In the late 1960’s, Gary Starkweather had a serious spat with his boss.  As an engineer in Xerox’s long-range xerography unit, he saw that laser printing could be a huge business opportunity. His manager, however, was focused on improving the efficiency of the current product line, not looking to start another one.  

The argument got so heated that Starkweather’s job came to be in jeopardy. Fortunately, his rabble rousing caught the attention of another division within the company, the Palo Alto Research Center (PARC) which wasn’t interested in efficiency, but inventing a new future and they eagerly welcomed Starkweather into their ranks. 

Within a decade, Xerox’s copying business declined sharply, but the laser printer took off and soon became the firm’s main source of revenue. In effect, the work that was squelched in one culture, thrived in another and saved the company. We tend to think innovation is about ideas, but it depends on people even more. Here’s how you create an innovative culture.

1. A Focus On Problem Solving

When you think about an innovative culture what probably first comes to mind is a bunch of fast moving hipsters guzzling down energy drinks and pulling all-nighters, pausing only to play a quick game of foosball or frisbee. Or maybe Steve Jobs on stage with a devilish grin just before he wows the audience with “one more thing…”

Yet in researching my book, Mapping Innovation, I found that very few of the organizations I studied looked like that. Some were fast moving startups, but most of the successful ones were led by executives that were mature and thoughtful, not brash or erratic. Others were large corporations and world class labs that tended to be fairly conservative.

The one thing I found in common in every fantastically innovative place I looked at was a disciplined passion for identifying new problems. Unlike most organizations, which are content to struggle with everyday issues, the enterprises I studied had a systematic method of finding new problems to work on that would take them in new directions.

The approaches vary considerably. IBM creates grand challenges, like building a computer that can beat humans at Jeopardy. Experian set up a Datalabs division to find out what’s giving its customers “agita” and launch new business off the solutions they build. Google’s “20% time acts as a human-powered search engine for new problems. 

We tend to think of innovation as fast moving, but the truth is that it usually takes 30 years to go from an initial discovery to a measurable impact. So the “next big thing” is usually about 29 years old. If you want to innovate effectively, don’t chase the latest trend, find a problem your customers will care about and solve it for them. 

2. Create Safe Spaces

In 2012, Google embarked on an enormous research project. Code-named “Project Aristotle,” the aim was to see what made successful teams tick. They combed through every conceivable aspect of how teams worked together — how they were led, how frequently they met outside of work, the personality types of the team members — no stone was left unturned.

However, despite Google’s nearly unparalleled ability to find patterns in complex data, none of the conventional criteria seemed to predict performance. In fact, what they found mattered most to team performance was psychological safety, or the ability of each team member to be able to give voice to their ideas without fear of reprisal or rebuke.

Interestingly,  highly innovative teams can be safe for some ideas, but not for others. For example, two of the scientists at PARC, Dick Shoup and Alvy Ray Smith, developed on a revolutionary new graphics technology called SuperPaint. Unfortunately, it didn’t fit in with the PARC’s vision of personal computing, the two were ostracized and eventually both left.

Smith would team up with another graphics pioneer, Ed Catmull, at the New York Institute of Technology. Later they joined George Lucas, who saw the potential for computer graphics to create a new paradigm for special effects. Eventually, the operation was spun out and bought by Steve Jobs. That company, Pixar, was sold to Disney in 2006 for $7.4 billion.

Xerox PARC is now a shadow of its former self. As it turned out, anything that didn’t have to do with the researchers’ vision for the future had no home there. So if you want to innovate consistently for the long term, you need to create a “safe space” for all ideas, not just the ones that fit with your initial mission.

3. Foster Informal Networks

In 2005, a team of researchers decided to study why some Broadway plays become hits and others flop. They looked at all the usual factors, such as production budget, marketing budget and the track record of the director, but what they found was that what was most important was informal networks of relationships among the cast and crew.

If no one had ever worked together before, both financial and creative results tended to be poor. However, if the networks among the cast and crew became too dense, performance also suffered. It was the teams that had elements of both — strong ties and new blood — that had the greatest success.

The same effect has been found elsewhere. In studies of star engineers at Bell Labs, the German automotive industry and currency traders it has been shown that tightly clustered groups, combined with long range “weak ties” that allow information to flow freely among disparate clusters of activity results in better innovation.

So before you embark on your next reorganization designed to “break down silos” you might want to think about how informal relationships develop within your enterprise. The truth is that innovation is never about nodes. It’s always about networks.

4. Promote Collaboration

All too often, we think of innovation as the work of lone geniuses who, in a flash of inspiration, arrive at a eureka moment. Yet the truth is that research shows that the high value work is done in teams, those teams are increasing in size, are far more interdisciplinary than in the past and the work is done at greater distances.

Just as importantly, there is growing evidence that it is crucial how these teams function. A study done by the CIA performed after 9/11 to determine what attributes made for the most effective analyst teams found that what made teams successful was not the attributes of their members, or even the coaching they got from their leaders, but the interactions within the team itself.

In another, more wide ranging study, scientists at MIT and Carnegie Mellon found that high performing teams are made up with people who have high social sensitivity, take turns when speaking and, surprisingly, the number of women in the group. There is also a wealth of research that shows diverse teams outperform more homogenous units. 

So the evidence is both abundant and clear, if you want to make your organization more innovative, don’t go searching for hard driving “A” personalities spouting off big ideas and interrupting others, but rather seek diversity, empathy and to network your organization so that teams interact more effectively.

As MIT’s Sandy Pentland has put it, “We teach people that everything that matters happens between your ears, when in fact it actually happens between people.” 

This article originally appeared at DigitalTonto.

Greg Satell is a popular speaker and consultant. His first book, Mapping Innovation, is coming out in 2017. Follow his blog at Digital Tonto or on Twitter @DigitalTonto