What is a lesson learned? Simply put, it is is knowledge or understanding gained by experience (whether positive or negative), and which adds significant, valid, and relevant new information that would be useful in accomplishing a business objective.

Most emphasis on lessons learned is on things that go wrong, rather than things that go right. The types of trouble for projects and product designs are well-known. However, these are categories of failure, and by themselves, don't adequately chronicle or prompt practitioners about all the countless ways that projects, product designs, and operations can find themselves in trouble. As an analogy, think of mountain climbing. When one does mountaineering, the types of hazards include falling, getting lost, and becoming overcome by weather conditions. However, while these threats are known, the specific way they manifest themselves is not equally evident, and so other means are appropriate to ensure that common sense can be practically translated into meaningful action. Having someone knowledgeable along to warn you of the risks of a specific crevasse is highly recommended. Sometimes, the best we can do is a thoughtful note from that knowledgeable person, rather than their active participation. That's what lessons learned can do.

For industries and professional practices in which such failures are highly important to businesses or society (such as safety-critical product applications) it becomes essential to utilize a more disciplined approach that ensures that when problems occur, the learning and new approaches are both applied to current products, and are woven into the fabric of practices of the organization for future products or operations. For interesting insights about how challenging this can be, and a fascinating background on the history of the discoveries of such lessons within engineering projects (and patterns which they tend to manifest), see the books Normal Accidents, Human Error, The Logic of Failure, and Inviting Disaster.

Because of a desire to learn such lessons, many organizations have formalized the lessons learned process itself; see links at the end of this article for examples. A lessons learned process has been institutionalized for over a century within teaching hospitals, who are now mandated by law to hold events such as M&M conferences to communicate and share learnings. An entire organizational development culture has also now been established around the concept of a learning organization, which draws upon these same ideas. There are even consulting businesses and commercial software tools now to support this! In light of this, it is worthwhile to step back and consider what lessons can be learned from these other lessons learned projects, when one is considering a similar pursuit themselves.

One of the things that is evident are the scenarios that a lessons learned will go through during it's lifecycle, and the issues which arise in these different contexts. They include: 

• collection - The means of collection can be active or passive, channeled or anonymous, structured or unstructured; the choice is yours, though each has it's advantages and disadvantages.
• development - The prioritization, resourcing, and classification of lessons must all be addressed; the quality of the products of this effort will have a direct impact on their ability to be used effectively by others in the future.
• verification - This typically is performed by one or more gatekeeping functions, which should also recognize the importance of retaining the ability to tailor the individual organization's learning's according to their own culture, processes and standards.
• storage - Issues of access, indexing, categorization, and representation must be addressed.
• dissemination - Approaches to transferring lessons into practice can vary widely - push, pull, facilitation, mentoring, training, and assessments are all options, and many different communications channels can also be utilized. However, the underlying stickiness of the ideas themselves should be enhanced to increase the likelihood of adoption
• application - This can involve improving processes, behaviors, and products; typically this will involve a user different than the submitter. May be helpful to provide a 'rating' system (similar to Amazon) to allow downstream users to provide feedback on the value of individual lessons.
• retirement or deprecation - Can be triggered by dates or based upon feedback on users with respect to value, accuracy, relevance, demand, duplication, or available implementation support

The numbers of items transitioning through each of these states over time can provide an effective performance measure for the effectiveness of the associated processes.

What are the critical ingredients to a lessons learned activity across these various scenarios?

• A team of knowledgeable experts that have the knowledge, skills, experience, and time to properly research the individual situations, what led to them, and how they could be avoided or exploited in the future.
• Techniques such as The Five whys and Failure mode and effects analysis need to be utilized to isolate problems which occur to the root causes which generated the learning opportunity.
• A set of well-formed criteria for documenting and reviewing each 'lesson' need to be established (for example, see the Bradford-Hill criteria used in epidemiology).
• A chartered body of experts must be available to approve (or gatekeep) proposed lessons for future action.
• A means needs to be established for assuring that effective communications of the lessons are achieved with all affected stakeholders.
• An accessible infrastructure must be established to provide access to appropriate findings.
• An accountability mechanism must be put in place to provide oversight, and assure that resulting lessons are actually being incorporated into use.
All this takes time, careful attention to detail, and a significant investment on the part of leadership to ensure that roadblocks are removed, and the desired results are developed and incorporated. One of the biggest risks to such efforts is that since results aren't immediately visible, and often aren't readily acknowledged as the sources for improvements, it can be difficult to maintain a sustained commitment to such efforts over a long period of time. As a result, the active involvement of proactive sponsorship is essential to the overall effort.  So ar periodic reviews of lessons learned, such as this one about NASA's Faster-Better-Cheaper initiative.