Landscaping Knowledge

The number of titles which have been dreamt up for this page are far too numerous to mention, so I'll spare you the details and just get on with it. Also note that this article is essentially a continuation of Creating Knowledge Objects.

The thought is that when developing content for the knowledgebase it is generally more appropriate to develop a set of objects which sort of blanket a subject area rather than simply develop a single solution. This idea was arrived at after a number of people looked at a single solution and subsequently asked a whole bunch of related questions for which there weren't answers in the knowledgebase.

The next question then became one of how to provide some guidance as to how to figure out what questions might be asked so there was some idea of what set of objects ought to be developed to pretty much completely address and area being considered. What follows are the perspectives developed to date.

Note that I attempted to develop this in a generalized fashion and found the result so complicated I didn't understand what I had developed. As such, after the general introduction the remainder of this is developed using a "Database Seeding" in a Solution Centered Support implementation as a context. Since it is expected that most will not be aware of what this subject is about to begin with, if what follows provides meaningful guidance then it probably works well. If not, then send me some mail and I'll work on it some more.

A Beginning

There are basically two ways to describe anything. You can describe a "thing" in terms of itself, wherein a "thing" is described by a pattern set of self-reference relationships. Alternatively you can describe a "thing" in terms of what it does, or has the potential to do. In this second approach a "thing" is described in terms of a pattern set of relationships with other "things." In this fashion the description is essentially the purpose of the "thing." More often than not we seem to mix these two perspectives.

The typical questions people ask about a "thing" are:

  • What is it?
  • What does it do?
  • What does it affect?
  • Who does it?
  • Who is it important to?
  • Why is it important?
  • Where is it done?
  • When is it done?
  • How is it done?
  • How is it evaluated?
  • What does it cost?
  • How long does it take?
  • Where can I get it?
  • What does it do for me?

These questions can be divided into 3 specific areas:

And somewhat more specifically as:

Now lets see how this would be applied to a specific subject area, i.e. Database Seeding.

Database Seeding

Database Seeding, in a Solution Centered Support implementation context, is an activity which puts solutions in the database. This essentially answers the question, "What is database seeding in a Solution Centered Support implementation context?" Yet, simply putting this definition in the knowledgebase leaves the knowledgebase a little short in terms serving as a source of support regarding the subject of database seeding.

In attempt to landscape the subject area my first action was to draw an interactions diagram to develop an understanding of the major components that interact in the activity.

From here I went on to ask what seemed to be the most relevant questions which arose while pondering this diagram:

  • What is database seeding?
  • When is database seeding done?
  • Who does the database seeding?
  • How many people should be involved?
  • Where is seeding done?
  • What has to be in place so it can be accomplished?
  • What's the ROI associated with seeding?
  • What feedback should be provided to the analyst during the seeding?
  • What is the value to coaches from participating in the seeding effort?
  • How does one evaluate the result of the seeding effort?
  • What are the analyst prerequisites prior to seeding?
  • How long should it take?
  • What solutions should be used to seed the database?
  • What is the value of database seeding?
  • What does seeding accomplish?
  • What is the role of a coach during database seeding?
  • How many solutions should be seeded in the database?
  • Should seeding be done into the production database?

From this it should be evident that to address database seeding in a Solution Centered Support implementation context will take more than one knowledge object in the knowledgebase. Admittedly you could develop a single object relating to database seeding which linked to a web page which contained all the questions along with their answers. I think this sort of defeats the whole concept of just-in-time knowledge-on-demand. Yet, at the same time I don't think that one question one object is quite appropriate either. Some of the questions are so closely related that trying to respond to them in different objects would fragment the foundation to a point of making seeding difficult to understand.

Presently I think the answer to this is to divide the questions first along the dimensions of information, knowledge, and wisdom as described above. Then review the questions and see which ones are so closely aligned that it would be in the best interest of the user if they were explained in the same object.

Here's an attempt (which I'm not quite comfortable with as of yet):

This essentially divides the landscape into What is seeding?, How should seeding be done?, Why should seeding be done? And, from the number of questions associated with the Knowledge dimension my initial thought is that I would probably create 3 objects relating to this area: preparation, doing, and evaluating.

Solutions Created

From the above considerations I actually ended up creating 4 solutions relating to database seeding. There are provided in the following sections.

Solution # 1

Goal: What is database seeding?
Fact: Activity which loads solutions into the database prior to production use
Fact: Information: definition, description, or perspective
Fact: Customer Support
Fact: Solution Centered Support
Fact: Implementing Solution Centered Support
Fact: Database Seeding
Symptom: Need to improve analyst's willingness to initially use the database
Symptom: Want to improve the speed with which the organization gets up to speed implementing Solution Centered Support
Fix: Because an empty database is of no value, seeding is a process that loads solutions into the database prior to production use. After seeding analysts can use the solutions in the database to resolve customer problems and as a foundation for creating new solutions. Analysts should find a seeded database of immediate value. This should reduce analysts reluctance to use the database when it initially goes into production use.
Fix: Ref: Some reference document on Database Seeding
Fix: Source: Gene Bellinger

Solution # 2

Goal: How does one prepare for database seeding?
Goal: What are the analyst prerequisites prior to seeding?
Goal: When is database seeding done?
Goal: Where is seeding done?
Goal: What has to be in place so database seeding can be accomplished?
Goal: What solutions should be used to seed the database?
Fact: Knowledge: strategy, practice, method, or approach
Fact: Customer Support
Fact: Solution Centered Support
Fact: Implementing Solution Centered Support
Fact: Database Seeding
Fix: There are several things that have to occur before the actual seeding process happens:
1. The technology infrastructure must be in place to support access to the production database for seeding;
2. Analysts must receive training on the technology used to create solutions;
3. Analysts must receive training on the Solution Content Standard which defines the particulars of solution content in the analysts organizational environment;
4. Facilities must be set up where analysts can do seeding away from the production environment. An attempt to do seeding in a production environment with the phones ringing and other interruptions will significantly reduce the effectiveness of the seeding effort;
5. A plan must be developed as to who is supposed to be doing seeding at what times, and who the coach(es) are that will be in attendance to support the effort. All analysts should have an opportunity to do some seeding as the actual seeding, and associated coaching, develops the analysts skill in developing good solutions;
6. Analysts should prepare solutions for entry prior to doing the actual seeding.
7. The best solutions are the most recent problems solved for which there are expected to be more calls, and the questions most often asked of the analyst by their peers;
Fix: Ref: Some reference document on Database Seeding
Fix: Source: Gene Bellinger

Solution # 3

Goal: How does one conduct database seeding?
Goal: Should seeding be done into the production database?
Goal: Who does the database seeding?
Goal: How many people should be involved in database seeding?
Goal: How many solutions should be seeded in the database?
Goal: How long should it take to seek the database?
Goal: What is the role of a coach during database seeding?
Goal: What feedback should be provided to the analyst during the seeding?
Goal: What is the value to coaches from participating in database seeding?
Fact: Knowledge: strategy, practice, method, or approach
Fact: Customer Support
Fact: Solution Centered Support
Fact: Implementing Solution Centered Support
Fact: Database Seeding
Symptom: Trying to plan a database seeding effort
Fix: Database seeding should be done in an environment that enables the individuals doing seeding to concentrate on the effort. Seeding should be as much a learning experience as a solution creation activity. Considerations are:
1. All analysts should participate in some portion of the seeding, and it is important not to automate the seeding process;
2. The seeding should happen immediately after analysts receive training on the solution capture technology and the Solution Content Standard;
3. The database should be seeded with roughly 50% of a week's worth of call volume with redundant calls eliminated. A group of 10 analysts averaging 10 calls a day handle 500 calls per week. If there is a 20% redundancy then the group really only handles 400 calls a week. Taking 50% of this would mean that the seeding process should populate the database with about 200 solutions. This is 20 solutions per analysts and should take about 2 days per analyst.
4. All solutions entered during the seeding process should be put into the production database in such a way that they are initially not findable by others until they have been reviewed.
5. Coaches should review all solutions entered during the seeding process and provide development coaching to the analysts on how to improve the solutions they are developing. This is also a good opportunity for coaches to develop their coaching skills.
Fix: Ref: Some reference document on Database Seeding
Fix: Source: Gene Bellinger

Solution # 4

Goal: How does one evaluate the result of database seeding?
Goal: What is the value of database seeding?
Goal: What does database seeding accomplish?
Goal: What's the ROI associated with database seeding?
Fact: Knowledge: strategy, practice, method, or approach
Fact: Wisdom: principle, insight, moral, or archetype
Fact: Customer Support
Fact: Solution Centered Support
Fact: Implementing Solution Centered Support
Fact: Database Seeding
Symptom: Need to figure out if the seeding effort provided a sufficient return on investment
Fix: The intent of the seeding effort is to make the production database immediately valuable to the analysts in the group that will use it. The only way to determine if this has been accomplished is to survey the analysts in the group.
Fix: Approximately one week after completion of the seeding effort, or one week after the database has gone into production use, whichever is later, survey the analysts in the group with the following questions:
1. During the past week, have you found solutions in the database, from the seeding effort, which helped you in solving customer problems?
2. What did you not find in the database that you would have expected to be there as a result of the seeding effort?
3. Is what you didn't find in the database now?
4. What recommendations would you offer in terms of how we should change the seeding effort for future groups to make the database more useful when it initially goes into production?
Fix: Based on the responses to these questions one should be able to determine if analysts actually found value from the seeding effort and what things might be done to improve future seeding efforts.
Fix: Ref: Some reference document on Database Seeding
Fix: Source: Gene Bellinger


Creating Knowledge Objects * theWay of Systems * Feedback * Musings
Copyright © 2004 Gene Bellinger