A Virtuous Mix Allows Innovation to Thrive
Skip to content
Entrepreneurship Innovation Nov 4, 2013

A Vir­tu­ous Mix Allows Inno­va­tion to Thrive

The right mix­ture bal­ances con­ven­tion­al­i­ty, nov­el­ty, and collaboration

Based on the research of

Brian Uzzi

Satyam Mukherjee

Michael Stringer

Benjamin F. Jones

On the one hand is the trope of the great sci­en­tist: a lon­er in a rum­pled lab coat, con­tent to spend every wak­ing hour chas­ing the extra­or­di­nary from the recess­es of his own con­sid­er­able mind. On the oth­er is the truth about how sci­ence actu­al­ly gets made.

Add Insight
to your inbox.

We’ll send you one email a week with content you actually want to read, curated by the Insight team.

In an influ­en­tial 2007 paper, Kel­logg School pro­fes­sors Bri­an Uzzi and Ben­jamin Jones, along with a col­league, ana­lyzed the near­ly twen­ty mil­lion research arti­cles in the Web of Sci­ence (WOS) data­base to deter­mine how the pro­duc­tion of research has changed over time. What we saw was that from the fifties up until today, there had been a shift to teams,” explains Uzzi, a pro­fes­sor of man­age­ment and orga­ni­za­tions and the fac­ul­ty direc­tor of the Kel­logg Archi­tec­tures of Col­lab­o­ra­tion Ini­tia­tive. Teams were not only becom­ing more promi­nent, but they were becom­ing big­ger each year.” Teams were also, across a major­i­ty of dis­ci­plines, increas­ing­ly pro­duc­ing the most impact­ful papers — those capa­ble of set­ting, or reset­ting, the research agen­da of an entire field. It became clear to us, says Uzzi, that sci­ence had made a fun­da­men­tal change.”

There are many rea­sons why the best sci­ence may have shift­ed to teams. Per­haps team research is sim­ply favored by fund­ing agen­cies, or maybe it plays well at tenure time. But one rea­son that struck Uzzi and Jones as both plau­si­ble and fas­ci­nat­ing is the idea that col­lab­o­ra­tions might fos­ter more cre­ative or nov­el research.

Make It New

There’s a very long­stand­ing idea that the cre­ation of a new thing is about putting exist­ing things togeth­er in a new way,” says Jones, an asso­ciate pro­fes­sor of man­age­ment and strat­e­gy and the fac­ul­ty direc­tor of the Kel­logg Inno­va­tion and Entre­pre­neur­ship Ini­tia­tive. That is, com­bi­na­tions are the key mate­r­i­al of cre­ative insight.”

As sci­en­tif­ic knowl­edge has expand­ed, indi­vid­u­als have had to spe­cial­ize; teams may allow for depth with­out sac­ri­fic­ing the abil­i­ty to com­bine ideas from dif­fer­ent sub­fields or even dis­ci­plines. And there is noth­ing like the inter­play between mul­ti­ple dis­tinct per­son­al­i­ties and per­spec­tives to fuse unlike­ly but inspired amal­gams. Take John Lennon and Paul McCart­ney. If it hadn’t been for the two of them bounc­ing ideas off of each oth­er — always want­i­ng to one up the oth­er — maybe we’d nev­er have Sgt. Pep­per,” says Uzzi.

But does the nov­el­ty of a research project affect its impact on a sci­en­tif­ic dis­ci­pline? In a paper pub­lished this week in Sci­ence, Uzzi, Jones, and col­lab­o­ra­tors Satyam Mukher­jee of the North­west­ern Insti­tute on Com­plex Sys­tems (NICO) and Michael Stringer of Datas­cope Ana­lyt­ics (for­mer­ly of NICO) find that nov­el­ty does indeed con­tribute to a research paper’s recep­tion — but only when bal­anced by high­ly con­ven­tion­al knowl­edge. The team also finds that research pro­duced by teams tends to bet­ter achieve this bal­ance than research by solo scientists.

Mea­sur­ing Nov­el­ty

In their new study, the researchers mea­sured the nov­el­ty of a giv­en research paper by deter­min­ing whether its influ­ences — the ref­er­ences it cites in its bib­li­og­ra­phy — go beyond the usu­al suspects.

Fit­ting­ly, they adapt­ed their method­ol­o­gy from an unex­pect­ed source. We wound up build­ing off a sys­tem that I heard that the Coen broth­ers had used to build nov­el­ty into their films,” explains Uzzi. After writ­ing a draft of the script for their first film, Blood Sim­ple, Uzzi explains, the broth­ers found it hum­drum. So they cut their script into small pieces and recon­struct­ed it hap­haz­ard­ly — a scene here, a moment there. And in fact when you watch Blood Sim­ple, you’re like, wow! What a twist in the sto­ry. How on earth could that hap­pen? Who would have thought of such a thing? And what you real­ize is, they did it by build­ing this ran­dom­ness into it. So we try to recre­ate some­thing like this in the paper.”

Specif­i­cal­ly, the researchers asked whether indi­vid­ual research papers draw large­ly from com­bi­na­tions of sources that appear togeth­er often in the WOS — which spans dis­ci­plines as dis­parate as soci­ol­o­gy and nan­otech­nol­o­gy — or whether the com­bi­na­tions are clos­er to what would be expect­ed if cita­tions were drawn ran­dom­ly, as if from a hat.

The Right Com­bi­na­tion

The researchers found that, in actu­al­i­ty, nov­el com­bi­na­tions of cita­tions are rare, and get­ting rar­er: in the 1990s, just 2.7% of arti­cles had a medi­an cita­tion pair­ing that was com­bined less often than expect­ed by chance — down from 3.5% in the 1980s.

You want to be ground­ed in some­thing that’s well under­stood and yet be adding in the piece that’s tru­ly unusu­al.” — Bri­an Uzzi

They also found nov­el­ty by itself was no hall­mark of suc­cess. What’s inter­est­ing,” says Uzzi, is most of the work done is con­ven­tion­al. And some of the work is tru­ly nov­el. And the chances of either one of those clas­si­fi­ca­tions of papers being hits is about the same.” Only about 5% of research papers that draw from only very nov­el or only very con­ven­tion­al sources were among the most high­ly cit­ed papers in the database.

But there was a third cat­e­go­ry of research that had near­ly twice the like­li­hood of mak­ing it big: papers that relied most­ly on con­ven­tion­al com­bi­na­tions of sources but also includ­ed a small sub­set of high­ly nov­el ones. It isn’t all about nov­el­ty or con­ven­tion­al­i­ty. It’s about both,” explains Jones, who was some­what sur­prised by this result. The researchers were also struck by how con­sis­tent the trend appears to be, span­ning over fifty years and across hun­dreds of disciplines.

But, says Jones, it also makes sense. You want to be ground­ed in some­thing that’s well under­stood and yet be adding in the piece that’s tru­ly unusu­al. And if you do those two things [and] stretch your­self in both direc­tions, then you rad­i­cal­ly increase your prob­a­bil­i­ty of hit­ting a home­run.” Uzzi con­curs. Many of these nov­el com­bi­na­tions are real­ly two con­ven­tion­al ideas in their own domains,” he says. You’re tak­ing estab­lished, well-accept­ed ideas, which is a won­der­ful foun­da­tion — you need that in sci­ence. But when you put them togeth­er: wow. That’s sud­den­ly some­thing real­ly different.”

The Role of Col­lab­o­ra­tion

Regard­less of whether papers were authored by a sole sci­en­tist or a team of them, this vir­tu­ous mix” — plen­ty of con­ven­tion­al­i­ty and a dash of orig­i­nal­i­ty — was most like­ly to pro­duce high-impact papers. But team­work played a role as well. Teams of three or more sci­en­tists pro­duced over­all more nov­el work than sci­en­tists work­ing by them­selves or in pairs. Teams were also over­rep­re­sent­ed as authors of papers that hit the con­ven­tion­al­i­ty – orig­i­nal­i­ty sweet spot. Final­ly, teams do more with the same mix­ture of nov­el­ty and con­ven­tion­al­i­ty. That is, whether a paper’s influ­ences are 90% con­ven­tion­al and 10% nov­el or 10% con­ven­tion­al and 90% nov­el, the paper has a bet­ter chance of mak­ing a splash if it is pro­duced by a team. Teams are bet­ter at find­ing nov­el­ty but also, it seems, at assim­i­lat­ing it.

The suc­cess of Jones and Uzzi’s own col­lab­o­ra­tion — this is the pair’s third Sci­ence paper in a decade, each pro­duced with con­tri­bu­tions from out­side col­leagues — is itself an endorse­ment of team-based sci­ence. Uzzi is trained as a soci­ol­o­gist, Jones as an econ­o­mist. There’s actu­al­ly pret­ty big gaps in the way that a soci­ol­o­gist and an econ­o­mist look at the world,” says Uzzi. Econ­o­mists are inter­est­ed in indi­vid­u­als behav­ing ratio­nal­ly, while soci­ol­o­gists study groups of peo­ple behav­ing in ways that are often not ratio­nal at all. It’s inter­est­ing that we study col­lab­o­ra­tion and one of our find­ings is that peo­ple who bring togeth­er exper­tise from dif­fer­ent domains may be an essen­tial fea­ture of progress,” agrees Jones. We’ve nev­er test­ed our own papers using these met­rics, but maybe we should!” He adds, It’s clear to me that we both bring impor­tant insights and exper­tise to the prob­lem in a way that makes the whole greater than the sum of the parts.”

Featured Faculty

Brian Uzzi

Richard L. Thomas Professor of Leadership & Organizational Change; Co-Director, Northwestern Institute on Complex Systems

Benjamin F. Jones

Professor of Strategy; Faculty Director, Kellogg Innovation and Entrepreneurship Initiative (KIEI)

About the Writer

Jessica Love is the staff science writer and editor for Kellogg Insight.

About the Research

Uzzi, Brian, Satyam Mukherjee, Michael Stringer & Ben Jones. 2013. “Atypical Combinations and Scientific Impact.” Science. 342 (6157): 468-472.

Read the original

Suggested For You

Most Popular


How Are Black – White Bira­cial Peo­ple Per­ceived in Terms of Race?

Under­stand­ing the answer — and why black and white Amer­i­cans’ respons­es may dif­fer — is increas­ing­ly impor­tant in a mul­tira­cial society.


Pod­cast: Our Most Pop­u­lar Advice on Advanc­ing Your Career

Here’s how to con­nect with head­hunters, deliv­er with data, and ensure you don’t plateau professionally.

Most Popular Podcasts


Pod­cast: Our Most Pop­u­lar Advice on Improv­ing Rela­tion­ships with Colleagues

Cowork­ers can make us crazy. Here’s how to han­dle tough situations.

Social Impact

Pod­cast: How You and Your Com­pa­ny Can Lend Exper­tise to a Non­prof­it in Need

Plus: Four ques­tions to con­sid­er before becom­ing a social-impact entrepreneur.


Pod­cast: Attract Rock­star Employ­ees — or Devel­op Your Own

Find­ing and nur­tur­ing high per­form­ers isn’t easy, but it pays off.


Pod­cast: How Music Can Change Our Mood

A Broad­way song­writer and a mar­ket­ing pro­fes­sor dis­cuss the con­nec­tion between our favorite tunes and how they make us feel.