Learning & Dev’s Positive Deviance Approach

positive deviance in learningWe, in learning & development, often focus on fixing what’s wrong. We try to find guidance on how to do it right, and then build learning around it. We’re reactive, in that sense. Can we do better? Positive Deviance is a different approach.

In addition to the subversive appeal of the label, there’s a record of success. The underlying idea is looking for new ways of working, that are better. And this can be problematic. It can be hard to go out, track down these instances that aren’t necessarily obvious, and analyze them. But new tools give us new possibilities.

The Concept

I first heard about positive deviance from Jane Bozarth of the Learning Guild. She cited the story about how, in Vietnam, a team were tasked with improving nutrition. They found, counter-intuitively, that the poorer members of the community were actually doing better in children’s health. And they built a program around the ways those folks were feeding their children.

As a more general approach, the focus is on applying useful insights. The process involved includes

  • a community wanting to improve
  • identifying the problem
  • finding instances where the problem is not occurring
  • documenting why
  • building an intervention
  • implementing the solution
  • evaluating the success
  • fine-tuning the approach
  • and rolling out more broadly

Which is different in some important respects than the typical approach. Instead of finding experts, the approach finds instances of success and documents what’s working, regardless of alignment with established process. Rather than rely on received wisdom, this is working intrinsically with successful demonstrations.

And there may be congruence between what’s working and what’s proposed as the ideal approach, but it’s not guaranteed. One of the big problems in organizational initiatives is replicating what worked elsewhere (best practices) instead of re-contextualizing the underlying theory (best principles). So, that received wisdom may not be relevant in this instance, for practical constraints.

And, of course, the discoveries may find nuances that the expert approach glossed over, or again are local refinements that go beyond theory. Or it may be a completely new approach that someone’s found that works. In the famous example of protein folding, an important chemical area of understanding, computers found some solutions people couldn’t see and vice versa. The point is that what may emerge from practice isn’t the same as what would be obtained from principle.


While you want to follow the best prescriptions, if someone’s succeeding in other ways, wouldn’t you want to know? How can you do this?

The problem, of course, is discovering those instances of positive behavior. Even if you know what the problem is, finding those different yet successful approaches isn’t always easy. What you want is a way to find those initial successes so that you can find out who to investigate.

These days, data gives us a handle. More and more, we can track finer-grained details about individual performance. Business intelligence systems can record sales, manufacturing errors, customer service success and more. And there are new opportunities.

One mechanism emerged from the foundations of web tracking. Marketers, interested in consumer behaviors, put in hooks to watch what customers were doing on sites. With mobile devices, they can do more. That data can be correlated with business outcomes, like sales.

For performance, the Advanced Distributed Learning initiative of the Department of Defense took notice. The originators of SCORM, they recognized the limitations of only tracking courses, and thus emerged the Experience API, a new standard. This standard lets you instrument any digital system to record individual activity. Moreover, there are mechanisms to capture non-digital activity as well. While not the only such, nor the only way to do this, with the weight of the DoD behind it this standard has gained traction.

The interesting element is the ability to align the information about performance with the information about business success. Here you can see who’s succeeding. And you might even be able to determine what they’re doing differently. Regardless, you can do more research if necessary to see if who’s succeeding is doing it in the prescribed way, or a new way.

The Opportunity

Too often, the response might be to get them back on track. But with a positive deviance mindset, you might instead investigate what they’re doing differently. You can see whether there are important differences that are worth propagating elsewhere.

As things move faster, organizations have to increasingly be agile. Finding new and better ways to work is a new imperative. Experimenting is a key way to trial and improve. The opportunity to uncover new performance potential is a gift.

Of course, opportunity uncovered doesn’t always realize a positive outcome. To be successfully capitalized upon, there are incumbent actions. The exact nature of the different must be understood. Making that work at scale is a necessary and separate endeavor. Then developing an intervention to accomplish that change is a non-trivial exercise. When you’re talking change, you’re talking chances to go wrong.

And the intervention doesn’t have to be ‘a course’. You have to look at the nature of what’s happening in the successfully deviant behavior. Is it a resource, a mindset, a process? What is necessary to support making that change systemic?

Yet the upside is worth the effort. This doesn’t stand alone, but systematically coupling the documented best principles with a continual evaluation of emergent possibilities optimizes the chances for good practice. And it’s easier. Even without xAPI, more and more performance systems are offering mechanisms to track more and more data.

The difficult thing with data is knowing what you want to do with it. Obvious choices are to look for courses that are not performing or, at a finer granularity, questions or resources that aren’t leading to good outcomes. Looking broader, for unexpected benefits, is non-obvious yet another opportunity.

Being positive is a plus. Being deviant has been viewed negatively, but putting a positive spin on it turns out to be a useful perspective. It’s time to be positively deviant (not just deviant). Are you game?