As NASA’s director of astrophysics in 1990, I launched the Hubble Space Telescope into orbit with a flawed mirror. The subsequent failure review board named the root cause as a “leadership failure.” Was it possible that something we had never thought about, much less talked about, had trumped the work of many of the best technical people in the world? I did not give this much thought at the time. The aftermath of one of the most colossal screw-ups in history consumed me, along with looking for a way to fix the telescope.

The nature of the flaw made a complete fix possible. We focused on development of the space repair mission. Interestingly, nobody blamed me personally for the flaw because Perkin-Elmer fabricated the mirror in 1977, and I was not director until 1983. When some normalcy returned, I began to wonder about leadership and its connection with success and failure of technical teams.

The University of Colorado’s Business School provided me a “professor of leadership” position. How could I contribute to such a crowded field? Perhaps the fact that I was different from most leadership scholars would be useful: I had a doctorate in physics, and a record of accomplishment as a successful executive-level manager. Recalling an expression from undergraduate physics — “The right coordinate system can turn an impossible problem into two really hard ones” — I struggled to find organizing principles in the lists of “things to do” that are in all business books.

I made no progress until I saw a “Dilbert” cartoon that said, “Every consultant makes his living with a two-by-two matrix.” I realized that the organizing system must be a simple X-Y system. Inspired by Carl Jung’s personality theory, I began experimenting with a system to analyze leadership and teamwork into four simple components, called “Dimensions.” These four human needs are:

1) To feel valued, appreciated;

2) To feel a sense of belonging;

3) To have a hopeful, realistic future to look forward to; and

4) To know what others expect of them and have the resources to succeed.

I reorganized my course around these principles.

With some distance from NASA, I realized that as program director, I was responsible for creating the flawed context at Perkin-Elmer that in turn caused the mirror flaw. How could I atone for this? My answer was total dedication to finding how to enhance technical team performance. No one should ever experience what we did with Hubble.

I left the business school and founded “4-D Systems” (D stands for Dimensions) and began providing leadership development services in the accounting industry. This launched a 15 year journey into “what works” based on my training as an experimental physicist. I converted my course into a three-day workshop. However, I grew skeptical about the effectiveness of workshops in creating the only thing that matters — behavioral change.

So, we added post-workshop coaching to minimize the effect of “workshop decay.” Then, we added online assessments to measure behavioral change. The 4-D behavioral modification system now had five components, usually applied in this sequence: Team Development Assessment, Individual Development Assessments, Workshops, Coaching and Reassessments. We built benchmarking scales for the assessments assigning team and team members’ benchmark ranks. During 2002, my NASA sponsor, Ed Hoffman, director of the Academy for Program/Project and Engineering Leadership, decided to implement 4-D processes across NASA project teams. The voluntary adoption of these processes since that time has been nothing short of astounding. We have measured more than 500 teams, 2,000 individuals, with 10,000 coaching sessions and 300 workshop-days. The teams in the bottom quintile are particularly interesting. They are overrunning, under attack by review boards and sponsors and prone to catastrophic failures. Wondering whether the individuals they attracted caused their problems, we ran correlation analyses of team assessment and individual assessment scores. The correlation coefficient, r, was less than 0.3, indicating weak correlation.

So, what’s causing the team to have performance limiting behavioral norms? The answer is context.

Context powerfully drives behavior. As we studied teams with Team Development Assessment scores in the bottom quintile, we noticed a recurrent set of context-breaking management/structural shortfalls.

  • Ineffective team leadership: This is first in the list because it is the most potent. A low score in a team leader’s Individual Development Assessment provides an easy diagnostic for this sin. In many, but not all cases, the leaders knew they were ineffective and were grateful to us for helping them develop, or seeing that they should change to less demanding jobs.
  • Task undoable because of inadequate resources: This was certainly a factor at Perkin-Elmer. The telescope was so underbudgeted that what began at about $500 million ultimately required $1.7 billion. We blamed the contractor when we were responsible for over-optimistic budgeting.
  • Flawed procurement implementation: All too frequently, inexperienced people take shortcuts with procurement. This results in having the wrong contractors, doing the wrong work, with the wrong incentives. This happens in NASA when a project has money they are afraid they will lose unless they spend it quickly.
  • Team affected by a larger “broken” context: Examples of this include a company’s top management relentlessly nagging staff to increase sales when suitable bid opportunities are virtually nonexistent. Imminent layoffs also break team contexts.
  • Team engaged in a power struggle with another organization: Power struggles are common across organizational interfaces. We humans are tribal at our core. When the power struggle has a power imbalance, as in a government-contractor interface, and the government is relentless in its criticism, the weaker party resorts to guerrilla tactics. Contractors tire of the blaming and withhold information that will result in criticism, if they can find a reason to do so. This was surely in play at Perkin-Elmer.
  • Team members temperamentally unsuited to required work: This is about the foundation of your innate personality; the one each of us is born with.
  • Flawed organizational structures: I believe technical managers place more emphasis on organization charts than is warranted. However, clear and unambiguous lines of authority are essential, and frequently missing.

We worked with the 20 percent of bottom-benchmarking team leaders to identify and correct context breaking shortfalls. Team Reassessments then moved teams to high quintiles, preventing context-driven failures. You can see the data graphically displayed at www.4-Dsystems.com.

I consider my debt repaid.

Charles Pellerin is founder of 4-D Systems and author of “How NASA Builds Teams: Mission Critical Soft Skills for Scientists, Engineers, and Project Teams” (Wiley, 2009).