Some Insights About Insights
- Sivan Halevy
- May 1, 2004
- 4 min read
When we build an insights repository, we face several goals and challenges. First, we aim to establish a dynamic and up-to-date repository where knowledge is continually added and accumulated. Second, we aspire for the repository to be useful and relevant to employees at the right time and place, and that at any given moment, they will remember its existence as a repository that can be consulted. How do we address these challenges?
Here are some insights about managing an insights repository:
Insight Approval
To ensure a quality repository where all insights have added value, are practical, and agreed upon, we can use two approaches:
Approve all insights by a single central authority appointed as the repository manager.
Bring insight and approval to team discussions.
We will naturally use the first approach when the repository manager is recognized as an authoritative figure with broad knowledge of the unit's overall activities. The primary advantage of implementing this technique is the ability to achieve resource savings and quickly introduce high-quality insights into the repository.
We will use the second approach when:
We want to gain a broad consensus regarding the insights
We want to create a commitment to implementing the insight among all stakeholders
We want to develop employee capabilities. We believe that by conducting team discussions, we can develop capabilities such as knowledge processing and creation, converting tacit knowledge to explicit, focused articulation, and developing generalization ability (within the discussion framework, we can learn together about all cases for which the insight might be relevant)
If we examine both approaches above, we can easily identify the significant added value that can be derived from team discussions beyond the accumulation of insights. To benefit from these advantages, we recommend exercising judgment (and not always sticking to what is "natural" to do). Thus, even when we do have an authoritative figure to determine what constitutes a quality insight, it's worthwhile to leverage the benefits of team discussions and decide in which cases and according to which criteria this technique should be used (for example, when there's concern that a particular insight will provoke disagreements, when there's danger that an undesirable situation will repeat itself and no solution has yet been found to circumvent it, etc.).
Contributing to and Using Insights
Our goal is to reach a state where employees contribute new knowledge as part of routine organizational work. For this purpose, we always consider natural anchors in work processes where contributions to and utilization of insights can be integrated. An example of such an anchor is collecting insights upon completion of the 'Lessons Learned' section in the documentation template and integrating significant insights into the template. This is indeed a lot, but not enough. It's important that we constantly "use our heads" in identifying additional anchors so that we indeed succeed in developing a learning organization culture. We can, for example, consider additional activity stages, such as team improvement activity summary meetings, significant events that occurred within that professional group, and discussions of project activities. In each of these, we can allocate space for reflection and documentation of "lessons so far." (The possibility of having a content expert join some of these meetings should be examined.)
Additionally, we can select a technique that ensures existing insights in the repository are not overlooked. Within team meetings, we can select, each time, 3 existing insights from the repository that have been there for longer than X time, and conduct a discussion about them. Are they still relevant? Did we make mistakes that could have been avoided if we had implemented the insight and remembered it in time? Is there a need to update it (for example, its ability to generalize to additional cases we hadn't thought of)?
Creative thinking about maximizing the integration of insight use and creation as part of a work routine will result in continuous, consistent learning that produces relevant knowledge assets.
Insight Characteristics
Every insights repository has characteristics that help the user twice. Once to retrieve knowledge in a focused manner according to their needs (for example, to filter all insights dealing with new equipment procurement), and once as thinking aids in the generalization process (is the insight relevant only to new equipment procurement or also to equipment maintenance after procurement?).
In this context, there are 2 types of characteristics: content characteristics (relevant to the specific content world, such as new equipment procurement) and system characteristics (appearing in every insights repository regardless of its content). Examples of such system characteristics: knowledge contributor, knowledge contribution date, link, date for reviewing currency, sensitivity, etc.
We would like to propose an additional system characteristic – insight direction. When we derive insights, there are insights intended to replicate success, to repeat activity. In these insights, we recommend acting in a way X. In contrast, some insights direct us to avoid mistakes by not acting in a way X. In short, we can say that all insights will be formulated as "do" and "don't do" recommendations.
If so, we can also use this general characteristic as an additional filtering tool by marking the direction of each insight ("do" or "don't do"). This marking will increase the focused retrieval capability of the end user. Such marking can be done using symbols such as a sad face for "don't do" recommendations and a happy face for "do" recommendations (as in the example below), arrows (down or up according to insight direction), or "+" and "-" signs.
Comments