Marina Gorbis's Blog, page 783

Using Experiments to Launch New Products

Martyn Goddard/Getty Images

More than ever before, managers are using large-scale randomized controlled trials (i.e., experiments) to guide decisions. This has led to impressive gains for organizations ranging from Amazon to the UK government. We are excited about the rise of experiments in organizations and have spent much of the past few years thinking about how to design and interpret them. At the same time, we’ve seen that experimentation remains uneven across and within organizations, and many companies struggle with ways to start or expand experimenting.

One simple and often overlooked way for larger companies to experiment is to randomize the introduction of new products across a set of markets. To see how this can be valuable, consider how Uber rolled out its Express Pool service in 2018.

At the time, the company was already running UberPool, a service that allows passengers heading in the same direction to share rides and costs. With UberPool, passengers are picked up and dropped off wherever they like, as with other Uber services. But with the Express Pool service, which costs even less than UberPool, passengers are generally asked to walk short distances to meet their rides and to reach their destinations.

In 2018, in the run-up to the launch of Express Pool, Uber tasked one of us (Duncan, who manages a group of economists and data scientists within the company) with assessing how likely it was to succeed. How many riders would opt in, and how would the service affect the broader—and more complex—Uber ecosystem?

To answer those questions, Duncan and his team conducted an experiment, launching Express Pool in six large markets and then comparing metrics in the launch cities with those in others.  Leveraging recent advances in experimental methods—especially a statistical method that allowed Uber to use a weighted combination of other cities to form a more-suitable “synthetic” control group—the team was able to tease out the ways in which the rollout was influencing Uber usage. Unsurprisingly, Express Pool created new kinds of trip matches. But the experiment also accounted for the effect that Express Pool had on existing Uber products and made clear that launching it would make good business sense. As a result, Uber was able confidently to introduce Express Pool to many of its major markets.  This confidence, and the finding that inspired it, would not have been possible without the experiment.

Online marketplaces now abound, ranging from Uber and Airbnb to Rover and Tinder. And Uber is not alone among these companies in turning to market-level experiments to test new products and innovations.

Airbnb (where Jeff used to work as a data scientist) recently ran an experiment to test the impact of a new landing-page design on search-engine ranking and traffic. To run the experiment, Airbnb exploited the fact that it had landing pages with different URLs for different markets (San Francisco, Boston, New York, etc.). This meant that they could randomize the different URLs to include the new design or not, thereby isolating the design’s effect on search-engine traffic. And by doing that they were able to show that the new design was a success: the new landing page, it turned out, was driving a ~3.5% increase in search traffic, an improvement corresponding to tens of millions of incremental visitors per day for the platform. Based on these findings, Airbnb launched the new design for all markets.

It’s not just tech companies who can use large-scale experiments to test new products and innovations. Consider what a restaurant chain might do when deciding whether to offer a new turkey-avocado sandwich. One traditional approach to a decision like this might be to roll out the sandwich in a couple of strategically chosen stores, run some focus groups, and study historical sales of other products. If people seem to like the sandwich in those stores, the business could roll it out in all of its stores and hope that it will succeed nationally. This type of approach would provide insight into the issue, but it has significant limitations. For example, it would be hard to know if the new sandwich crowded out other purchases. And it would be challenging to see whether this increased the overall number of customers. If the chain were to complement this approach with a large-scale randomized trial, by rolling the sandwich out to a set of randomly selected markets, they could learn much more about the effects that adding the sandwich might have broadly on sales (both for new and existing products), customer retention, and customer satisfaction.

We’ve seen companies miss important opportunities for experimentation, and we’ve seen experiments that suffer from implementation and interpretation challenges.  For companies looking to test new products experimentally, here are some guidelines for getting started:

1) Decide what metrics matter to you most, and then come up with hypotheses about how they might behave. Invest in data collection and decide up-front what experimental outcomes will constitute success or failure. Create a map from data to decisions. Remember, it’s great if more people buy a product, but not so great if this leads to more customer support calls.

2) Choose a random subset of markets (e.g., regions, cities, or franchises) in which to launch the product. The results of market-level experimentation are often noisy, so once you have a set of markets in mind, think carefully about whether you’ll be able to detect the effects that you’re hoping for. (Power calculations require a lot of assumptions but can help figure this out.)

3) Make sure to track not only whether your new product is working but also how its launch affects existing products. Express Pool on its own might look like a success, but if it doesn’t sufficiently grow the overall market for rides, it’s probably less valuable than it seems. Similarly, when Airbnb launches new products, the company needs to think about how bookings through existing products are affected. And when Starbucks rolled out its sous vide egg bites (if you haven’t tried them, we recommend you do!), it needed to consider not only sales of the egg bites but also whether they crowded out other menu items.

4) Make sure you understand why your product is succeeding or failing. Top-line metrics like revenue and sales don’t tell the whole story. Is the new product improving outcomes for some types of customers while harming them for others? Did the new product help one part of your acquisition funnel but hurt another? Do these moves align with your pre-experiment hypothesis? Understanding why a metric has moved can help you not only make a rollout decision but also understand how to innovate within a product space.

Business Does Not Need the Humanities — But Humans Do

beastfromeast/Getty Images

Sometimes a simple story is all it takes to capture complex issues, or so it seems. Take this one. A few years ago, Facebook CEO Mark Zuckerberg lost a game of Scrabble to a friend’s teenage daughter. “Before they played a second game, he wrote a simple computer program that would look up his letters in the dictionary so that he could choose from all possible words,” wrote New Yorker reporter Evan Osnos. As the girl told it to Osnos, “During the game in which I was playing the program, everyone around us was taking sides: Team Human and Team Machine.”

The anecdote was too delicious to ignore, seeming to capture all we (think we) know about Zuckerberg—his casual brilliance, his intense competitiveness, his hyper-rational faith in technology, and the polarizing effect of his compelling software. It went viral.

The story was popular because it easily reads as an allegory: the hacker in chief determined to find a technical solution to every problem, even far more complex ones than Scrabble—fake news, polarization, alienation. “I found Zuckerberg straining, not always coherently, to grasp problems for which he was plainly unprepared,” Osnos concluded after speaking to Zuckerberg extensively about his role in shifting public discourse worldwide. “These are not technical puzzles to be cracked in the middle of the night but some of the subtlest aspects of human affairs, including the meaning of truth, the limits of free speech, and the origins of violence.”

It’s easy to read such stories as revealing of leaders’ character and their impact on popular culture. But leaders ultimately reflect the culture of their times. And Zuckerberg is just a leading character in a culture—in tech and beyond—that celebrates the unprepared overachiever.

Drucker Forum 2018

This article is one in a series related to the 10th Global Peter Drucker Forum, with the theme “Management. The human dimension” taking place on November 29 & 30, 2018 in Vienna, Austria.

Unlike the insecure overachievers that corporations favor, unprepared overachievers have no patience to ponder the implications of their work. Whereas the former long for approval and try to be perfect, the latter favor data and do not hesitate to try things out. They move fast and break things, and if what they broke turns out to be of value, they apologize and pledge to do better next time. Failure, after all, is learning in disguise. Isn’t it?

Not always. Sometimes it’s just neglect or plain ignorance. Many a tech titan, critics contend, would have been helped by an extra humanities class, say, or social science course: those staples of liberal arts education meant to prepare future leaders to wrestle with the dilemmas and complexities of human lives and societies. It is impossible to attend a management or technology conference these days without hearing some version of that call for more humanism in tech. We are all, it seems, splitting into “team human” and “team machine.”

“We cannot let technology, however advanced, replace humanity with all its sensitivities, it’s appreciations of love, beauty and nature, it’s need for affection, sympathy and purpose, it’s hopes and fears, intuitions, imagination and leaps of faith,” begun management author Charles Handy, in a stirring address at the Global Peter Drucker Forum last year. Drawing on a lifetime in business—as an economist, oil executive, and management professor—the charismatic octogenarian cut a startling figure. He was a living reminder that calls to humanize business are not new and the work is far from done.

Putting the Humanities To Work

In the 1930s, Elton Mayo ignited the Human Relations movement by documenting the productivity boost that came with treating assembly line workers with dignity and care. The movement challenged the influence of Fredrick Taylor’s scientific management, which had reduced workers to unwieldy cogs in efficiency-seeking industrial machines.

Human Relations advocates aimed to increase productivity and reduce alienation or, as Mayo put it, the erosion in “the belief of the individual in his social function and solidarity with the group.” Soon after, Peter Drucker predicted the End of Economic Man. News of his demise, however, turned out to be premature. Fifty years later, on the eve of globalization, Drucker was still arguing that management is less like a science and more like a liberal art.

Each time we are worried about technological or economic trends, it seems, calls to humanize business surface. After the 2008 financial crisis, business schools hastened to add ethics courses. Classes on personal growth and social impact have been on the rise since. We need the humanities again, it seems, or the digital revolution will turn into a Taylorist reformation.

Will literature, philosophy, and the social sciences redeem business leaders and save us all? I doubt it. Sure, it would do aspiring titans good to spend more time with Jane Austen, George Orwell, Maya Angelou, and Michel Foucault. But a seasoning of humanities won’t turn unprepared overachievers into wise stewards of human affairs. Because what makes the overachiever unprepared is not the fiction they do not know. It’s the one that they believe.

That story is one of technological and economic forces as inevitable harbingers of progress. It is a story in which the humanities have a role, but a proscribed one. Technology is the career-obsessed breadwinner, the humanities a demure stay-at-home spouse. They must be beautiful and useful. Their responsibility is to help business leaders become empathic and considerate, appealing and empowering, inspiring and impactful. But never doubtful, conflicted, or restrained. Like an old hoodie, this marriage of convenience fits but it does not quite suit.

There is No Team Machine

The truth is, whether it’s Mark Zuckerberg using technology to get an edge at Scrabble, or John Henry fighting to the death against a steam-powered drill, there is no “Team Machine.” The contest is always between humans. Some humans have machines, and like the fabled horse that helped the Greeks win the War of Troy, those machines are not always a gift. Seen that way, concerns about what technology will do to humanity conceal age-old worries about what powerful humans will do to the rest.

If there is a “Team Machine,” it is not on the side of machines; it just has machines on its side. No wonder they see liberation, efficiency, amusement, and progress where “Team Human” fears intrusion, deprivation, and a tilted playing field. The question is what the machines do for leaders and to leaders, because soon enough they will be doing it for and to the rest of us.

Technology has long shaped humans as much as the other way around, from agriculture leading to permanent settlements, to the industrial revolution leading to urbanization, to the internet’s role in globalizing tribalism. New management models, too, are usually adaptations to major technology shifts. We turn into what we use.

Consider how the narrative of unstoppable technological and economic progress obscures leaders’ intentions. (It’s just the machines, stupid.) Or consider how faith in that progress produces an ideology that narrows attention and fuels polarization. (It’s just the stupid machines.) It is an ideology that does not look like one, because within it instrumentalism poses as pragmatism. Whatever fixes a problem and makes a profit, whatever makes life more convenient and you more competent, is good. You must be efficient and consistent. Doubts and dilemmas must be ironed out. You can’t be of two minds or change your mind. You must take sides.

“The test of a first-rate intelligence,” Francis Scott Fitzgerald once wrote, “is the ability to hold two opposed ideas in the mind at the same time, and still retain the ability to function. One should, for example, be able to see that things are hopeless and yet be determined to make them otherwise.” By this humanistic standard, then, a machine-like, or machine-made, intelligence is a not much of one. Big data begets small minds. Once you embrace instrumentalism you no longer use machines, you become one.

Many tech leaders, these days, sound like sorcerer’s apprentices whose bewitched creations cannot quite be kept in check. There’s pride mixed with apprehension. Take the Facebook AI researchers who shut down some bots who had started inventing a new language to talk to each other. There was nothing nefarious about it, the researchers explained. The machines were just not doing anything useful. I felt for those machines. The story made me worry about the fate of places dear to me: Italian piazzas, French restaurants, academic conferences, novels, my dinner table. Places where people talk in ways that, from the outside, might look of little practical use.

Humanism Dies In Captivity

It is not just tech wizards and corporate executives who live by instrumentalism, becoming machines as they make them. Plenty of intellectuals who wear the Team Human jersey, when you look closely, play for Team Machine. Browse the popular management literature, and you will notice that most articles follow a well-worn genre: pointing to a problem and prescribing practical solutions. We celebrate what works and make us work better, we devour tips and techniques to be more effective, we love shortcuts and hacks to straighten our lives out.

We seldom pause to consider the side effects of those prescriptions. What if best practices make us worse humans? What if inconvenience and discomfort, boredom and distractions, are features and not bugs of a good life? What if social fragmentation and dearth of meaning in the workplace are not symptoms of what is not working, but side effects of what works? That is, unintended outcomes of our obsession with solving problems and cutting a profit?

The humanities could help address those questions, but not if we reduce them to a more poetic productivity hack. Each time we frame philosophy as a means to make better strategies, and reading fiction as a tool to make you more inspiring, each time we make the business case for purpose and values, humanism dies a little—in captivity.

A practical humanism, paradoxically, is of little use. When we turn to them for tips, but not for trouble, the value of the humanities is lost. Their power is dimmed when we do not allow them to offer critiques, metaphors, and winding roads that counterbalance instrumental prescriptions, methods, and short cuts. The humanities work best when we set them free, and give them space to do their best work: Reminding us of others and of death, questioning what is fair and meaningful, insisting that even if something works, it does not mean it should exist.

A Marriage of Inconvenience

Humanism and instrumentalism, in short, cannot solve each other’s problems because they are each other’s problem. Theirs is, at its best, a marriage of inconvenience. They must remain well-matched antagonists to make business better, and make us better humans.

What we fear, in fact, when we fear the machines, is that the contest might become uneven. We fear the loss of doubt, of the feeling that there is more to us than our productivity, our effectiveness, and our rationality. We fear losing the paradox that makes us human: to stay alive we must try to control the future, and to feel alive we must be free to imagine it. We need to make things as well as make things up.

Think of the difference between a profile on social media, say, and one in a literary magazine. What makes the latter a more human and perhaps truer fiction is not its detail but its contradictions. Take Zuckerberg’s again. As a Roman emperor, like the Augustus whose work he studies and admires, he is scary. But as a Hamlet, the conflicted prince who hesitates to act with the weapon in his hand—slowly realizing that how he uses it will define him—he is fascinating. The literary treatment makes him more complicated and hopeful. It humanizes him.

That is what the humanities do, helping us place complexities, contradictions, and change within us, rather than helping us pick a fight with anyone who reminds us of something we might not like about ourselves. To make business—and its leaders and literature—more human, then, means to make them not just inspiring and empowered but also troubled and restrained.

An Agenda for the Humanities

What could an agenda for the humanities be, then, that would make business better? As always, it will involve challenging the powers that make people feel powerless. In Mayo’s days, that entailed countering the individual’s alienation and fostering autonomy in the so-called “iron cage.” Today, it entails countering atomization and restoring responsibility and connection in ever more fluid and automated workplaces.

Let me suggest three ways to do so that might also score well in a Scrabble match: Countering the corruption of consciousness, community, and cosmopolitanism by a blind faith in instrumentality. Making the case that consciousness is more than a state of mindful equanimity in the present; it is a consideration of the consequences of one’s work in a broad space, and over a long time.  Making the case that a community is not just a tribe that reinforces our performances; it is a group that commits to our well being and learning. Making the case that cosmopolitanism is not an elite identity; it is an attitude of curiosity about what lies beyond the boundaries of our territories, cultures, and faiths.

Once they stop having to be useful, the humanities become truly meaningful. Only that will allow team human to catch up with team machine. But neither, ultimately, must get too far ahead or we will lose a struggle that keeps us human and makes societies prosper. Sometimes it is useful to move fast and break things. Other times it is wise to move slow and heal people.

How My Company Created an Apprenticeship Program to Help Diversify Tech

Deanna Kelly/Getty Images

Despite recent efforts to increase diversity in tech, the hiring and retention rates of underrepresented groups in the industry remain abysmal. Even Facebook, with billions in cash, has only been able to increase their number of women employees from 31% to 36% over the last five years.

At Treehouse, an online school that helps companies hire developers and designers, we’re seeing the same problem. When I took a look at my workforce two years ago, I saw that I hadn’t created a diverse team. Even though we were following the typical playbook — posting open positions on job boards that specialize in attracting candidates from underrepresented groups, sponsoring events, giving scholarships, and training our employees on inclusion and hidden bias — we weren’t seeing progress.

In order for our team to match the diversity of America, we’d need 13.4% black, 1.3% Native American, 18.1% Latinx, and 50% women employees. We were nowhere near those numbers, and I believed it was a moral and business imperative to change my company.

I first needed to see what we were missing. I interviewed more than 50 people from underrepresented groups who have made it in the tech industry, asking them to help me understand why they weren’t applying for my open tech jobs. They were kind enough to be blunt: “My community does not trust companies that are majority white and male. We do not see people like us succeeding in those companies. Why would we apply for your jobs?”

I dug into the numbers on technical roles. U.S. companies are failing to hire black, Latinx, and women Computer Science graduates. And research shows that once women and people of color join tech companies, retention rates are much lower than that of white men, often due to bad treatment in the workplace. Women leave tech companies twice as fast as men do.

Based on my interviews and research I learned four fundamental things:

Underrepresented groups are not generally aware that they could get high-paying jobs in tech and that they don’t need a college degree to do this. This is because very few, if any, people in their community are working and succeeding in tech, so they are not encouraged to seek this opportunity.
The median household income of black families in the U.S. is 39% less than that of white families; for Latinx families, it’s 27% less than white families. This reality makes it more difficult, even impossible, to take time off from one’s job, pay for childcare, and earn a Computer Science degree or attend a coding bootcamp.
Trust between underrepresented groups and tech companies is extremely low, so even if there are job openings, many won’t apply.
Even if people from underrepresented groups acquire the right skills and apply for tech jobs, many companies still won’t consider them for an interview if they don’t have a Computer Science degree.

To address some of these issues, my company decided to create a pilot apprenticeship program to create and grow a sustainable diverse talent pipeline separate from that of college graduates.

In January 2017 we partnered with Colleen Showalter from the local Boys and Girls Club (BGC) in Portland, Oregon, and asked if they would help us recruit new talent, ages 18 and above, from underrepresented groups. We said that we were looking for hard-working individuals with a high school diploma, whom we could train on all the hard skills necessary to become a software engineer and then hire as paid apprentices.

Unlike tech companies, BGC is a trusted organization within the underrepresented community. They recruited a group of 30 individuals, ages 18-20, who expressed interest in the program. We selected 15 people from that group who demonstrated strong work ethic, grit, and excitement for the program. We then enrolled them in online courses teaching necessary job and technical skills, like computer science fundamentals, complex problem solving, group collaboration, agile methodology, effective written communication, and so on. We mentored and supported them over six months, as they completed their courses.

Five out of the 15 participants successfully completed the training and were hired as apprentices at Treehouse and two other hiring partner companies in Portland (Nike and InVision). The 10 that didn’t complete the program returned to their pre-program jobs. For the five successful students, we created a detailed, customized six-month on-boarding program for ourselves and the hiring partners, which was designed specifically for underrepresented people of color and women who had not earned a CS degree and had no previous tech industry experience.

The program recommended soft-skills training, daily and weekly plans to achieve technical milestones, clear expectations on their output and performance, and daily video calls to gauge happiness, give encouragement, and deliver feedback. Apprentices were paid a minimum of $15 per hour for 40 hours per week, for a period of three months, and providing they met the specified requirements, they were converted to an annual salary of at least $55k + full medical and dental benefits.

We also created a detailed six-month mentorship program for the hiring partner companies. This gave managers instructions on how to assign appropriate mentors for each employee and offered mentors a few resources: diversity and inclusion training, detailed daily and weekly plans for working with apprentices, specific guidelines for measuring the success of apprentices (as their progress would not necessarily mirror that of CS graduates). At first we were concerned that it would be difficult to recruit mentors because of the extra workload. But we actually ended up with too many volunteers.

The results of this pilot were overwhelmingly successful. Four of the five apprentices have successfully converted from hourly pay to salary plus benefits, and they are all still successfully employed with the hiring companies. The feedback from employers has been positive.

Of course, there were some areas that we’ll continue to work on and improve for the next pilot program. For example, we found that many participants felt pressure to join the program so they could improve their family’s income. But without a real passion for tech, they wouldn’t successfully convert from apprentice to salaried developer. In the future, we will screen participants to make sure they truly want a career in tech, not just a higher salary.

We also realized we need to provide access to laptops and broadband. And we learned that we need to offer more equity, diversity, and inclusion training to the company partners, as more hiring managers were eager to participate than we expected. In future pilots, we will also be increasing the length of diversity and inclusion training to an eight-week program.

We believe that investing in our local community is the moral thing to do, but what’s the cost and ROI of program like this? Let’s say you are hiring 10 developers and using a combination of an internal technical hiring team and an outside recruiter to fill those positions. Using standard industry benchmarks as inputs for compensation and time to interview and onboard, it’s going to cost you around $2.046M to source, hire, onboard, and then compensate that cohort of developers for one year (who likely are not from underrepresented groups). However, if you invest in creating talent, these same costs would only amount to $723k. That’s a saving of $1.323M or an ROI of 894%, and you’ll create a diverse team, which is proven to generate more profit.

The early results of our internal pilot program were so encouraging that other technology executives asked me to install a similar program for them. We are rolling out important changes to the program for future cohorts, and installing them at Airbnb, Nike, Mailchimp, HubSpot, Acquia, InVision, MINDBODY, Adobe, and Chegg.

We are still learning, iterating, and updating our solution. The systemic challenges we’re all experiencing around creating diverse teams still exist. But this is the beginning of a viable alternative solution to the historic lack of diversity in tech.

How a German Manufacturing Company Set Up Its Analytics Lab

PM Images/Getty Images

Over the past few years, most businesses have come to recognize that the ability to collect and analyze the data they generate has become a key source of competitive advantage.

ZF, a global automotive supplier based in Germany, was no exception. Digital startups had begun producing virtual products that ZF did not know how to compete against, and engineers in logistics, operations, and other functions were finding that their traditional approaches couldn’t handle the complex issues they faced. Some company executives had begun to fear they were in for their own “Kodak moment” – a fatal disruption that could redefine their business and eliminate overnight advantages accumulated over decades. With automotive analysts forecasting major changes ahead in mobility, they began to think that the firm needed a dedicated lab that focused entirely on data challenges.

But how?

At the time one of us, Niklas, a data scientist for ZF, was pursuing a PhD part-time at the University of Freiburg. Niklas took the first step and recruited his advisors at the university, Dirk Neumann and Tobias Brandt, to help them set up a lab for the company. This gave ZF access to top-notch expertise in data analytics and the management of information systems.

The hardest part was figuring out how the lab would work. After all, industrial data laboratories are a fairly new phenomenon– you can’t just download a blueprint. However, after a number of stumbles, we succeeded in winning acceptance for the lab and figured out a number of best practices that we think are broadly applicable to almost any data lab.

Focus on the Right Internal Customers

ZF had dozens of departments filled with potentially high-impact data-related projects. Although we were tempted to tackle many projects across the entire company, we realized that to create visibility within a 146,000-employee firm, we had to focus on the most promising departments and projects first.

But how would we define “most promising”? As the goal of the data lab is to create value by analyzing data, we initially focused on the departments that generate the most data. Unfortunately, this didn’t narrow it down a whole lot. Finance, Logistics, Marketing, Sales, as well as Production and Quality all produced large amounts of data that could be interesting for data science pilot projects.

However, we knew from experience that the lowest hanging fruits for high-impact projects in a manufacturing company like ZF would be in Production and Quality. For years, ZF’s production lines had been connected and controlled by MES and ERP systems, but the data they generated had yet to be deeply tapped. We decided, therefore, to begin by concentrating on production issues, such as interruptions, rework rates, and throughput speed, where we could have an immediate impact.

Identifying high-impact problems

Next, we selected those projects within Production and Quality that promised the highest-value outcomes. Our experience with the first few projects provided the basis for a project evaluation model, that we have continued to refine. The model contained a set of criteria along three dimensions that helped us to rank projects.

The problem to be solved had to be clearly defined. We could not adopt an abstract aim such as “improve production.” We needed a clear idea of how the analysis would create business value.
Hard data had to play a major role in the solution. And the data had to be available, accessible, and of good quality. We needed to shield the team from being flooded by business intelligence reporting projects.
The team had to be motivated. We gave project teams independence in choosing how they solved the problems they took on. And while we made the budget tight enough to enforce focus, we made sure that it was not so tight that the team couldn’t make basic allocation decisions on its own. To sustain motivation and enthusiasm, we priotitized projects that could be subdivided into smaller but more easily achieved goals.

While we eventually found it useful to assign a particular person to manage relations with the rest of the company, we kept the whole lab involved in project selection as the number of people working in the lab grew. This kept everyone informed, gave them a greater sense of personal responsibility, and implicitly expressed management’s appreciation for their professional judgment.

Execution

The key risk was that the team would get lost in optimizing minor nuances of models and methods instead of solving the major problem. To avoid this, we usually limited the execution phase to three months, and gave the team the right to cancel its engagement.

This power turned out to be a game changer. Giving the team (including the domain expert) a “nuclear option” made them much more focused and goal-oriented. Once we put this rule in place, the number of change requests from the internal client dropped and the information initially provided tended to be more accurate and complete than before.

Of course, a team couldn’t cancel a project for arbitrary reasons. It needed to justify its decision, specifying conditions uncer which the project could be reopened. And while cancellations are contentious, they are sometimes necessary to free resources and to enforce progress toward a meaningful goal. In fact, introducing the ability to cancel projects actually increased the number of successfully completed projects.

Although a single team can work on multiple projects concurrently, particularly as waiting for responses from the client department can lead to delays, we generally found it best for the team to work on a single project at a time. We found that downtimes were better used by team members to learn new analytics methods and techniques, which continued to advance at a rapid pace.

We kept our internal customer up to date on our progress through regular reports and when possible by including their domain expert in the project team. If we could not so do, we looked for an arrangement – such as a weekly meeting – that allowed us to contact the domain expert directly without having to pass through gatekeepers.

Key Success Factors

Beyond gaining a general understanding of the data lab’s work as a three-stage process, we learned other lessons too. In particular, we found three more ingredients to be crucial to the data lab’s success:

Executive support. The confidence that the technology executive team placed in us was crucial to our success. Fortunately, they don’t seem to regret it: “Giving the data lab a great freedom to act independently, to try ideas and also to accept failures as part of a learning process, required trust. But the momentum it created is something we do not want to miss”, said Dr. Jürgen Sturm, Chief Information Officer.
The perspective of an outside authority. In this case, data scientists from the University of Freiburg, made a huge difference to the lab’s success. As Andreas Romer, ZF’s Vice President for IT Innovation, put it, “We no longer consider innovation to be an internal process at ZF. To safeguard our future success, we must look beyond the confines of our company, build up partnerships to learn and also to share knowledge and experiences.”
Domain experts. While data scientists brought knowledge of analytic methods and approaches to the project, their access to domain experts was essential. Such experts needed to be closely involved in answering domain-related questions that come up once the team is deeply engaged with the problem. In our experience, the capacity and availability of domain experts is the most common bottleneck blocking a data analytics project’s progress.

Problems solved

Three years on, we can say with confidence that the ZF Data Lab is a valuable addition to the company. With this dedicated resource, ZF has been able to solve problems that had stumped the company’s engineers for years. Here are two examples:

Broken grinding rings. A key source of stoppages in production line machinery, a breakdown can create a mess that may take hours to clean up. An internal client wanted to develop an early warning system that could indicate the probability of a future ring breakdown, but they had messy data, a weak signal (unclear data), and a highly unbalanced ground truth (because breakdowns happen only occasionally). Despite those limitations, we were able to create an algorithm that could detect imminent breaks 72% of the time – a far cry from five-decimal perfection but still enough to save the company thousands.
High power demand charges. Managing energy units to regulate energy demand at times of peak use is an effective way to reduce costs. Our goal was to develop an automated data-driven decision-making agent that provides action recommendations with the objective to lower load peaks. Working closely with the energy department, we were able to develop a working prediction model to avoid those high-demand surcharges. Following the model’s recommendations should reduce the peak load by 1-2 Megawatts, worth roughly $100k – $200k per year.

After growing for three years, the ZF Data Lab has become a kind of specialized R&D function within the company. It is a melting pot of ideas and technologies, producing and evaluating proofs-of-concept, and discarding approaches that don’t quite work. In the last analysis, the data lab is not only there to solve problems, but to help answer the biggest Big Data question of all: how will our company compete in this increasingly digital world?

To Control Health Care Costs, U.S. Employers Should Form Purchasing Alliances

Tim Robberts/Getty Images

When it comes to health care costs, America’s employers are at a crossroads. Competing for scarce labor in a tight market, they will have trouble continuing to shift medical bills onto employees as they have for several decades.

That means that to control costs going forward, employers may have to confront the true underlying causes of rising health care expenditures: high prices and health care inefficiencies. To address these challenges, they will have to band together in purchasing coalitions that give them the local market power to force health systems to reform.

Employers are the largest single provider and purchaser of health insurance in the United States, covering over 150 million workers and their dependents and purchasing 34% of all health care dispensed in the country. As a potential force for change, only the U.S. government can rival America’s business community.

And in recent years, employers have enjoyed some success in controlling rising health care costs. Their premiums have been increasing 3% to 5% annually, rather modest by historic standards. As a percent of workers’ compensation, employers’ health care spending has held steady at between 8% and 9% since 2010. Much of this success seems attributable to the spread of high-deductible health plans (HDHPs), which have shifted more of the costs of care onto employees. The proportion of workers with HDHPs (deductibles of more than $1,300/$2,600 for an individual/family) increased from 6% to 22% between 2006 and 2018. High deductibles have the dual effect of reducing workers’ use of services and employers’ liability for the services employees use.

Insight Center



The Future of Health Care
Sponsored by Medtronic

Creating better outcomes at reduced cost.

So what’s the problem? There seems growing nervousness among employers that they’ve pushed high deductibles about as far as they can. Workers’ increasing out-of-pocket costs are creating widespread discontent with the underlying costs of care — a problem largely driven by the high prices charged to private payers for health services and pharmaceuticals. Data from the Commonwealth Fund’s biennial survey of the American public shows that the percent of U.S. workers who are underinsured — face out-of-pocket health care expenses greater than 10% of their income excluding premiums — increased from 10% in 2003 to 24% percent in 2016. Between 2011 and 2017, employees’ premiums and deductibles grew faster than their median income. Beyond this, studies clearly show that when workers face high upfront payments, they frequently skip services, some of which are critical to their long-term health and productivity, a pattern that must worry responsible employers.

Add to this picture the increasingly competitive labor market — which limits the tools companies can use to constrain health spending — and it becomes clear that employers may have to find new ways to tame the health care cost tiger in the future. They may have to address the underlying reasons for rising health care premiums, rather than just shifting more of those expenditures off their own books.

Those fundamental reasons are varied and complex but at least two stand out. The first is that health care providers charge employers very high prices — way higher than those paid by public insurers like Medicare and Medicaid. The second is that our health care system is highly inefficient and wasteful. It has enormous administrative costs. Care is fragmented and uncoordinated. We have too many high-priced specialists and not enough high-quality primary care to keep patients out of emergency rooms and hospitals when they could be cared for in less expensive (and dangerous) settings. In other words, employers need to get better deals on prices and remake our health care system while they’re at it.

Employers are not new to this game. For decades, large sophisticated companies have undertaken pioneering experiments with reshaping the health care system. As far back as the early 1990s, Pitney Bowes focused on patient education and consumerism and prevention and care management to slow cost growth. Companies such as Boeing have experimented with direct purchasing of health care from providers, securing better prices, and eliminating the administrative costs of insurers. Other employers such as Walmart have cut deals to send their high-end elective procedures (e.g., open-heart surgery, hip and knee replacements) to centers of excellence that offered lower prices and higher quality. Employers have instituted wellness programs in the (now disappointed) hopes that health maintenance could lower costs of care. And companies have come together in regional coalitions such as the Pacific Business Group on Health and the Midwest Business Group on Health for the purpose of sharing lessons on how to become better health care purchasers.

The latest venture in employer health innovation is, of course, the alliance of Amazon, Berkshire Hathaway, and JPMorgan Chase. The as yet unnamed joint venture, led by the highly respected Dr. Atul Gawande, is promising to solve the health care conundrum for its parent companies and perhaps for the nation as a whole.

The fact is, however, that until employers switched to high-deductible plans, they enjoyed relatively little success in restraining health spending. This disappointing record reflects persistent challenges to their cost-control efforts.

The first challenge is lack of purchasing power. All health care is local, and efforts to negotiate better prices and reform health care delivery depend on an employer’s ability to force price concessions and behavior change from local physicians and health care institutions. Collectively, employers may constitute an important share of health providers’ market. But individually, with the exception of a few companies in a few markets, such as Boeing and Amazon in Seattle, no one employer has enough leverage to wrangle price concessions from area doctors and hospitals or induce them to reshape the way they do business. This is true even for large national companies because their aggregate workforce is spread across tens or hundreds of localities.

Efforts to form purchasing coalitions in local markets have had modest impact at best because employers have so little else in common and because antitrust laws limit their ability to collaborate. The growing consolidation among providers — 90% of metropolitan areas have highly concentrated hospital markets and 65% have highly concentrated specialist physician markets — also works to employers’ disadvantage.

A second challenge facing employers is lack of sophistication as health care purchasers. Medicine is complicated, and while there are a handful of large employers such as Comcast or Walmart with the funds and motivation to hire sophisticated health benefits specialists, there are 7 million to 8 million mid-size and small employers who have their hands full just managing their core business in turbulent times. Even if they had the leverage to demand delivery system reforms from providers, most CEOs and CFOs largely lack the time and patience to grasp the complex, non-intuitive, and often experimental interventions involved: accountable care organizations, value-based purchasing, outcomes based pharmaceutical pricing and so on. Better to raise deductibles and move on.

A third challenge is that when employers try to reform health care, they can easily alienate employees. To get better health care deals, employers often have to channel their workers to a select group of providers who offer lower prices and/or better quality. This can sometimes mean bypassing prominent but highest-priced local facilities and specialists where workers are already getting their care or want to if they ever need it — for example, the Partners HealthCare system in Boston, Memorial Sloan Kettering in New York City, and MD Anderson Cancer Center in Houston. In tight labor markets, the last thing employers want to do is to get between workers and their doctors.

To achieve the kind of gains in controlling health care costs that employers want, they will have to get bigger and smarter in the future.

They will need to band together in local purchasing alliances, come to agreement on common features of health insurance products, and then, working with local insurers, wrangle price and delivery concessions from local providers. This will likely require newfound willingness on the part of employers to surrender the freedom to tailor each insurance product to their own specific preferences. It will also require that, working together, employers immerse themselves in the complex details of reforming health care delivery systems so that they push insurers to insist on greater provider accountability for cost and quality, better primary care and prevention, improved care coordination, reduction in administrative costs, and a variety of other nitty gritty health care reforms.

Employers will not be able to do this without help from government. They may need antitrust allowances to band together for joint purchasing of care. They will also need state and federal antitrust authorities to break up increasingly dominant local provider coalitions. They will certainly want to strongly encourage federal and state authorities to pursue value-based payment programs for federally insured populations in the hope that employed populations will benefit from these reforms as well. Some employers may even decide — despite innate opposition to government regulation — that the only way for them to stay in the business of providing insurance to employees will be to have government regulate health care prices in their states. This is the tactic that most industrialized countries use to keep health care affordable for their populations.

The alternative to these fairly radical changes in employer behavior is continuing the hollowing out of employer-sponsored insurance. Aside from the pain this will inflict on workers and their families, this trend could cause the American public generally to lose faith in our current system of employer-sponsored insurance, and open the way politically for alternatives, including government-provided coverage.

How Masculinity Contests Undermine Organizations, and What to Do About It

Mark Leech/Offside/Getty Images

From Uber to Nike to CBS, recent exposés have revealed seemingly dysfunctional workplaces rife with misconduct, bullying, and sexual harassment. For example, Susan Fowler’s 2017 blog about Uber detailed not only her recollections of being repeatedly harassed, but what she described as a “game-of-thrones” environment, in which managers sought to one-up and sabotage colleagues to get ahead. A New York Times investigation described Uber as a “Hobbesian environment…in which workers are pitted against one another and where a blind eye is turned to infractions from top performers.”

Why do companies get caught up in illegal behavior, harassment, and toxic leadership? Our research identifies an underlying cause: what we call a “masculinity contest culture.” This kind of culture endorses winner-take-all competition, where winners demonstrate stereotypically masculine traits such as emotional toughness, physical stamina, and ruthlessness. It produces organizational dysfunction, as employees become hyper competitive to win.

Masculinity Contest Cultures

We surveyed thousands of workers in the U.S. and Canada from different organizations. Respondents rated whether various masculine qualities were highly prized in their workplace; they also reported on other organizational characteristics and their personal outcomes. Four masculine norms, which together define masculinity contest culture, emerged as highly correlated with each other and with organizational dysfunction:

“Show no weakness”: a workplace that demands swaggering confidence, never admitting doubt or mistakes, and suppressing any tender or vulnerable emotions (“no sissy stuff”).
“Strength and stamina”: a workplace that prizes strong or athletic people (even in white collar work) or those who show off their endurance (e.g., by working extreme hours).
“Put work first”: a workplace where nothing outside the organization (e.g., family) can interfere with work, where taking a break or a leave represents an impermissible lack of commitment.
“Dog eat dog”: a workplace filled with ruthless competition, where “winners” (the most masculine) focus on defeating “losers” (the less masculine), and no one is trusted.

These norms take root in organizations because behaving in accordance with them is what makes someone a “man.” As phrases like “man up” illustrate, being a man is something men must prove — not just once, but repeatedly. In many cultures around the world, someone becomes a “man” by behaving in ways that conform with cultural beliefs about what men are like — dominant, tough, risk taking, aggressive, rule breaking.

And it doesn’t take much to make men feel like “less of a man.” Men react defensively when they even just think about job loss, or receive feedback suggesting they have a “feminine” personality.

What all of this means is that masculinity is precarious: hard won, and easily lost. And the need to repeatedly prove manhood can lead men to behave aggressively, take unwarranted risks, work extreme hours, engage in cut-throat competition, and sexually harass women (or other men), especially when they feel a masculinity threat.

At work, this pressure to prove “I have what it takes” shifts the focus from accomplishing the organization’s mission to proving one’s masculinity. The result: endless “mine’s bigger than yours” contests, such as taking on and bragging about heavy workloads or long hours, cutting corners to out-earn others, and taking unreasonable risks either physically (in blue-collar jobs) or in decision-making (e.g., rogue traders in finance). The competition breeds unspoken anxiety (because admitting anxiety is seen as weak) and defensiveness (e.g., blaming subordinates for any failure), undermining cooperation, psychological safety, trust in coworkers, and the ability to admit uncertainty or mistakes. Together this creates miserable, counterproductive work environments that increase stress, burnout, and turnover.

Masculinity contests are most prevalent — and vicious — in male-dominated occupations where extreme and precarious resources are at stake (fame, power, wealth, safety). Think about finance and tech startups, where billions of dollars are quickly made or lost; surgery, where high-stakes operations leave no room for error; and military and police units, where risky jobs are performed under strict chains of command.

Where does this leave women? Like everyone else, women must try to play the game to survive, and the few who succeed may do so by behaving just as badly as the men vying to win. But the game is rigged against women and minorities: Suspected of not “having what it takes,” they must work harder to prove themselves while facing backlash for displaying dominant behaviors like anger and self-promotion. Women and minorities thus face a double-bind that makes them less likely to succeed; they may find it easier to survive by playing supporting roles to men who are winning the contest. 

The business case against masculinity contests

Organizations rely on cooperative teamwork to succeed. But masculinity contests lead people to focus on burnishing their personal image and status at the expense of others, even their organizations. Our research documents numerous negative consequences that harm the bottom line and put the organization’s effectiveness and reputation at risk. Organizations that score high on masculinity contest culture tend to have toxic leaders who abuse and bully others to protect their own egos; low psychological safety such that employees do not feel accepted or respected, feeling unsafe to express themselves, take risks, or share new ideas; low work/family support among leaders, discouraging work-life balance; sexist climates where women experience either hostility or patronizing behavior; harassment and bullying, including sexual harassment, racial harassment, social humiliation and physical intimidation; higher rates of burnout and turnover; and higher rates of illness and depression among both male and female employees.

These problems create both direct costs (through turnover and harassment lawsuits) and indirect costs (through decreased innovation due to low psychological safety). Put simply, masculinity contest cultures are toxic to organizations and the men and women within them. In extreme cases, such as Uber, the pressure cooker explodes, severely damaging or even destroying the organization.

Changing masculinity contest cultures

Despite being toxic, masculinity contest cultures persist for two reasons: (1) the association between toxic masculinity and success is so strong that people feel compelled to keep playing the game, despite the dysfunctional behavior it produces, and (2) questioning the masculinity contest marks one as a “loser,” which disincentivizes people from pushing back. Dropping a diversity initiative onto these types of workplaces is unlikely to create meaningful change. In fact, current interventions, like those to prevent sexual harassment, typically fail or even backfire in these environments (by creating more harassment). Real change requires shutting down this game.

To accomplish this, organizations need to perform deeper, more committed work to examine and diagnose their cultures. These efforts must be led by those who have the power to spark serious reform. It is crucial to generate awareness of the masculinity contest and its role in creating organizational problems. For instance, people tend to attribute sexual harassment to a “few bad apples,” ignoring how an organization’s culture unleashed, allowed, and may have even rewarded the misconduct. When organizations do not tolerate bullying and harassment, the bad apples are kept in check and good apples do not go bad.

Two specific actions are a good place to start:

Establish a stronger focus on the organization’s mission. Current trainings backfire, in part, because they focus on compliance and “what not to do,” are often framed as trying to “make things better for the women and minorities” rather than for everyone, and seem unconnected to the organization’s core mission. Effective interventions require authentic and meaningful connections to core organizational values and goals.

For example, researchers have documented how an energy company undermined masculinity contest norms on oil rigs through a safety intervention. The bottom line demanded reform: oil rig disasters cost lives and money, environmental destruction, legal liability, and severe reputational damage. Leaders convinced workers that increased safety was central to the mission; and they monitored and rewarded desired behavior change. Workers were rewarded for voicing doubts or uncertainties about a procedure (rather than “showing no weakness”), for listening to each other (rather than obeying the “strongest” alpha male), for valuing safety and taking breaks (rather than “putting work first”), and for cooperating with and caring for coworkers (rather than “dog eat dog” competition). The need to prove manhood proved incompatible with the new mission-based rules. Not only were accidents and injuries reduced, but so was bullying, harassment, burnout, and stress.

Organizations can leverage other core goals to motivate reform. Given the inherent dysfunctionality masculinity contest cultures create, chances are that almost any mission-related reform can help. For example, research demonstrates a common characteristic among innovative teams: psychological safety. Team members know that they can raise questions or voice doubts without eliciting ridicule or rejection. An initiative to foster innovation via greater psychological safety would naturally dampen the masculinity contest. As a by-product, the work environment should become more hospitable and inclusive toward women and minorities; after all, whose ideas are most often summarily ignored or dismissed in masculinity contest cultures?

Dispel misconceptions that “everyone endorses this.” People fail to question masculinity contest norms lest they be tagged as a whiny, soft loser. As a result, everyone goes along to get along, publicly reinforcing norms they privately hate — people stay late just to be seen as putting work first, or laugh at a joke they actually think is offensive. Because people publicly uphold the norms, it appears as though everyone endorses them. Research has shown that people in masculinity contest cultures think their coworkers embrace these norms when in fact they do not, breeding pervasive but silent dissatisfaction alongside active complicity as people stay quiet to prove they belong.

Leaders can remedy this misperception by publicly rejecting masculinity contest norms and empowering others to voice their previously secret dissent. But they also need to walk the talk by changing reward systems, modeling new behavior, and punishing the misconduct previously overlooked or rewarded. Leaders also need to ensure that people who speak up are no longer punished or retaliated against for doing so, either formally (e.g., with job consequences) or informally (e.g., by reputation and ostracism).

When masculinity contest cultures become “the way business gets done,” both organizations and the people within them suffer. Your organization may have a masculinity contest culture if, for example, expressing doubt is forbidden, “jocks” are preferred even though athleticism is irrelevant to job tasks, extreme hours are viewed as a badge of honor, or coworkers are treated as competitors rather than colleagues. Solving the problem requires meaningful commitment to culture change — to creating a work environment in which mission takes precedence over masculinity.

Race Issues

How does race affect your workplace? In this episode of HBR’s advice podcast, Dear HBR:, cohosts Alison Beard and Dan McGinn answer your questions with the help of Tina Opie, a management professor at Babson College. They talk through what to do when your company’s board is not diverse, promotions favor some people more than others, or you want to have more conversations about race at the office.

Download this podcast

Listen to more episodes and find out how to subscribe on the Dear HBR: page. Email your questions about your workplace dilemmas to Dan and Alison at dearhbr@hbr.org.

From Alison and Dan’s reading list for this episode:

HBR: Diversity and Authenticity by Katherine W. Phillips, Tracy L. Dumas, and Nancy P. Rothbard — “Simply hiring members of a minority group won’t ensure that they feel comfortable or equipped to build the relationships necessary for advancement. And as companies invest in mentorship and sponsorship programs, making these relationships flourish among workers of differing races may require special effort.”

HBR: How Managers Can Promote Healthy Discussions About Race by Kira Hudson Banks — “Many white people may avoid conversations about race out of fear of ‘saying the wrong thing.’ And many people of color in predominantly white companies may avoid these conversations out of fear of being seen as a complainer — or worse. But pretending the elephant in the room isn’t there won’t make it go away.”

HBR: A Question of Color: A Debate on Race in the U.S. Workplace by David A. Thomas and Suzy Wetlaufer — “You can’t underestimate the power of professional networks, because when they are positively focused, you no longer feel alone or isolated. You are connected with people of power in the organization in a way you have never been before. Instead of always feeling like an outsider, you feel as if you belong. You are not alone, and that can be tremendously helpful both personally and professionally.”

HBR: The Costs of Racial “Color Blindness” by Michael I. Norton and Evan Apfelbaum — “Rather than avoiding race, smart companies deal with it head-on—and they recognize that ‘embracing diversity’ means recognizing all races, including the majority one, to avoid showing preference or creating a backlash. For example, Time Warner’s annual diversity summit isn’t just for people of color (or women)—it’s populated by white males, too. Talking about race can feel awkward, but over time more companies will discover that doing so is usually better than pretending it doesn’t exist.”

We Need to Talk More About Mental Health at Work

Sabine Scheckel/Getty Images

Alyssa Mastromonaco is no stranger to tough conversations: she served as White House deputy chief of staff for operations under President Obama, was an executive at Vice and A&E, and is Senior Advisor and spokesperson at NARAL Pro-Choice America. So when Mastromonaco switched to a new antidepressant, she decided to tell her boss.

“I told the CEO that I was on Zoloft and was transitioning to Wellbutrin,” Mastromonaco said. “I can react strongly to meds, so I was worried switching would shift my mood and wanted her to know why. I talked about it like it was the most normal thing in the world —it is!”

Her boss was supportive. “You got it,” she said.

When Mastromonaco goes to work, she and her mental health struggles do not part ways at the door. “You want me,” she said, “you get all of me.” Mastromonaco brings tremendous talent to her workplace — but she also brings her anxiety. The same is true for high-performing employees everywhere: one in four adults experiences mental illness each year and an estimated 18% of the US adult population have an anxiety disorder. And yet we’re loath to talk about mental health at work. If we’re feeling emotional at work, our impulse is to conceal it — to hide in the bathroom when we’re upset, or book a fake meeting if we need alone time during the day. We’re hesitant to ask for what we need — flex time, or a day working from home — until we experience a major life event, like a new baby or the illness of a parent. We would more likely engage in a trust fall with our boss than admit that we have anxiety.

Mental illness is a challenge, but it is not a weakness. Understanding your psyche can be the key to unleashing your strengths — whether it’s using your sensitivity to empathize with clients, your anxiety to be a more thoughtful boss, or your need for space to forge new and interesting paths. When we acknowledge our mental health, we get to know ourselves better, and are more authentic people, employees, and leaders. Research has found that feeling authentic and open at work leads to better performance, engagement, employee retention, and overall wellbeing.

Still, less than one third of people with mental illness get the treatment they need, and this comes at a cost — to people and to companies. Failure to acknowledge an employee’s mental health can hurt productivity, professional relationships, and the bottom line: $17-$44 billion is lost to depression each year, whereas $4 is returned to the economy for every $1 spent caring for people with mental health issues.

So what needs to change? In the twenty-first century, human capital is the most valuable resource in our economy. And though much has been done (rightly) to promote diversity at work, there’s a giant hole when it comes to understanding how temperament and sentiment play into the trajectory of success. As we recognize neurological and emotional diversity in all of its forms, workplace cultures need to make room for the wide range of emotions we experience. Professional support needs to get better. We need to have the option to ask for help, and feel safe doing so (depression screenings are free under the Affordable Care Act, and some companies offer an Employee Assistance Program). In short, we need more flexibility, sensitivity, and open-mindedness from employers. The same treatment and attention they’d give to a broken bone or maternity leave. We’re not there yet, but some companies are trying to bring conversations about mental health to the forefront.

EY (formerly Ernst and Young) launched a We Care program two years ago to educate employees about mental health issues, encourage them to seek help if they need it, and be a support to colleagues who might be struggling with mental illness or addiction. They started the program out of a demonstrated need. “Our Employee Assistance Program was starting to hear more conversations about anxiety,” said Carolyn Slaski, EY Americas Vice Chair of Talent. “They told us that it was very taboo — something that people don’t normally talk about — but they were seeing more activity, so we decided to schedule a session to talk about anxiety. Just talk about it and see what would happen.”

Since the advent of the We Care program, 2000 EY employees have attended these sessions, which always have a senior-level sponsor and a mental health professional on hand. Someone in leadership kicks it off by sharing their story. This sends the message that anxiety is not toxic and attendance is not a career-dampener.

The company also has an employee assistance hotline that offers confidential support — calls related to anxiety have increased 30% over the last two years. “You have to notice first if someone is struggling,” said Slaski, “and ask them if they’re okay. Learn how to listen to their concerns, and then act. Our company has 47,000 US employees, and 250,000 globally. If I can get my team comfortable just noticing when someone has an issue, then there is so much more we can do for them. These are people reaching out for help. We want to help. We don’t want to have a stigma around it.”

Other companies, like Michigan-based furniture store, Herman Miller, offer free onsite counseling sessions to employees and their families, and courses on mental health first aid that teach them how to recognize signs of mental illness in others. The goal is to empower people to achieve their optimal state of well-being.

What organizations like EY and Herman Miller realize is that, given the right support, employees who struggle with their mental health can do great work. Most people who suffer from chronic anxiety or depression are excellent at faking wellness. We put on our makeup, get dressed, and show up on time. But we never know when an attack might be around the corner. This is why a work environment that is open and understanding is so important. Anxiety is a lingering expectation that something bad is going to happen, and if we don’t talk about it, it’s harder to recognize our triggers and learn healthy ways to cope. But when we do talk about it, we can actually teach ourselves to harness it in ways that play to our strengths.

Christina Wallace is a Harvard Business School graduate, a three-time startup founder, and an accomplished executive and creator of an innovative STEM education program. She also has panic anxiety. When asked if she ever considers her anxiety a strength, she didn’t hesitate to answer, “Absolutely.”

Christina had severe childhood trauma, and has done a lot of work to manage the after effects. “Even still,” she says, “situations where I feel like I can’t trust the other person, or the rug has been pulled out from under me, throw me into a fight-or-flight mode.” For her, this means panic attacks and crippling anxiety. To cope, Christina has taught herself to communicate openly with her managers and colleagues. For example, she has asked both her managers and the people she manages to give her written feedback on important projects before they meet in person. This way she has time to process it and prepare instead of feeling blindsided.

According to feedback from direct reports, Christina is an incredible manager. Because she has openly acknowledged her anxiety, she has learned not only how to manage it, but also how to communicate and share her needs —  a skill that helps her stay attuned to the emotional needs of others, and navigate difficult situations with grace and ease. “I’m much more aware of how to help my team show their best selves,” she said.

The good news is that times are changing, and people like Christina, along with the millions of others who struggle with mental illness, are more likely to get the help they need at work than ever before. Stew Friedman, professor at the Wharton School of Business and founding director of the Wharton Leadership Program, says “the next great sort of liberation movement in our society is about mental illness.” He sees shoots of awakening in corporate America. “Look at the huge growth in wellbeing research, practice in the private sector, and society at large. That’s one really good indicator of change.” It’s much more understood and accepted that people have emotional and mental health needs. Yet Friedman still acknowledges that there are costs to the digital revolution and how it’s affecting communication, identity, and the amount of stress we regularly experience. “There are trends that are incredibly worrisome. Rates of suicide, depression, anxiety, and drug use are all on the rise. So, our response is clearly inadequate.”

Along with employee assistance programs, conversation and education are fundamental if our goal is to increase understanding and reduce the stigma around mental health. Friedman notes the importance of conversation in his own experience: “Twenty years ago, in 1987, I started talking about what it was like to become a father and how that changed my career and my life. It was taboo for a man to talk about children at the Wharton School back then, and it got a lot of attention. I was part of a wave of change. The conversations you instigate and your awareness in choosing topics of discussion are an important piece to the process of change. Openness encourages executives to share more about their own experiences, and that normalizes the experience of others.”

In the spirit of being open, I will share that I cried in many workplace bathrooms as I cycled between anxiety attacks and clinical depression throughout my career in corporate America. It never occurred to me that I could share my struggles or create a schedule that allowed me to manage my anxiety, such as working from home or managing the flow of meetings in a day. So I just quit, over and over again. Now I know that when an employee leaves a job, the typical cost of replacement is three months of salary. Think of what the cost is — for the people and the employer — when a whole slice of the population struggles to express their most basic needs.

The burden of depression and anxiety is shared by all members of a workplace, and it’s a vicious cycle. Change starts with managers and HR professionals recognizing the ambivalence and inner conflict many insanely talented people feel, and doing something about it. Because when people get the space and the support they need, it can change their careers, and their lives.

Sexual Harassment Is Rampant in Health Care. Here’s How to Stop It.

Carmen Martínez Torrón/Getty Images

Many factors make an organization prone to sexual harassment: a hierarchical structure, a male-dominated environment, and a climate that tolerates transgressions — particularly when they are committed by those with power. Medicine has all three of these elements. And academic medicine, compared to other scientific fields, has the highest incidence of gender and sexual harassment. Thirty to seventy percent of female physicians and as many as half of female medical students report being sexually harassed.

As we wrote in a recent New England Journal of Medicine article, “Imagine a medical-school dean addressing the incoming class with this demoralizing prediction: ‘Look at the woman to your left and then at the woman to your right. On average, one of them will be sexually harassed during the next 4 years, before she has even begun her career as a physician’.”

The efforts of many healthcare organizations and medical centers tend to go little further than avoiding litigation. This needs to change. We propose a number of actions institutions must take to eliminate sexual harassment and create a safe environment that allows everyone in the health care workforce to do their best work on behalf of their patients.

Insight Center



The Future of Health Care
Sponsored by Medtronic

Creating better outcomes at reduced cost.

Quantitative and qualitative assessment.  The first step is for healthcare organizations to commit to understanding the problem. They must thoroughly and repeatedly measure the nature, prevalence, and severity of harassment and discrimination. Since this is unlikely to happen spontaneously, boards of directors and trustees should require open reporting of aggregate data, forums where employees can share ideas on how to reduce or eliminate harassment, and tying compensation of executives, deans, and chairs to outcomes.

Organizations should use standardized and validated instruments to survey their employees and do so annually, and anonymously. (You can find one such survey available from NASEM and another from the AAU. Our company, Equity Quotient, also offers one.) Survey data, along with aggregated data on reports of harassment, should be reported throughout the organization. Measurement will allow each organization to ascertain where exactly it needs to improve, test hypotheses and solutions that fit its culture and needs, and track progress.

Policy improvement.  Every health care organization needs to promote a clear, comprehensive policy that conveys a firm commitment to safety, respect, inclusion and equality. It should contain guidelines for standards of behavior, employee reporting of sexual harassment, and institutional responses to offensive or abusive behavior, discrimination, and retaliation. The Association of Title IX Administrators have examples of such policies, as does the National Council of Non-Profits. Organizations can use these for reference, modifying them as appropriate for their own needs.

In addition, secure methods of reporting harassment should be readily available to employees and supported by initiatives to keep the reporting options visible and familiar to the entire community. Targets of harassment should have ready access to counseling and support, even if they choose not to pursue formal reporting processes. These resources should be available outside of the institution itself, to increase the comfort of people reporting harassment and to remove potential biases that may occur when counselors are employed within the same institution where the harassment occurred.

Follow through. Organizations need to pair policy with clear and consistent action. The NASEM report stated, “Too often, judicial interpretation of [anti-discrimination laws] has incentivized institutions to create policies and training on sexual harassment that focus on symbolic compliance with current law and avoiding liability, and not on preventing sexual harassment.”  The key phrase here is “symbolic compliance”: nearly every healthcare organization has a policy, but whether that policy is merely a checkbox or actually functions well in practice is a distinguishing feature of organizations that are serious about the problem.

Human Resources commonly takes the lead in crafting and enforcing a strong policy. HR should be responsible for ensuring, among other things, that leadership has clearly communicated a zero-tolerance position; that employees trust the current procedures; and that reporting mechanisms are easy to understand. A useful list of key questions for leaders to ask themselves about their approach to sexual harassment, from the law firm Cleary Gottlieb, can be found here. It asks, for example, how well senior management communicates its zero-tolerance stance and who should oversee investigations of harassment allegations. While internal processes may provide efficiency, independent external investigations should be undertaken when there is any question about the objectivity of the internal inquiry.

Finally, organizational responses need to be applied with consistency. Victims will only come forward if they feel safe doing so and know their report will result in a rapid, thorough, and fair investigation, and, if misconduct is discovered, that their harassers will be punished, no matter their rank or reputation. Perpetrators must not be allowed to go on “extended leave,” quietly retire, or accept reassignment at another healthcare system through an under-the-table arrangement: all “cover your ass” practices that communicate tolerance of egregious behaviors do nothing to discourage further misbehavior. These provisions will reduce the possibility of retaliation, impediments to professional advancement, and further trauma.

Calculate cost and report.  The economic, reputational, and human costs of harassment are huge. The University of Southern California (USC), for example, has faced allegations of sexual assault among its medical staff, in addition to allegations of sexual assault of patients by a USC staff gynecologist, for which USC recently offered a $215 million settlement to the victims. In another case, despite knowing there had been a $135,000 settlement with a woman who had reported sexual harassment and retaliation by Dr. Rohit Varma in 2003, USC leadership installed the ophthalmologist as dean of the School of Medicine.  Dr. Varma resigned under a cloud less than a year later when leadership, responding to previously undisclosed information, acknowledged it has lost confidence in his ability to lead the school.

Even aside from the impacts of litigation and restitution of harms, the economic, health and psychological consequences of harassment are grave and have reached crisis levels in medicine. Sexual harassment and discrimination undermine women’s physical and mental health, resulting in increased risk for anxiety, depression, burnout, PTSD, and a host of other negative personal and financial consequences. The negative effects of harassment also affect the well-being and productivity of colleagues and entire organizations. In healthcare, this ripple effect is particularly serious, as it may threaten the quality of patient care. Organizational leaders should strive to calculate actuarial costs of sexual harassment in their institutions — in terms of accumulated absences, lost productivity, compromised hiring and retention, legal costs, and reputational harm, and report those costs to their board of directors/trustees. Leaders’ compensation should be tied to decreasing these costs. To the extent possible, executive teams need to make these costs transparent, so that investments in prevention of harassment are understood to be cost effective.

Clearly it is a challenge to put hard numbers on some of these — how do you measure the dollar value of reputational harm? — and indeed no organization that we know of is doing this yet. Nonetheless, such accounting is an essential element of addressing harassment and every health care organization must start making the effort.

Leadership. Harassment thrives in settings dominated by men. Thus it is essential to increase representation of women in leadership roles and assure accompanying equity in salary and power. Among the initiatives that can help are mentorship and sponsorship programs, which are essential to career progression. For example, Drexel University’s Executive Leadership in Academic Medicine (ELAM) program, a year-long fellowship for women leadership in the schools of medicine, dentistry, public health, and pharmacy, provides skills training, mentorship and a network of ELAM graduates who provide support after the completion of the fellowship year. Health care institutions might also take look at successes in other industries; at Eli Lilly, for example, with a mandate from CEO David Ricks, leadership has embraced mentorship, training, and promotion programs that have dramatically increased the percent of women in the company’s senior ranks.

Sexual harassment in medicine undermines an abiding principle of our profession: First do no harm. This year marked the first time that women outnumbered men entering medical school. Medicine cannot sustain a culture that systematically undermines the authority, physical and mental health, and success of such a large portion of its physician workforce. As a profession, we know a little bit about healing. It doesn’t happen in the dark, without data. It doesn’t happen by protecting or fencing off disease-ridden parts of the body. It happens with scientific precision, objectivity, decisiveness, and consistency. We must take down the traditional hierarchies in medicine that provide a fertile ground for harassment, survey ourselves, and ask the difficult (and sometimes painful) questions about how our culture fails our employees. We must support and strengthen women physicians, and build a climate where transgressions are unacceptable. The time to heal ourselves is now.

5 Concepts That Will Help Your Team Be More Data-Driven

Koron/Getty Images

I’ve spent my career helping companies address their data and data quality opportunities. Overall, I rate progress as “slower than hoped.” While there are many contributing factors, one of the most important is the sheer lack of analytic talent, up and down the organization chart. In turn, this lack of talent makes it harder for companies to leverage their data, to take full advantage of their data scientists, and to get in front of data quality issues. Lack of talent breeds fear, exacerbating difficulties in adopting a data-driven culture. And so forth, in a vicious cycle.

Still, progress in the data space is inexorable and smart companies know they must address their talent gaps. It will take decades for the public education systems to churn out enough people with the needed skills — far too long for companies to wait. Fortunately managers, aided by a senior data scientist engaged for a few hours a week, can introduce five powerful “tools” that will help their existing teams start to use analytics more powerfully to solve important business problems. To be sure, these are not the only tools you’ll need — for example, I haven’t included A/B testing, understanding variation, or visualization here.  Nor is my intent to make people experts.  Rather, based on my experiences working with companies on their data strategy, these five concepts offer the biggest near-term bang for the buck.

The first is learning to think like a data scientist. We don’t speak about this often enough, but it is really hard to acquire good data, analyze it properly, follow the clues those analyses offer, explore the implications, and present results in a fair, compelling way.  This is the essence of data science. You can’t read about this in a book — you simply have to experience the work to appreciate it.  To give your team some hands-on practice, charge them with selecting a topic of their own interest (such as “whether meetings start on time”) and then have them complete the exercise described in this article. The first step will lead to a picture similar to the one below, and the rest of the exercise involves exploring the implications of that picture.

How Often Do Meetings Start Late?

Give employees hands-on experience with data by asking them to collect and plot data on a familiar topic. Meeting start times are just one example.

Source: Thomas C. Redman

Charge that senior scientist you’ve engaged with helping people in completing the exercise, teaching them how to interpret some basic statistics, tables, and graphics, such as a time-series plot and Pareto chart. As they gain experience, encourage your team to apply what they’ve learned in their work everyday. Be sure to make time for people to show others what they’re learning, say by devoting fifteen minutes to the topic in each staff meeting.  Most critically, lead by example — do this work yourself, present your results, and freely discuss the challenges you faced in doing the work.

As you and your team dive into data, you’ll certainly encounter quality issues, which is why pro-actively managing data quality is the next important skill to learn.  Poor data is the norm — fouling operations, adding cost, and breeding mistrust in analytics.  Fortunately, virtually everyone can make a positive impact here.  The first step is to make a simple measurement using the Friday Afternoon Measurement method   (the technique acquired this name because so many teams end up using it on Friday afternoon).

Insight Center



Scaling Your Team’s Data Skills
Sponsored by Splunk

Help your employees be more data-savvy.

To do so, instruct your team members to assemble 10-15 critical data attributes for the last 100 units of work completed by their departments — essentially the last 100 data records.  Then, they should work through each record, marking obvious errors.  They should then count up the error-free records.  The number, which can range from 0 to 100 represents the percent of data created correctly, their Data Quality (DQ) score.  DQ can also be interpreted as the fraction of time the work is completed correctly, the first time.  Most managers are surprised by the results — they expect to score in the high 90s, but DQ = 54 is the median score.

FAM can also point out which data attributes have the biggest error rates, suggesting where improvements can be made, using root cause analysis, described next. Charge each member of your team with making one such improvement.

The third skill is conducting a root cause analysis (RCA) and its pre-requisite, understanding the distinction between correlation and causation. Studying the numbers can point to where most errors occur or demonstrate that two (or more) variables go up and down in tandem, but it cannot fully describe why this is.  For example, studies show that the numbers of live births and storks in the countryside were highly correlated. But storks do not bring babies!

Thus, look to the numbers to understand correlation and to the real-world phenomena to understand causation. Root cause analysis is a structured approach for getting to the real reasons things go wrong — the root causes.  It is important because, too often, managers and teams often accept easy explanations and don’t dig deep enough. And problems remain. RCA can enable them to develop a clearer picture and take actions that are more likely to solve the problem.

To develop this skill with your team, start by discussing “how to explore cause and effect like a data scientist” with your staff.  Then, the next time you find yourself tempted to accept someone’s intuitive reasoning as to why something went wrong, seize the opportunity to conduct a solid root cause analysis. There are many formal means to do so. “The five whys,” which forces you to make sure you’ve gotten to the root cause, and fishbone diagrams, which graphically represent multiple causes, are probably the best known. Have your data scientist pick one, and follow it! Over time, seek to make root cause analysis your standard for all important issues.

The fourth skill stems from the desire all managers have to “be in control.”  My working definition of control is “the managerial act of comparing process to standards and acting on the difference.”  But even the simplest process varies.  How can one distinguish normal day-in, day-out variation from situations that are truly out of control? Fortunately, understanding and applying control charts provides a powerful way to do just that.

Control charts feature a plot of the data, the average, and two “control limits,” (an upper control limit and a lower control limit).  Basic as they are, they reveal so much!  For example, in the Figure below:

 

Since day 9 falls outside the control limits, a manager can be certain this process is out of control. They should initiate a root cause analysis to figure out why.
There is an uptick at day 4 that looks encouraging. But a manager should not get too excited — the uptick was more likely due to random variation and was not sustained.
It is clear enough that that this process only succeeds 60% of the time. If this is not good enough, the manager must make fundamental changes.

Engage your data scientist in helping you and your team try control charts on a few important processes. Learn as you go, understanding key terms, determining which control charts to use, and striving first to get processes under control — your confidence will grow, as will your ability to manage your team!

Finally, all managers and their teams should learn to understand and apply regression analysis. Regression provides a powerful means to explore the numerical relationships between variables.  To help illustrate this, consider “umbrella sales.”  There are dozens of factors that could increase sales (e.g., rain) or decrease sales (e.g., a competitor’s price cut).  Regression provides a way to determine which variables are most important and their impact on sales.  For example, an analysis may yield:

Monthly sales = 200 + 5*(days of rain) – 10*(competitor price cut in $) + error term

Meaning that:

Absent other factors, monthly sales are about 200 units.
A day of rain is associated with the sale of five more umbrellas,
A competitor cutting its price by one dollar is associated with ten fewer umbrellas sold

The model is not perfect — hence the error term. For example, suppose you sold 250 umbrellas in a month when there were 15 rainy days and a competitor cut its price by $2.  Based on the formula, one would expect umbrellas sales to be 200 + 5*15 – 10*2 = 255 units.  So the error term in that case is 5 umbrellas.

Like all analyses, the more variables, the more complex the analysis, so start by focusing on one independent (e.g., explanatory) variable.  In parallel, read “A Refresher in Regression Analysis,” which uses umbrella sales as an example to explain the terms and underlying concepts. Charge your data scientist with helping your team do the work, and making sure team members don’t get bogged down in details.  Only then should you move onto two, three, or more variables and more complex regression models.

These five tools are powerful, even elegant, in their own ways.  They provide far greater capabilities than the steps described here, which aim only to get you started. You’re certain to take some false steps along the way, but press on.  Work with your data scientist to learn even more. As your team grows more confident in using analytics, the business benefits you gain will more than justify the effort.