The New Killer Apps: Teamwork and Weak Signal Detection Lessons from the Military

Killer App: a feature or component that in itself makes something worth having or using. Merriam–Webster

I recently had the pleasure of speaking at the eXperience Agile conference on Military Secrets of High-Performing Teams, and was asked by Ben Linders to share some additional insights with the InfoQ community on that talk. Ben and I decided to turn my InfoQ interview into an article where I could dive a little deeper into how teamwork and weak signal detection lessons from the military are becoming “The New Killer Apps.”

Are We Really That Different?

In this volatile, uncertain, complex, and ambiguous (VUCA) world, the U.S. Navy (1) is in deep competition to attract and retain top talent, (2) must find a way to stay competitive as near-peer competitors now have equal access to information and technology, and (3) has recognized that the rate of change is outpacing its ability to field relevant technology to the warfighter.

If you replace the words “the U.S. Navy” and “warfighter” with “business” and “customer” respectively in the previous sentence, you may notice that the problems facing the U.S. Navy are the same challenges facing today’s business leaders.

Low-Energy Cost Approach to Scaling Agility

When it comes to finding a low-energy cost approach to scaling agility, we must agree that scaling agility, like scaling safety, is a misnomer. Agility and safety are emergent properties of complex adaptive systems, not something organizations do or scale. Therefore, we should look to complex adaptive systems to discover a low-energy cost approach to creating agility and safety. And teams are a great place to start, as they happen to be complex adaptive systems.

A team is best defined as two or more people who work interdependently, adaptively, and dynamically toward a shared and valued goal/mission/objective. If you read that definition of a team again, you will notice that a team, by definition, is an agile organism. If organizations focus on building networks of real teams, not just teams in name only but teams that meet the definition of a team, then the organization and their customers will enjoy the benefits of agility, improved safety, and resilience.  

A low-energy cost approach to agility and safety is simply copying others who are great at building teams, understanding why their particular teaming approach works, trying it and adapting it to your context, and then scaling the teamwork approach across your organization. Also, natural science informs this low-energy cost approach.

Contrast this with the high-energy cost or “islands of disconnected effort” approach to agility, where an organization hires a bunch of certified consultants who argue over frameworks. Some of these consultants swear by engineering or scaling agile approaches to fix human systems. Organizational leaders measure agility by the number of teams in name only who are following an Agile framework. Subjective surveys are thrown around that only measure work as imagined. Myers-Briggs testing and other pseudo-science approaches are commonplace, and the latest business fad seen in HBR is applied across the organization to create leaders.

I call this high-energy approach to agility the “Suck Less” strategy—where the only way this approach will work is if the organization sucks less than their competition. Boards and shareholders of companies whose leaders are wasting resources on a “Suck Less” Agile strategy should be concerned if their competitors decide to embrace the low-energy cost approach to agility. And some are.

Teamwork Lessons from the U.S. Navy

During the Vietnam War, naval aviators were flying technologically superior fighter aircraft compared to their adversary, but aerial engagement outcomes suggested that superior technology did not translate to superior results. In 1968, a report was released by the U.S. Navy recommending the creation of a quality circle or center of excellence focused on improving aircrew tactical performance in air-to-air combat. With this mandate, a sense of urgency, and, of course, limited resources, nine naval aviators launched a startup in a small trailer that would forever change the way carrier-based fighter pilots would improve continuously.

If you consider that air-to-air combat is an extreme team activity—even more so for highly interdependent, dual-cockpit aircraft such as the F-4 Phantom II used by the Air Force and Navy during the Vietnam War—the problem these nine startup founders had to solve was a human systems problem. The startup became the Navy Fighter Weapon School (TOPGUN), and the solutions that moved the performance needle from 2.6:1 to 13:1 in air combat victories for the U.S. Navy were found in how aircrews transferred knowledge and experience through the art and science of briefing and debriefing.

Today, TOPGUN’s Plan-Brief-Execute-Debrief (PBED) cycle is burned into muscle memory of every fighter pilot. But it is not the PBED cycle that makes a team; it is the quality of interactions in each event that ultimately turns a team of experts into an expert team. The same is true about Scrum. Scrum follows the same team life cycle as TOPGUN’s PBED, yet many Scrum teams are teams in name only simply because members do not know how to interact or work together as a team. Let me be clear: following a framework does not make a group a team.

After a series of high-profile incidents that occurred at sea in the late 1980s—incidents that involved groups of people working with advanced technology—the U.S. Navy initiated a program that advanced research on team performance and the science of teams. The aim of this U.S. Navy program, known as Tactical Decision Making Under Stress (TADMUS), was to help the navy understand how leaders and teams make high-stakes decisions under dynamic and uncertain conditions. Researchers including Gary Klein learned a lot about teamwork, team training, naturalistic decision making, situational awareness, sensemaking, and planning through this navy-sponsored program.

During the same period, commercial aviation was having great success with Cockpit Resource Management—a human factors program designed to reduce hierarchical structures and improve teamwork skills in the cockpits of commercial airliners.

In the early 1990s, the navy combined lessons from TADMUS with lessons from commercial aviation’s Cockpit Resource Management to launch Naval Aviation’s Crew Resource Management (CRM) program. The U.S. Navy’s CRM program targets non-technical skills training (teamwork), with the goal of improving mission effectiveness by minimizing crew preventable errors, maximizing crew coordination, and optimizing risk management.

Today, CRM training is the most widely applied human factors and team training technique to operations personnel in High Reliability Organizations (HRO). An interesting connection between the U.S. Navy and the study of HROs is that the concept was coined by researchers who observed flight operations aboard a Nimitz class aircraft carrier back in the mid-80s. HROs have what is known as a generative culture—a mission-focused culture where messengers are trained, decisions are made by those closest to the problem, and members are always looking for failure. A generative culture is the same type of culture the DevOps movement is trying to create. Just a thought here, perhaps DevOps should look to carrier flight ops?

What Can We Learn from This

The most important lesson: learn from unusual contexts. With product and company lifecycles rapidly decreasing, there is little time to figure out on your own what others have already learned. It is okay to learn teamwork and leadership skills from unusual contexts such as carrier aviation. Why? Because teamwork and leadership skills are transportable.

Research on military teams and leadership has advanced to what is now known as the science of teamwork. This science looks at things such as team cognition, team effectiveness, effective debriefing, planning, multi-team systems, shared mental models, leadership, how to foster psychological safety, types of teams, crews, and how to measure team performance.

When I came into “Agile” coaching out of the military, I was shocked that few of these concepts were known by the coaches who claim to coach “Agile” teams. I even had senior leaders tell me that this “team coaching thing you do is not Agile, that teams should self-organize when it comes to teamwork.” I have news for you: that approach and thinking kills people in the air, in operating rooms, and anywhere safety incidents can be expected. And it’s the same approach and thinking that’s killing programs, products, and organizations.

If we look at the factors behind failed and successful programs, products, and companies, they look a lot like the factors behind why aircrafts crash, why surgical teams make mistakes in the operating room, and why we lost two space shuttles. Lessons from other contexts suggest that teams should self-organize around task work only after they learn how to work together as a team.

Teams are central to the way the military accomplishes its mission. The military has a 60-year head start on industry when it comes to building teams and multi-team systems. One of the biggest advantages the military has over industry is the time allocated to train teams. The military spends 99% of its time training and 1% doing; industry sets 1% of its time aside for training and 99% of the time for doing. But what industry leaders do with that 1% matters. And right now, I argue that industry is spending 80% of its 1% training time and resources on 20% of its problem—those trivial many things that matter less than teamwork skills.

In a nutshell, the military knows how to turn ordinary people into exceptional teams, while industry struggles to find extraordinary people and put them on ordinary teams.

An industry that has recognized the value of teamwork training inspired by the military is healthcare. Right now, the third leading cause of death in the United States is human error. But when we dive deeper into the human error data, we find that poor teamwork skills are what really kill. Team Strategies and Tools for Enhancing Performance and Patient Safety (TeamSTEPPS®) is one of the teamwork programs used by the healthcare industry to address this serious problem, and this teamwork program happens to be informed by the U.S. Navy’s TADMUS and CRM programs.

Industries benefiting from teamwork approaches from military aviation include oil and gas, railroad, spaceflight, emergency medical systems, and even shipping companies. We are starting to see an uptick in other industries calling for teamwork training inspired by the military, as well. Nigel Thurlow mentioned during the eXperience Agile conference and in his Scrum the Toyota Way interview that Toyota Connected is leveraging the Cynefin framework and lessons from the science of teamwork to build one of the most agile organizations in the world. I spent a lot of time talking with Nigel about these concepts over the past two years and it is great to see a Lean-Agile thought leader see the light. There’s hope!  

Second lesson: teamwork skills must be learned, practiced, and reinforced at every level of an organization. There are a couple of points I want to make here, and the first point should be the most obvious from what we already discussed: high-technical expertise and advanced education are not the same as teamwork skills.

The next point is about teamwork skills above the “team-level.” By the way, I hate this term. It is condescending in so many ways. Not all teams in name only at the “team-level” need to work as teams.

Being promoted to a leadership position does not make a person a leader—especially if the promotion was based off of technical skills. The higher up a person is in an organizational hierarchy, the higher their task interdependence. Context, specifically the level of task interdependence, determines if a team is needed. I should point out that leadership is a component of teamwork—and it is hard to be a leader if you lack teamwork skills.

The people who need teamwork skills the most in an organization are not necessarily the ones closest to the customer. Those who need teamwork training the most are the ones who generally tell everyone else to work as teams, while they act as wrestlers; we call them managers.  

Finally, to reinforce teamwork, organizations need to change their performance management practices from rewarding individual performance to developing and rewarding teamwork. To do this, HR or people operations leaders need to work closely with coaches who can break down the components of teamwork into observable behaviors and feed those desired behaviors back into the company’s performance development program. Failure to align performance management with teamwork is a recipe for failure.

Weak Signal Detection

What causes us to miss weak signals? In a word… humanness.

As humans, we are all susceptible to ego and fear, cognitive biases, linear thinking, the desire to drive complex systems toward order, cultural influences, we confuse correlation with causation, and in some contexts, such as launching from the front end of an aircraft carrier, we have physiological limitations. These and other human factors are what cause us to miss weak signals.

The biggest barriers to weak signal detection are ego and fear. As leaders, our job is to create conditions that enable teamwork and foster the psychological safety needed to allow team members to speak up when they notice things that others don’t see. But to create these conditions, leaders have to get over their illusion of knowledge—they need to admit to themselves and others that they do not and cannot know everything—and they need to display to others that they are fallible, that they are human.

This is tough to do, considering most executives and managers are promoted on their technical competencies and not their leadership skills. And this is why I pointed out earlier that teamwork skills must be learned, practiced, and reinforced at every level.

It gets worse. Inattentional blindness prevents us from seeing what we do not expect to see. This cognitive bias, combined with automation bias and linear thinking, for example, make picking up and acting on weak signals nearly impossible.

Improving Our Non-Technical Skills to Pick-Up on Weak Signals

We have to recognize and experience these barriers before we can move on to techniques that mitigate them. The way we do it at AGLX is by pulling lessons from the Cynefin framework, putting teams through simulations that build their non-technical skills, and during workshops we play simple games where attendees can experience different barriers.

Once people have experienced these barriers, they are more receptive to techniques that mitigate them. Some of the techniques we use include complex facilitation techniques from Dave Snowden and Gary Klein’s work and lessons from Red Teaming—a military concept that helps organizations overcome the complex problems introduced by our humanness. These approaches work in ideation sessions, planning, meetings, reviews, and debrief events.  

In addition to addressing our non-technical skills, we have to examine the tools we use, as they often prevent us from detecting weak signals. For example, traditional surveys and focus groups are often biased either by a facilitator or a pre-determined hypothesis. Additionally, survey fatigue is real, and when survey respondents do have time to think about the survey, they will often parrot back what they think the surveyor wants to hear. By answering subjective survey questions such as, “On a scale of 1-7, is your Scrum Master effective at leading events?” we capture work as imagined, not work as done.

With the desire to make better decisions through Big Data, we are seeing the rise of natural language processing (NLP) tools on the market. But NLP tools are still biased, as they transfer the interpretation of meaning from a facilitator to an algorithm. Although Big Data can predict what employees or customers may do next, it does not tell us why. And it is knowing why—the attitudes, beliefs, and emotions of our customers and employees—that helps leaders make sense of the operational environment so they can make better decisions. Leaders need Thick Data to get to why.

Thick Data is best collected through stories or narratives where those that tell the story add meaning to or self-signify the stories. This approach of adding meaning to a story told or experience is called disintermediation; it is how we remove the intermediary layers between the raw data and the people who need to act on the insights, and it is the fastest way to move from research to action.

This next generation approach to capturing weak signals or Thick Data is found in what are known as scans and pulses powered by SenseMaker®. One of the secrets behind this weak signal detection capability is its roots in the Defense Advanced Research Projects Agency, the U.S. Department of Defense agency that gave us the basis for the modern Internet. You can read more about the connection this sense-making capability and the Cynefin framework have with the military in Dave Snowden’s Cognitive Edge blog.

Improving Team Performance

Team performance is a function of task work and teamwork. Task work is what teams do, and teamwork is how team members interact. The great thing about teamwork is that it is observable and therefore measurable.

So how does the military measure teamwork?

To measure their teamwork skills, fighter aviation crews and military medical teams use behavioral markers. A behavioral marker is a rating scale comprised of a skills taxonomy with examples of good and poor behaviors. Behavioral markers are observable, non-technical behaviors that contribute to superior or substandard performance within a team environment; they provide an objective approach to measuring team performance.

During my eXperience Agile keynote, I shared some insights on how we are combining behavioral markers with team frameworks such as Scrum and TOPGUN’s PBED cycle to help teams monitor and improve their teamwork skills, regardless of industry. You can learn more about our Team Performance Indicator (TPI) here.

There is a psychological phenomenon known as Dunning-Kruger effect, where people with low-ability in a subject or skill will tend to mistakenly overinflate their competence in that subject. When deploying our TPI in organizations, we’ve seen that individuals who have not been through teamwork intervention training tend to rate their team performance has high. Following teamwork intervention training, the same individuals will then rate their teams as underperforming.

A team improves its performance by increasing insights (weak signal detection) and reducing errors. We already covered weak signal detection, and the reduction of errors is accomplished through effective teamwork. The single most important team event that improves team performance is an effective debrief or retrospective. There is some recent research coming out of the science of teamwork that suggests well-conducted or effective debriefs improve team performance by 25%.

The expectation of the debrief or retrospective helps create accountability. And accountability, according to eXperience Agile speaker Christopher Avery, PhD., means that if called on, you can tell others what happened. In an effective debrief, we improve insights by leveraging multiple perspectives to understand what happened before moving on to how and why. Gathering data is good, but we are more interested in building individual and team situational awareness so we can learn to notice more (detect weak signals).

In fighter aviation, debriefs focus more on teamwork than task work, and outcomes are separated from decisions. In VUCA, sometimes a good outcome is the result of a bad decision. One of the most important aspects of to holding an effective debrief is overcoming the Dunning-Kruger effect. And you do this through non-technical or teamwork skills training.  

Concluding Thoughts

There are a lot of great teamwork and weak signal detection lessons from the military that can help forward-leaning leaders create the organizational agility and safety they need to survive and thrive on their own terms in this VUCA world. And perhaps, instead of hiring coaches with multiple certifications, hire a person who has lived what they coach. Hire a veteran. They lived it!

About the Author

Brian Rivera, is the founder of AGLX Consulting, creator of the ZONEFIVE™ Team Performance Indicator, and an international speaker on teams and high-performance. He is a member of the Cynefin.net network and a former F-14 instructor and demonstration team member. Rivera recently spoke about military secrets of high-performing teams at eXperience Agile 2018. InfoQ is covering this conference with Q&As, summaries, and articles.