Harassment Training Doesn’t Work, So Here’s How We Fixed It

By November 21, 2018 No Comments

Sexual harassment isn’t stopping. In fact, the needle isn’t moving up or down – reports from the EEOC over the last 8 years shows a pretty consistent number of reports filed every year. What can be done?

For decades, the workplace has been flooded with courses, policy reform, discussions, and even social movements. Sadly, these efforts have done nothing to improve the situation, and training to prevent harassment actually makes it WORSE. In studies where conventional training was done for sexual harassment, those that engaged in the training actually fared worse than those who did no training at all.

Even for companies who use training merely as a legal buffer to protect them from losing a lawsuit, we have more bad news. By the time you end up in court, you’ve already lost. Back in 1988, it was estimated the average fortune 500 company spent $6.7 million per year in indirect harassment costs. That was 30 years ago…. Today, that number is astronomical.

Don’t believe me? To help businesses assess and quantify their unique financial risk, we built an online calculator as a no-obligation tool for HR professionals and learning designers to leverage buy-in from upper management. The numbers are insane. I couldn’t believe them myself. But I want to invite you to play around with the app and see what comes up. It helps if you give realistic data, but I don’t care if you just want to see some results:


You may be shocked at what you find from the calculator, and it may even be depressing; but there is hope! At Sage, we spent two years compiling research on the social, psychological, and educational influences that have made anti-harassment training a failure for the past 30 years. We took all of this knowledge and built an e-learning course that actually provides real solutions. Here’s a short list of what we changed that actually improved the effectiveness of training:

We changed the learning objectives.

Most training places undue emphasis on the responsibility of victims and perpetrators. The problem with this is that 94% of victims don’t advocate for themselves, and if perpetrators have grown up with this false mental model of the world that causes them to harass, how can we expect a 20-minute compliance training to reverse that? These are unreasonable goals. Instead, we focussed on getting bystanders who witness harassment to intervene.

Peruse standard learning objectives through available courses on the market, and you’ll see things like “Define harassment.” This is ineffective because it equates defining a behavior with a behavior change, and that’s incompatible. Knowing what stealing is or murder doesn’t stop criminals from doing it; you have to consider root causes.

This wasn’t a willy-nilly decision, and we spent months consulting with SPHR-certified HR experts to confirm that this choice wasn’t so radical that it disrupted the initial focus of anti-harassment training: to protect the company.

We used storytelling to drive empathy.

Our unique storytelling methodology uses analogies and high-end video production to engage learners more enthusiastically – safely distancing them enough from the content so they could think about it critically and how it applies to their personal lives. This echoes back to psychologists like Bruno Bettelheim, who would often use fables and folklore to help children work through extreme trauma. By using analogy, our stories, in Bettelheim’s words, “personify and illustrate inner conflict, but they suggest ever so subtly how these conflicts may be solved, and what the next steps in the development toward a higher humanity might be. [In stories,] internal processes are externalised and become comprehensible as presented by the figures of the story and its events.” Analogies work wonders for emotional engagement and learning subtle, nuanced soft-skill concepts.

We used a more effective motivation model.

While the above breakdown video has a lot of information, there were literally thousands of neurological, psychological, and andragogical features at work in this course – all to provide a seamless learning experience while actually working to change behavior. Sadly, I can’t list every feature, but wanted to include the element of motivation because it’s so underutilized in compliance training.

The general perception people have about motivation is that it operates on a pleasure-pain stimuli model: people are more motivated to do things that cause them pleasure and less motivated to do things that cause them pain. But that isn’t how human motivation works; we have a variety of models, and not all of them serve us in the way we want.

In Israel, a study was conducted with parents at a daycare. The daycare issued a new policy: all late arrivals to pick up children would be fined. The expectation is that, if we follow this rudimentary model of motivation, fewer people would arrive late. The exact opposite occurred, and parents even began to view the fine as a means to “buy more time.”

Our model followed the works of business leaders and acclaimed researchers ranging from “Flow” author Mihaly Csikszentmihalyi to Ericsson’s former VP Stefan Falk to Stanford University professor Carol Dweck. We also balanced the emphasis of our course between how to intervene, but made sure to spend ample time on the why. Learners that understand why an action must occur are far more motivated to engage in it – and with fervor.

All of this implemented research gets learners excited to go through the course. The fact that our consistent feedback has been that learners have fun is a monumental achievement. I have never heard that about any online sexual harassment prevention course.

We never show harassment.

Harassment is personal, contextual, and manifold: there are endless ways harassment takes place, and it continues to evolve as technology does. (I heard about a situation where an employee was harassing another through Yelp, the local-search app.)

Avoiding displays of overt instances of harassment allows learners to discern for themselves what constitutes harassment, and they are taught how to gauge their level of personal involvement. This process reinforces our trend of getting learners to think more critically in how they assess unique social circumstances.

Further, by not showing specific examples, we avoided a psychological phenomenon called “anchoring” – where people tend to fixate on specific content rather than the context. For example, if you’ve ever heard learners come out of a training bemoaning how “I guess we can’t tell jokes anymore,” you’ve seen anchoring at work.

You might think that learners would need to see examples to better interpolate how harassment looks in their own organization, but the research suggests that approach is flat-out wrong, and actually detrimental – increasing instances of the exact harassment you tried to avoid. Netflix rolled out a policy to their film production sets that outlined a “5-second Eye Contact” rule – whereby employees could be reprimanded and fired if they looked at another employee for longer than five seconds. How did employees respond? They would gawk at each other, bug-eyed, and loudly count to five before averting their eyes. They not only mocked a policy that was supposed to be designed for employee safety, but engaged – regularly – in behavior that made employees uncomfortable.

We let learners build their own personal intervention strategies.

Autonomy was a huge goal for our online course, getting learners to think and act for themselves. The included course book corresponds to each lesson, and learners work through the psychological barriers that keep most people from speaking up – as well as how to overcome them. When they’re done, employees have drafted a tangible, personalized, and actionable strategy for how they will comfortably approach harassers. We took it further: this empathy-building course allows for victims and perpetrators to become active bystanders.  In priming learner empathy, viewers become more receptive to self-reflection and to reading the thoughts, feelings, and emotions of others.

Bringing it all together.

All of the research that went into this course had the singular goal to improve bystander intervention. While some of the stuff we designed may seem hokey – like the binaural music that plays during the writing activity in order to stimulate the emotional processing parts of learner brains – these features do not act in isolation. Doing just one of them doesn’t make the course particularly revolutionary. However, doing all of these things (and there are thousands of features included) all contribute in a way that transforms learners, helping them hone their intuitive senses, empowering them to take action when they witness questionable behavior, and providing logical and emotional tools to better communicate with their team.

If you’d like a more in-depth look at what went into building this course, we recently gave a free, 45-minute webinar discussing its learning design. You can view it here:

If you’d like to purchase our course, you can get a copy here (bulk pricing available):