear under names like “goodwill” or related terms. Most analysts today rarely go into more detail than recording these metrics alongside all the others. Only very few pick them apart in a quantitative and scientifically sound way to understand what exactly is driving a company’s success.
Even more concerning is the oversight of nonfinancial risks. Environmental liabilities, governance failures, or a lack of diversity in leadership may not immediately impact the bottom line, but their long-term consequences are undeniable. Ignoring these factors is not just negligent; it is a recipe for disaster.
In an era where information is abundant and the business landscape is rapidly evolving, the continued reliance on outdated financial metrics is not just insufficient—it borders on reckless. It is time for analysts to broaden their horizons and incorporate nonfinancial variables into their assessments. The future of financial analysis depends on it.
Nonfinancial metrics to the rescue!
Enter nonfinancial metrics. This is no wizardry; in fact, companies need to disclose more and more of this type of data.
It is also not limited to the biggest, stock exchange listed ones. As per new EU directives, companies with operations in the bloc need to publish a long list of nonfinancial data from 2024 onwards. Other jurisdictions in the world are watching this closely and are adopting similar laws.
Beyond this, many companies have voluntarily been disclosing such data for years. Some corporates are ahead of their investors in the sense that they understand the value that nonfinancial data has for understanding a company deeply.
We will here show two examples of phenomena we have encountered frequently in our work: In male-dominated industries, women managers and profits move together. And in heavy industries, worker injuries do not just correlate with depressed stock prices; they are not that great for profits either.
Women in male-dominated sectors drive profits
We encounter this all the time with companies that have an overall percentage of 30 percent or less women in the overall workforce: The higher the percentage of women, the higher the profits. This effect is particularly pronounced for the percentage of women in a company’s management.
Below is a big correlation matrix for Salesforce, a company that sells customer relationship management (CRM) software. We used 12 years of historical data, gathering both financial and nonfinancial data. (We talked about how to gather such data efficiently in previous posts.)
We then performed a time-series analysis and examine the correlation of each pair of variables. If the correlation is not statistically significant, we ignore it; else, it gets included.
There is a lot of information in the above matrix, but let us focus on the entries women_nb_pc, women_nb_mngt_pc, and women_nb_tech_pc. These refer to the percentage of women and nonbinary people in the overall workforce at Salesforce, the percentage in management, and the percentage in tech-related roles, respectively.
When you can see a blue number, the correlation is significant and positive; when it is brown it is significant and negative. White and no number means that the correlation was not significantly significant.
Note that the rather strong correlations for co2_scope3_up_kt — the kilotonnes of carbon emitted upstream of the value chain — are likely spoofs because there were few datapoints to go by. Similar for the community donations, community_donations_M. The other results are interpretable though.
One can see that there is a 0.21 correlation between the percentage of women (and nonbinary people, who are just a fraction of a percentage point) in management and profits in that same year. This is pretty huge. There is also a similarly sized correlation between profits and the number of total employees; note, however, that one would not have to go into massive hiring efforts to hire a few more women in key positions, compared to hiring thousands of additional people for the overall workforce.
While the values of these correlations might be similar, the effort to increase either metric is therefore very different. This justifies our focus on women and nonbinary people, versus total people.
We also see that there is a significant but less pronounced correlation between profits and women and nonbinary people in the overall workforce, and between profits and women and nonbinary people in tech-related roles (0.13 and 0.1, respectively).
Clearly, if Salesforce wants to make more profits this year, a statistician would tell them to hire more women and nonbinary people in key management roles. No financial metric (in the top left corner of the correlation matrix) has such a correlation with profits.
At risk of repeating ourselves, we have seen this phenomenon over and over in male-dominated industries. We also find that the effect of increasing the percentage of women in management peaks three to five years after hiring them.
Of course we cannot tell from this alone whether women are really causing higher profits, for example by working harder or communicating differently. Perhaps a common cause, such a strong company culture, both leads to more women and more profits. This can be verified with causal statistics such as a Bayesian analysis. (We’re on it, to be followed up in a future post.)
Injuries are not good for profits and share prices alike
It is not just diversity metrics. The below matrix is similar and was produced for ArcelorMittal, a steel-producing giant.
In the top third of the matrix you will find the entry losttime_injuries_steel_per_mh. This is the amount of injuries that resulted in lost time (i.e. more than a small scratch on the arm that does not make you go to the doctor) per million hours worked. You will see that it is negatively tied to profits with a coefficient of -0.18.
One might assume that the number of injuries would be positively correlated to profits; the more that workers work, the more injuries occur but the more profits they generate. This is indeed true; however, investments in safety still pay off because the number of injuries per million hours worked negatively affects profits.
With similar companies, we have found that the amount of injuries — total or as a rate per million hours — is also negatively correlated to the share price. It seems intuitive that shareholders might be scared away by negative headlines about harmed workers.
Also, it is just not a good look for any company to be known for wrecking their workers. That being said, we are happy to report that injury rates have decreased massively within the past decade in all companies that we have analyzed so far.
Caveat: Correlation is not causation
In the above text, we have been careful to use the word correlations and to state our interpretations as such. This comes from my previous training as a scientist (I was a particle physicist in a previous career), but it goes deeper than that: Just because two variables co-move does not imply that the one causes the other. There might be another common cause for both that is captured by the dataset or not, or it might be a mixture of reasons.
We are now working on developing our Bayesian approach to better describe causal relationships between pairs of variables. I did such things a lot during my days as a particle physicist; we will be sure to report back to you on our findings in upcoming posts.
Financial analysts and investors should care about this — or they might lose out on returns
Traditional financial analysts may scoff at the idea of integrating nonfinancial metrics into their models, but ignoring them is increasingly a costly mistake. The financial landscape is evolving, and so are the factors that drive profitability. Failing to account for data like leadership diversity or workplace safety risks falling behind competitors who do. Let’s be clear: these aren’t just “nice-to-have” metrics—they’re critical signals that offer a competitive edge.
Why Financial Analysts Need to Pay Attention
The data speaks for itself. Our findings consistently show that nonfinancial metrics such as the percentage of women in management or workplace injury rates correlate with profits and stock performance. These aren’t abstract numbers—they’re predictors of real financial outcomes. For example, the 0.21 correlation between women in management and profits in Salesforce is stronger than any traditional financial metric we analyzed. Analysts who continue to dismiss such insights risk undervaluing high-potential companies or misjudging risk exposures.
Moreover, the regulatory environment is shifting rapidly. With the EU's CSRD requiring detailed nonfinancial disclosures starting in 2024, and other jurisdictions following suit, this type of data is no longer optional. It’s becoming a standard part of corporate reporting. Investors who adapt quickly to incorporate these metrics will have an information advantage over those clinging to outdated methodologies.
Why Investors Should Care
For investors, the implications are equally profound. Nonfinancial metrics provide insights into a company’s culture, resilience, and operational risks—factors that don’t show up in quarterly earnings reports but have a significant impact on long-term value. Consider this: a company with a strong safety culture might avoid costly disruptions and reputational damage, while one with diverse leadership is more likely to innovate and attract top talent.
BlackRock CEO Larry Fink’s 2022 letter to CEOs highlighted this shift, emphasizing that “stakeholder capitalism” isn’t about politics but about driving long-term value creation. Fink’s message is clear: companies that manage sustainability issues well are better positioned to outperform their peers. Investors who ignore these variables risk not only financial losses but also reputational damage as markets and stakeholders increasingly demand accountability.
It is worth noting that Fink has since walked back on his messaging due to political backlash. Judging from the steady number of job openings for sustainability-related and nonfinancial data at BlackRock, the internal policy remains very much pro nonfinancial data.
Losing Out on Returns Isn’t Just Hypothetical
Investors already have examples of what happens when they fail to account for nonfinancial risks. From the Boeing 737 Max disaster (linked to governance and safety culture failings) to the reputational fallout of workplace misconduct scandals, the financial implications of ignoring these variables are real and measurable.
On the flip side, companies like Google and many others have demonstrated how focusing on diversity, sustainability, and employee well-being can drive consistent outperformance.
The Cost of Inaction
Financial analysts and investors who disregard these insights are leaving money on the table—or worse, exposing themselves to unnecessary risks. The tools to analyze these variables are readily available, and the data is only becoming more abundant. By failing to adapt, you’re not just sticking with tradition—you’re falling behind.
As data scientists and analysts, we have an opportunity to lead this shift. By integrating nonfinancial metrics into our models, we can uncover hidden drivers of value and better navigate an increasingly complex financial landscape. The question isn’t whether this matters—it is whether analysts are prepared to act before it is too late.
The future of financial analysis is nonfinancial
The future of financial analysis isn’t about abandoning traditional metrics—it’s about expanding them. Analysts will need to combine financial expertise with data science and statistical tools to extract actionable insights from nonfinancial metrics.
Imagine a valuation model where leadership diversity, safety culture, and emissions data sit alongside EBITDA and cash flow. A company struggling with workplace safety isn’t just risking lawsuits—it’s signaling deeper operational inefficiencies that could cripple long-term growth. Conversely, a firm with rising diversity in management isn’t just checking a box—it’s laying the foundation for future profitability.
Bayesian models and causal inference will replace vague correlations with concrete cause-and-effect analyses. Analysts who fail to develop these skills will find themselves eclipsed by those who understand how to quantify the unquantifiable.
The bottom line: We need more creative financial models
Make no mistake: This transition will create winners and losers. The winners will be the analysts and firms that lead the charge, incorporating nonfinancial data into their models to uncover hidden drivers of value. (Full disclosure: At Wangari we develop such insights and help companies use them for their financial models.) They will be the ones who spot opportunities before the market does and who hedge risks others can’t even see.
The losers? They will be the ones who refuse to adapt, clinging to spreadsheets that tell only half the story. They will miss the next Salesforce, the next Google—companies using diversity, sustainability, and culture as competitive advantages. Worse, they’ll misjudge the risks of companies like Boeing, where governance failures or safety oversights led to catastrophic financial fallout.
Nonfinancial metrics have the power to reshape financial analysis, but creativity is key. Leadership diversity and safety culture are just the beginning. What about metrics like employee satisfaction, supply chain resilience, or community engagement? Could these factors hold hidden insights into a company’s future profitability?
Here’s where you come in: what nonfinancial metrics do you think deserve more attention? Have you seen unexpected correlations in your own work? Let me know in the comments—I’d love to explore these ideas further with you.
What we’re reading at Wangari
We’re all about financial innovation. We were therefore delighted to read about neo-brokerage firm Lemon Markets, which enables any firm to offer investment products to its clients.
pens this in a very beginner-friendly way. What's new in Finance ? Embedded investing is worth a read!Wangari wrangles data every single day. What Does Data Maturity Mean for Data Engineers? is a good reminder that all data is not created equal. Briefly and understandably put by
, this is worth a read for data analysts and data engineers in particular.Say about Elon Musk what you want (we have our opinions too), he does seem to be pulling off quite a feat with SpaceX. Its rather ambitious mission of sending space missions to Mars might just work out. By monetizing its satellite communications service, the company is poised to generate more than enough cash to finance many Mars missions.
shows how a creative businessman might just manage the impossible. How SpaceX will finance Mars is a great read for our more business- (and space-)oriented readers.