Another way to incorporate "learning from failure" into a company's culture might be to recast the entire notion of failure by creating a culture of experimentation. Over on the Fortune Business Innovation blog, Dominic Basulto cited an article by James Cash of Harvard Business School that highlighted the importance of creating a culture of business experimentation within a company, pointing out examples from a broad spectrum of companies: Wal-Mart, Capital One and General Electric. Creating this culture of business experimentation is harder than it sounds, since you can't always know whether or not an idea will work until you put it into action, and most companies are not prepared for the downside risk of failure.
Cash's article contains some interesting comments on failure tolerance and the need for a “culture of experimentation. For example, Cash says:
"It's not an easy feat to create an environment that walks the line between so-called failed experiments—where the discipline of data collection, analysis and iteration results in learning even if the experiment itself doesn't produce a desired result—and the frivolous waste of resources, where ideas are tested in an undisciplined manner."
What I'm thinking is that you could undercut the negative psychological energy around the concept of "failure" by incorporating a culture of experimentation. Here's my guess at what this culture of experimentation might look like:
Experiments would be fashioned from specific innovation challenges framed to meet certain strategic objectives. The resulting concepts would be prototyped in successive iterations, each designed to get closer to the "finished" product or process. At each prototype level an evaluation would take place using criteria developed out of the original objectives. The decision to kill the concept or develop it further would be made. If the decision was to kill, then a full report on what worked and what didn't would be made.
Here's what's most important:
1. Set up a specific process for experimentation and identify experiments as such from the start so that projects that don’t go through the process aren’t allowed to “bring down” the process.
2. Start the process with a specific challenge tied to strategic objectives – this kind of disciplined approach will ensure the relevance of both the process and the results.
3. Make evaluation criteria as specific as possible and relevant as well (which it will be if it’s tied back to the strategic objectives). This is critical to a disciplined approach, and it’s also critical to avoid blame and finger-pointing when failures do occur.
4. Subject all prototypes to rigorous analysis. The ability to build on what’s working and change what’s not is based on the understanding that would come from such an analysis.
5. When a project is killed, document all possible learnings, including things that can be learned about the process itself.
Renee Hopkins Callahan started IdeaFlow and serves as chief blog-wrangler. She is 
1. Tim Bailen on October 30, 2005 2:53 PM writes...
The Experimental Gameplay Project at Carnegie Mellon implemented some of your points. Over the course of a semester, four grad students pumped out over 50 games. They found point 2, "start the process with a specific challenge", to be especially important.
An article on what they learned can be found here:
Permalink to Comment"How to Prototype a Game in Under 7 Days"
http://www.gamasutra.com/features/20051026/gabler_01.shtml
2. Renee on October 31, 2005 11:30 AM writes...
Tim -- Thanks for the comment. The article is great! I'll point it out in a post as well.
It's truly amazing how just two things -- challenge framing and constraints -- can boost creativity. But think about it -- many people say they are more creative on deadline, and a deadline offers both those things -- a frame and a constraint. When you set those up consciously, it's very powerful.
Permalink to Comment3. Daniel Williams on November 5, 2005 8:05 PM writes...
Wow! What a concept! I plan to forward this to my director, but was curious if this was original thought, or a derived from other principles.
Permalink to Comment4. Renee on November 7, 2005 3:15 PM writes...
Thanks, Daniel!
Yes, this is original thought. I definitely think business needs to have a culture of experimentation in order to make innovation work, but haven't been able to find anything specific published about business experimentation beyond the "design thinking" that's all the rage these days. I made this up based on some of those principles (such as prototyping) and my experiences in a creative profession -- journalism. There's also a little bit of CPS in there (Creative Problem Solving), which is a rigorous process for solving problems. It doesn't specifically recommend experimentation, but it does involve validation, which scientific method tells us is critical for experimentation.
Permalink to Comment5. Peter White on November 9, 2005 10:40 AM writes...
We're working with 29 earliest-stage bioscience projects, and our methods and processes are pretty much what's described. Results are wonderful, with two scientists getting investment within six months of proof of concept. Reason it all works is that the 'entrepreneurs' are all skilled scientists and the whole principle of goal-driven failure is good scientific practice: also 'fail fast'. Our processes started out described as enterprise prototyping, which encourages a 'back to the drawing board*' philosophy where failure is simply a gate.
* See the classic New Yorker cartoon by Peter Arno
Permalink to Comment6. Mike on November 21, 2005 10:38 AM writes...
We have found at the intelligence agencies and certain of the military institutions that one can “short-cut” the issue of failed experiments, by building graphical models, called Grapho constructs, of the project and then testing the model for Threats, Impacts, Risks, etc. These models not only eliminate the risk of investing in a “failure”, they foster a great deal of creativity and “risk taking” because the risk is in the mind and not the purse. We have modeled everything from competitor attacks to the human personality using subjective probabilities. See http://www.futureknowledge.biz/AdvancedMarketModel-3.htm
Thanks for this site…its great.
Kindest Regards,
www.futureknowledge.biz
Permalink to Commentmcahill@futureknowledge.biz
San Jose, CA 95125
7. Mike on November 21, 2005 10:44 AM writes...
We have found at the intelligence agencies and certain of the military institutions that one can “short-cut” the issue of failed experiments, by building graphical models, called Grapho constructs, of the project and then testing the model for Threats, Impacts, Risks, etc. These models not only eliminate the risk of investing in a “failure”, they foster a great deal of creativity and “risk taking” because the risk is in the mind and not the purse. We have modeled everything from competitor attacks to the human personality using subjective probabilities. See http://www.futureknowledge.biz/AdvancedMarketModel-3.htm
Thanks for this site…its great.
Kindest Regards,
www.futureknowledge.biz
Permalink to Commentmcahill@futureknowledge.biz
San Jose, CA 95125
8. RICK KELLER on November 22, 2005 4:47 PM writes...
As a leadership professional who works deep within the cultures of many companies, the enticement of a so-called, "culture of experimentation" is only possible if the organization is able to provide the necessary leadership (people and resources) that allows its current culture to accept failure that comes from taking risk. When failures are no longer being measured in terms of dollars lost, as they are currently, and become measured by experiences and knowledge gained, or thinking processes being rewarded, and solutions being advanced that actually impact the bottom line, will this concept be readily accepted into the company mainstream. I am working hard to allow my clients to make this transition. I look forward to working with you culture-changers!
Permalink to Comment9. Wyszukiwanie on April 27, 2007 3:42 AM writes...
I Think in bussines culture is a problem to do money :P
Permalink to Comment