[Printer Friendly Version]

Opening the Silos

Related Links   If you like this article, you may also be interested in: Seminar: Nonprofit Knowledge Management   If you found this article interesting or helpful, please consider making a donation to Nonprofit Online News. It will probably feel good!

Leveraging Communication for Foundation Knowledge Management

By Michael C. Gilbert, October 2005

This article was first published in the October 2005 edition of the Nonprofit Online News Journal.
 

In 2001, we worked with the W.K. Kellogg Foundation on a pilot project to explore and influence the learning related communication patterns of their program officers. It was a classic example of applying communication centered techniques to a knowledge management goal. The following article is an adaptation of our report on our research into those communication patterns.
 

Background

Like many grantmakers, the Kellogg Foundation has well developed program areas, with all the strengths and weaknesses that go along with that staffing and grantmaking model. The strengths include, of course, that program areas allow for efficient payout and a solid level of subject matter expertise. The weaknesses include the classic problems of any staffing 'silos', such as impaired innovation and challenges to considering cross cutting issues.

The foundation had several strategies to mitigate the problem of programmatic silos. One was to fund multi-year initiatives that were in many ways larger than any single program area in their vision. The other was to establish several cross cutting themes and work to introduce those themes into the work of the program officers. Making these cross cutting themes work was the focus of this project.

In particular, we were asked to look for new ways to build the Information Communication Technology (ICT) knowledge of their program officers, so that new technology could be considered as an integral part of program work, rather than as something separate. At Kellogg, this is called the ICT Cross Cutting Theme (ICT CCT). This issue is a large scale problem that still persists in most foundations.

The fatal flaw of most knowledge management initiatives is that they focus upon and organize themselves around the content that leaders are hoping to teach, rather than around the people to be taught and their learning processes. The results therefore are often things like new web sites and reports, or sometimes, lunchtime presentations. These are by no means useless, but their impact is limited and they are often regarded as "yet another thing to read", or process, by those they are intended to benefit.

We took a classic Communication Centered approach to this problem. We were determined not to give overburdened program officers more to do, but to integrate new knowledge into their existing flows of communication and learning. This report covers our research into those flows.

Abstract

We conducted eight interviews and surveyed 25 program staff with an eye to determining the best possible tools for supporting the ICT Cross Cutting Theme.

We found that long reports and large websites were unlikely to find their way into the normal workflow of the staff, especially if not seen as being absolutely central to their program area. We found that information was as important for its ability to introduce staff to experts to whom they might turn, as for its ability to convey lessons of value.

We concluded that a system organized around program areas, which emphasizes connection with trusted experts, and delivers content in a highly digestible form should be the key principles for development of the tool. We are especially interested in leveraging the participation of people who are motivated to provide content, especially grantees.
 

1. Goals and Objectives

The overall goal of this engagement was to explore knowledge sharing mechanisms that will support the ICT CCT. In other words, we're asking the questions: What can we do to support the incorporation of useful Information and Communication Technology within the program work of the foundation?

The objectives of the research portion of our work were to: (1) engage the target audience of program staff in identifying high leverage knowledge and communication flows, and (2) explore the existing patterns of communication in order to inform the development of a prototype knowledge sharing mechanism.

The following diagram illustrates how the research and the prototype will play off of each other, in order to maximize the probability of the project's success:

The research focused on mapping how relevant information currently flows through the organization, where the opportunities are for leverage, efficiency, and substitution, and understanding how people currently learn to improve their program work. The research also focused on establishing a very strong sense of the "as is" situation within the foundation.
 

2. Activities and Findings

A. Interviews with Program Directors

We interviewed seven program staff, plus one technology staff member. There were three parts to the Interview Protocol, as follows:

• 1. Setting Context
  • Introduction
  • Thanks for their time
  • Explanation of procedure for the interview
  • Review of goals/purpose of preliminary interviews
  • Explanation of information mapping
  • Offer to answer questions

• 2. Interview Questions
  • What do you do?
  • To do your job WELL, what information do you need to take in?
    • Who and where do you get it from?
    • By what media?
  • To do your job WELL, what information do you need to disseminate?

    • Who and where do you disseminate it to?
    • By what media?

• 3. Open Conversation

The final open conversation section of the interviews helped us prepare for further investigation by looking at the questions we asked, how they were presented, how they might be received by a larger audience, how we might improve them, and other ways in which we might gather relevant information. We also solicited suggestions as to who we should talk to next, in order to help improve response rates for the upcoming survey.

In analyzing the results of the interviews, certain patterns emerged having to do with the way program directors (PDs) work with and absorb information about technology and other, more general, information. The most important recurring patterns include:

• Long reports and other documents are not generally considered useful. Summaries of reports are much preferred. Email messages containing notes about why an item may be of interest, along with a link to the web site containing the article (or possibly an attachment) are very popular.
• In terms of delivery mechanisms, most of the PDs don't seem to know what technology system a given resource may reside upon or just how to access it. Given two resources on the same system, they may not remember that they are on the same system - they just think about the resource as a place or tool.

• Most of the PDs seem to ignore any technology platform or system WKKF has, unless someone has shown them how to find, and use, some specific tool on it. Most do not and will not actively explore new systems and channels to see what is available.
• PDs strongly prefer to ask a person about a technological issue or service than to use technology to find the same information. This could be about confidence in the source or lack of familiarity with the tools in question.
• Announcement mailing lists, electronic newsletters and the electronic versions of journals were mentioned frequently as a tremendous resource. PDs subscribe primarily to announcement lists, rather than discussion lists. Several mentioned that discussion lists take too much time to read and absorb, or that they are difficult to manage. This may relate to the fact that announcement lists are, in essence, curated. It may also result from a lack of knowledge of all but the most basic email skills. Filtering skills are particularly absent.
• While PDs rely heavily on email lists and electronic versions of journals for information about their program areas, none mentioned contributing content to them.
• An apparent exception to the PD's discomfort with email discussion groups seems to be WKKF's in-house discussion lists, which were mentioned repeatedly as being valued.
• PDs use the foundation's intranet to find contact information, search for grant information and read the daily news. Few other uses were mentioned in the interviews, although presumably some documents are retrieved frequently.
• Four of the seven PDs interviewed specifically mentioned the Council on Foundation's Affinity Groups as an important networking/learning opportunity.

Caveat: We did not interview or survey communications staff, so some material that gets published may come initially from the PDs and be routed through communications. By the same token, some of the information delivered to PDs may be coming in through communications staff.

B. Survey of Program Directors and Program Associates

With the support of a team from the foundation, we surveyed 33 program staff. We received an astonishing 25 replies to this survey. We conducted this survey through an email message to the staff members, directing each of them to a personalized web form. In a few cases, we provided technical assistance and clarification to program staff.

There were eleven questions to the survey, in addition to open comments, as follows (multiple choice answers not included):

1. Looking around your desk, please list the most valuable communications (memo, presentation, newsletter, etc.) that you have recently received.
2. Please list the names and web sites of the key electronic mailing lists to which you currently subscribe.
3. Looking in your email inbox, please copy and paste the three most valuable incoming communications you have recently received, which you are comfortable sharing. Please include the To and From fields.
4. Your bookmarks are a great resource and would provide us with potential insight into what you regard as important enough to record for future reference. Would you be willing to share your bookmark file with us?
5. Your browser history is a similarly valuable resource. Would you be willing to share your browser history file with us?
6. What are the last three things that you have written that you believe were of some value to others in your field or in your work and how were they distributed?
7. What are the key reference sources (books, web sites, databases, etc) that you would be most likely to turn to if you had questions about a technology component of a grant proposal?
8. Who are the three people you would most likely turn to if you wanted to evaluate a technology component of a grant proposal?
9. If there were an online resource designed to help you evaluate ICT components of your work, how often would you be likely to consult it?
10. If there were an online resource designed to help you evaluate ICT components of your work, how would you want it integrated into your ? existing work?
11. Comments, Questions and/or Suggestions

For each of the questions, certain patterns emerged:

(1) Valuable Incoming Communications

Incoming information falls primarily into three categories: information supporting the making and managing of grants; information helping PDs stay current with, and connected to, their issue area and to the sector in general; and information that helps PDs to keep current with, and connected to, the foundation's leadership, policies and procedures.

Grant proposals, budget updates and status reports fall into the first category. Journals, papers, newsletters and guides, as well as correspondence from colleagues outside of the foundation, fall into the second. Administrative documents, memos and meeting minutes fall into the last.

Information sent electronically, including email messages, email newsletters and attached reports, are often printed out to be read/edited/stored in paper form. Phone messages are often transcribed onto paper and passed onto PDs by support staff.
 

(2) Key Electronic Mailing Lists

Email bulletins from respected general news sources such as The New York Times ranked high in terms of perceived value, as did electronic newsletters and updates from well known organizations within a given issue area or geographical area. Also valued are messages from announcement lists run by a PD's grantee.

Announcements of conferences and other events, and publications listing issue-related resources or opportunities were frequently mentioned as valuable.

Discussion lists, on the other hand, were absent from people's lists of valued resources, with the notable exception of internal foundation lists. The internal lists mentioned were either issue-oriented, such as Africa Staff, or function-based, such as WKKF Project Directors.
 

(3) Valuable Incoming Email

Messages from mailing lists also turned up here, similarly with a focus on the PD's issue area. Also cited were messages from team members and grantees, having to do with the monitoring and evaluation of grants, issue resolution, keeping in touch, and sharing resource and opportunity information.

Email from extra-foundational colleagues tended to be either about shared projects, useful resources, coordinating travel plans, or passing along the content of conversations and meetings relevant to the recipient.

Here are a few interesting statistics about incoming email messages considered valuable by the recipient, pulled from the survey data: Messages tended to be short, 20% of the messages had documents attached. Few included URLs.
 

(4) Browser Bookmarks

Bookmark files we analyzed were generally neither well developed nor out of control. Most were for practical tools such as maps, weather, finance, etc., with a few personal interests represented as well.

Other bookmarks represented general and philanthropic news sites, Kellogg-related links, especially tools, and programmatic links.
 

(5) Browser History

The time constraints of the project did not permit a content analysis of the web sites or documents pointed to by the history files. A visual review of history files indicated that there is relatively little use of search engines, routine use of practical tools such as weather, investment sites, and internal foundation systems, and fairly frequent use of grantee web sites and sites related directly to program and initiative work.
 

(6) Valuable Outgoing Communication

Most of the information listed by survey participants in response to this question was oriented toward an external audience such as grantseekers, the general public, researchers, students, members of the press and staff of other foundations. Another audience that materials are produced for is the foundation board.

This external or board focused material often appears on the foundation's public web site, in brochures, journal articles, presentations at conferences or is sent via email to grantees. None of these information streams were listed as part of the daily workflow of the foundation's program staff in terms of information consumption. Ironically, it could be that colleagues within the foundation have less access to this information than do colleagues outside of the foundation.

It should be mentioned that a possibility of data skewing exists for this item - some people appear to have focused excessively on the "distributed" aspect as though it meant 'publicly distributed.' As a result they may have under-reported their internal contributions. As an example, all one participant told us was "Did not publish anything that was disseminated √??in my field.
 

(7) Key ICT CCT Reference Sources

One of the more surprising results in the survey was that 10 out of the 25 respondents stepped outside of the suggested answers to tell us that they would turn to a trusted individual, rather than a book, website, database or some other inorganic resource, when seeking answers about the technology component of a grant proposal. It is perhaps even more interesting to note that 8 respondents said that they didn't know where they would turn for such answers.

The remaining respondents listed various search engines, books and web sites, with no noticeable pattern to them.
 

(8) Key ICT CCT Experts

Nine people were listed more than once in response to this question, and 19 more received a single vote each. There were 13 other, less specific, responses, all of whom boiled down to essentially 'someone who is more knowledgeable on the subject than I am.'

We noticed what might be a bit of sensitivity about question 8: 'There is a big assumption here. That is that we get many, many proposals with a major chunk for technology. I am not sure that's the case.'

A possible source of data skewing is suggested by the fact that the instigators of the survey appeared as the two experts most frequently cited.
 

(9) ICT CCT Tool: Frequency of Use

When asked how often they would be likely to use an online ICT evaluation resource, participants were split fairly evenly between 'A few times a month' and 'Rarely - if ever,' at 11 and 10 votes respectively. Two respondents answered 'Several times a week.'
 

(10) ICT CCT Tool: Preferred Delivery Mechanisms

When asked about the form an online ICT evaluation resource should take, the most popular choice was "On demand search engine," with "Online library" and "Program-related ICT examples" tying for second and "Periodic email summaries" tied with "Support through Program Services" for third favorite delivery method.

We find these results to reflect the typical tension found between what people say they want and how they actually behave. Search engine usage is actually very low, with response to trusted senders of email messages much higher. Furthermore, the spread between most of these answers was very narrow.
 

(11) Comments, Questions and/or Suggestions

The only pattern that we could determine from this final question is consistent with our other findings: Program staff prefer a form of "just in time knowledge management." In other words, they need answers to specific questions, and see themselves as having no time for development of resources. They need to know who to turn to.

3. Conclusions

We focused first on information consumption, because this project had the education of program staff as its emphasis. There were three important conclusions related to the consumption and use of ICT information:

• Program staff are interested in content that they perceive as related to their program area, initiatives, and issues of interest. (This is in contrast to material that is viewed through a technology lens. For example, a piece on the value of mailing lists might be seen as unimportant, whereas a piece on mailing lists for rural development might be intriguing.) This was an overwhelming pattern in all our research. There is a secondary pattern of what might be called "personal context" information, which is information that contributes to the context of their work as they see it. Often this is general news.

  Therefore, the most important organizing principle of the ICT CCT Tool should be programmatic, rather than technological. This should apply to all aspects of the system: email bulletins, the main web interface, and the content partnerships and flows.

• Program staff prefer a form of "just in time" knowledge management, at least as far as all those areas regarded as outside of their programmatic expertise. They want to find the right person to ask, ask them quickly, and get a quick answer.

  Furthermore, the source of a communication was the most important filter in determining the degree of attention it would receive. Sources close to a PD's programmatic work were the most significant.

  Therefore, the second most important organizing principle of the ICT CCT Tool is to reduce the path of access to trusted experts. This principle should apply to the length of most communication (short), providing phone numbers and email addresses a short click away from items, and so forth.

• There is great skepticism about the use of the ICT CCT Tool. Although surveys such as this tend to overestimate participation in future systems, the projected participation is very low, given current perceptions of what it might be.

  Therefore, the ICT CCT Tool should strike a balance between engineering and staffing. It should not be overdesigned. It should have two independent forms of staffing support: engineering support and content curation.

  The information that program staff will consume has to come from some place. Furthermore, our past research indicates that it is often knowledge capture that is the most important bottleneck in a knowledge maangement project. There were three conclusions related to the production of ICT information:

• The circumstances under which program staff produce content (not specifically ICT content) on a routine basis include: (1) when it is required of them in order to move their grantmaking forward, (2) when it is made very easy or is integrated into their workflow, (3) when they see it as advancing their field, or (4) when they have a personal interest or relationship involved.

  Therefore, the ICT CCT Tool must not expect the program staff to make substantial contributions to its knowledge base until one or more of those circumstances is established. The prototype should be designed to make those circumstances more likely over time, for example through a voluntary ICT grant evaluation checklist.

• The media that are used by program staff to produce content include personal electronic mail (although it is not perceived as a medium), workflow documents (supporting grantmaking, initiative teams, ARs, IFRs, etc), and what might be called "prestige documents" (such as white papers, articles, and books).

  Outside of administrative communication, the audience for content produced by program staff include: executives and trustees of the foundation, team members at the foundation, outside experts in the program area, and grantees.

  Therefore the ICT CCT Tool should either become a means to support communication with those core audiences, or it should support such communication, if at all possible. At the very least, it should not detract from such communication, which is perceived to be at the core of the program staff work.

• Furthermore, we conclude that the ICT CCT Tool should gather content in its early phases from those most motivated and ready to contribute to it. We identify those likely contributors as grantees and consultants.

If you found this article interesting or helpful,
please consider making a donation to Nonprofit Online News.
It will probably feel good!