Efficient Meetings: A Knowledge-centric approach

Introduction

Meetings are a cornerstone of collaboration, yet they often fall short in delivering clear outcomes or valuable insights. In today’s fast-paced world, where time is one of our most precious resources, we need to rethink how meetings are designed and executed. Rather than seeing them as routine events, what if we viewed meetings through a knowledge-centric lens, where the primary goal is not just discussion but targeted information discovery?

In this article, we explore how applying the principles of information theory can transform meetings into efficient, purpose-driven activities that maximize the value of each participant’s time. By focusing on acquiring essential knowledge and optimizing attendance, we can ensure that meetings become productive tools for driving decisions and reducing uncertainty.?

What is the Knowledge-Centric Perspective?

The knowledge-centric perspective shifts the focus from traditional software development metrics—such as lines of code or feature completion—to the cognitive effort that drives development. It emphasizes that developers are engaged in knowledge work, not just manual coding.

Knowledge work is the cognitive effort needed to bridge the gap between individual skills and experience and the knowledge required for task completion.?

When tasks exceed a developer’s existing understanding, they must engage in knowledge discovery to fill those gaps. Success depends on their ability to acquire and apply new information effectively and efficiently.?

One of the most common methods for gathering this missing knowledge is through meetings. From a knowledge-centric perspective, a meeting is primarily a knowledge discovery activity.

Designing Efficient Meetings

Meetings can be designed and executed as an efficient knowledge discovery and sharing activity. We fill knowledge gaps by asking questions and receiving answers.

But how can we gauge if a meeting is efficient? By measuring the information acquired. For this, we can use bits of information.

In Information Theory[1], a bit represents the amount of information gained when uncertainty about a situation is reduced by half. A binary question—one with two possible answers, such as "yes" or "no"—helps narrow down possibilities.?

A binary question that removes 50% of the possibilities provides 1 bit of information.

However, not all questions carry the same amount of information. Some questions are broad and general, while others are more specific. Specific questions might only remove a small fraction of uncertainty. For instance, confirming a minor detail may reduce possibilities only slightly, contributing less than 1 bit of information.

IIn a meeting, it's essential to focus on strategically framed questions that target the biggest unknowns and reduce the most uncertainty. These questions provide the most insight and move the discussion forward efficiently. The broader the question, the more uncertainty it can potentially eliminate. Ideally, a well-structured question can narrow possibilities by more than 50%, giving you at least 1 bit of information.

That said, it's not always realistic or necessary for every question in a meeting to reduce at least 50% of the possibilities. However, by aiming for questions that maximize uncertainty reduction, you can make your meetings more efficient. The main goal is to maximize the information gained per question.

In order to better show the difference between broad and specific questions let's consider the following game of hiding a coin.

The game of hiding a coin

Suppose we have eight boxes as shown below. There is a coin hidden in one of the boxes.

In order to find the coin we are allowed to ask binary questions i.e. questions with “Yes” or “No” answers. The boxes are with equal size hence there is a probability of 1/8 of finding the coin in any specific box.

There are many ways we can decide what binary questions to ask. Here are two extreme and well-defined strategies.

Below is an animated example of Brute force strategy when we search for a gold coin hidden in 1 of 8 boxes.

When using the brute force strategy with the first question we remove 1/8 of the possible coin locations. With the second question we remove 1/7 of the remaining possible coin locations. With the seventh question we remove 1/2 or 50% of the remaining possible coin locations.

Below is an animated example of Optimal strategy when we search for a gold coin hidden in 1 of 8 boxes.

Using the optimal strategy with each question asked we remove 1/2 or 50% of the possible coin locations.

Applying the brute force we have a small 1/8 probability of finding the coin with one question but we may need to answer up to seven questions. With the optimal strategy we need to ask exactly three questions. If we compare the two strategies we see that we need to ask at least one question and at most seven questions. The average number is three questions.

Not All Bits Are Equally Easy to Acquire

The final bit of missing information may be much harder to acquire than the first few. This last bit may require deeper analysis, more difficult questions, or even follow-up meetings, creating a diminishing returns scenario. In practical terms, without this final bit, you may not be able to make a fully informed decision.

Example in a Meeting Context:

Consider a product planning meeting where the goal is to decide whether to launch a new feature. You need to resolve three major uncertainties:

• Market Fit: Will the feature meet customer needs?
• Technical Feasibility: Can we develop the feature with our current resources?
• Budget: Do we have the budget to support development and launch?

Instead of asking broad, general questions like "What do you think about the feature?", structure your questions to directly address these uncertainties:

1. “Does data from customer research show a strong need for this feature?” If the answer is "yes," this resolves around 50% of your uncertainty about the feature's success in the market.
2. “Can the feature be developed within our current tech stack and timelines?” This question would eliminate another 50% of uncertainty regarding technical feasibility.
3. “Does the current budget allocation support the feature’s launch and marketing?” This would resolve the final uncertainty around financial viability.

By asking targeted, well-structured questions, you efficiently reduce uncertainty and gather the information necessary for decision-making.

Method for Designing Efficient Meetings

Based on the knowledge-centric perspective outlined in this article, you can design efficient meetings by focusing on maximizing information acquisition while minimizing resource use (in terms of both people and time). Here's a method for designing efficient meetings using the principles of information theory and the Information Efficiency per Person (IEP) metric.

1. Preparation: Define the Information Goal (Focus on Key Uncertainties)

Before the meeting, clearly define the specific information you need to acquire—i.e., the uncertainties that must be resolved. Frame these uncertainties as questions that will be answered during the meeting.

• Identify the Set of Alternatives: List all possible options (solutions, scenarios, etc.) that need to be considered in the meeting.
• Formulate Broad Questions: Use binary questions (yes/no or either/or) that can eliminate roughly half of the possibilities. These questions should focus on the most uncertain or impactful factors, aiming to reduce uncertainty by at least 50%, or 1 bit of information.
• Prepare Follow-up Questions: Once broad questions have narrowed down the possibilities, prepare follow-up questions to further refine your understanding. This can be done systematically using concepts from decision trees and multistage decision-making.
• Prioritize High-Impact Questions: Focus on the questions that will have the greatest impact on decision-making. These high-value questions should aim to reduce the most uncertainty.

2. Estimate the Number of Bits of Information Needed

Each question should be evaluated based on how much uncertainty it will reduce. This is akin to estimating the number of bits of information required to reach a decision.

For example, if you need to answer 3 key binary questions, you can estimate that the meeting needs to resolve 3 bits of information.

3. Optimize Attendee Selection

Only invite attendees who are critical to answering the key questions. This step minimizes unnecessary participants and maximizes the value each person brings to the information-gathering process.

The goal is to optimize the IEP (Information Efficiency per Person) by keeping the number of attendees proportionate to the amount of information expected to be gathered.

4. Structure the Meeting Around Information Acquisition

Organize the meeting into focused segments that target the key questions identified earlier. Each segment should aim to resolve one of the uncertainties.

• Address Key Questions First: Start with the most critical questions to ensure that, even if the meeting is cut short, the most important bits of information are acquired.

5. Monitor Information Acquisition During the Meeting

Track how much information is being acquired throughout the meeting. For each key question answered, measure how much uncertainty has been reduced (in terms of bits of information).

If significant uncertainties remain unresolved after a certain period, consider adjusting the structure or planning follow-up actions.

6. Calculate IEP After the Meeting

After the meeting, calculate the IEP (Information Efficiency per Person) using the formula provided in the Appendix. This metric provides feedback on how effectively the meeting gathered information relative to the number of attendees.

7. Iterate and Improve

• Follow an Iterative Process: Many meetings require a series of questions, with each subsequent question removing further uncertainty. Prepare follow-up questions to deepen understanding as the meeting progresses.
• Be Aware of Diminishing Returns: Not every question will reduce uncertainty by a large amount, but by prioritizing high-impact questions, you can make meetings more efficient.
• Refine Future Meetings Using IEP: Use the IEP from previous meetings to improve future meeting design. If meetings consistently show low IEP values (meaning too many attendees relative to the amount of information gathered), consider reducing the size of the meeting or refining its focus on key information-gathering objectives.
• Identify Patterns: Are certain types of meetings more efficient than others? Are there participants who consistently contribute to reducing uncertainty? Use this data to inform and optimize future meetings.

This approach helps ensure that meetings are productive, focused, and efficient by aligning them with the principles of information theory and maximizing the Information Efficiency per Person (IEP).

Example of the Method in Action:

Let’s walk through an example to illustrate the method:

1. Information Goal: You’re planning a meeting to decide whether to launch a new feature. The key questions (uncertainties) you need to resolve are:
2. Estimate Bits: You determine that answering these 3 questions will fully resolve the uncertainty around launching the feature, so you need 3 bits of information.
3. Attendee Selection: You invite 5 key people who can directly contribute to resolving these questions: a product manager, a technical lead, a financial analyst, a customer insights expert, and a project manager.
4. Structure the Meeting: The agenda is structured around these 3 questions. Each question gets a specific time slot where the relevant attendees provide input to resolve the uncertainty.
5. Monitor Information Acquisition: As the meeting progresses, you track how much uncertainty has been resolved. If by the end of the meeting you’ve answered 2 of the 3 questions, you’ve acquired 2 bits of information, but 1 bit remains unresolved.
6. Calculate IEP:
7. This IEP of 0.1 indicates that the meeting was relatively inefficient, with KEDE = 0.1 per person. You may consider follow-up actions to resolve the final bit of uncertainty.
8. Improve: Based on this result, you can reflect on whether the meeting was overstaffed, whether it could have been better structured, or if the final question required more preparation or a follow-up meeting.

Conclusion

The knowledge-centric perspective on meetings shifts the focus from traditional time-based measures to information acquisition and efficiency. By applying the method outlined in this article, organizations can transform meetings into precise, data-driven activities that aim to resolve key uncertainties effectively.

The advantages of this approach are clear:

• Focused Information Acquisition: By targeting key uncertainties, meetings stay on track and avoid being bogged down by irrelevant details.
• Optimized Use of Time: The IEP metric provides a measurable way to assess whether the right number of participants were invited and how efficiently their time was used.
• Continuous Improvement: With the built-in feedback loop, you can continuously refine meeting processes, leading to better outcomes with fewer resources over time.

Incorporating these principles will lead to meetings that are not only more productive but also more efficient, ensuring that every participant’s time is well spent and that the knowledge gathered drives meaningful decisions.

Appendix

1. Measuring Meeting Efficiency

How can we gauge if a meeting is efficient? By measuring the information acquired. For that we can use bits of information.

In Information Theory, a bit represents the amount of information gained when the uncertainty about a situation is reduced by half. A binary question is a question with two possible answers, such as "yes" or "no," "true" or "false," or "0" or "1." If you have a situation where multiple outcomes are possible, asking a binary question helps narrow down the possibilities.?

A binary question that removes 50% of the possibilities provides 1 bit of information.

Imagine I have a meeting organized to acquire some missing information about a feature for our next product release. I have assessed that I need answers to 3 binary questions i.e. I need 3 bots to acquire. However, after the meeting I have acquired only 2 bits, meaning there is still 1 bit of missing information. The 1 bit that wasn’t acquired represents the remaining uncertainty or the critical piece of information that wasn’t addressed during the meeting.

Now let’s calculate the efficiency of the meeting based on the information gathered using Knowledge Discovery Efficiency (KEDE).?

Knowledge Discovery Efficiency (KEDE) quantifies how efficiently meetings reduce uncertainty or discover missing knowledge[2]

It is a measure of the effectiveness of the meeting in resolving key unknowns, where H represents the amount of remaining uncertainty (in bits).

As more information is gathered and uncertainty is reduced, KEDE increases, indicating higher efficiency. However, if critical information remains unresolved, KEDE reflects the diminished efficiency due to the remaining gaps in knowledge.

Let’s apply your formula for KEDE:

Where H is the remaining missing information, which is 1 bit in this case (since 3 bits were expected, and only 2 were acquired, leaving 1 bit).

In our example,? KEDE=0.5 means that the meeting was 50% efficient in terms of the information it provided. You aimed to gather 3 bits but only gathered 2, so half of the expected information was obtained.

This result highlights that there’s still some unresolved uncertainty, and additional follow-up is necessary to fully close the information gap.

This model can help evaluate the success of the meetings in terms of information efficiency and can also guide improvements in how you structure and conduct meetings.

If the meeting is significantly less than 100% efficient, it indicates that more planning or better communication strategies might be needed to ensure all required information is acquired.

2. Combining KEDE with Number of People:

We introduce an adjusted efficiency metric Information Efficiency per Person (IEP) that combines KEDE to take into account the number of people in attendance, so that you're not only measuring how efficiently the meeting reduced uncertainty (gathered information) but also how efficiently it used the human resources (the attendees).:

• H reflects how much uncertainty remains unresolved (the number of missing bits).
• N is the number of people attending the meeting.

This metric can help in planning future meetings by revealing whether the number of people attending is proportionate to the information gained.

Higher IEP: Indicates that the meeting was highly efficient in terms of reducing uncertainty and using people's time. A higher IEP suggests that fewer people were able to resolve significant uncertainty, meaning the meeting was well-organized and productive.

Lower IEP: Indicates that although some information may have been acquired, the meeting involved a large number of people, which could indicate an inefficient use of resources. A low IEP might suggest that the same amount of information could have been gathered with fewer participants.

IEP emphasizes both the informational aspect and the human resource efficiency in the meeting, which aligns perfectly with the goal of optimizing both information gathering and attendee participation.

3. Preparing Follow-Up Questions to Refine Understanding

Once you’ve narrowed down possibilities with broad questions, follow-up questions help you refine your understanding and move toward a decision. Here, concepts from decision trees and multistage decision-making help you systematically reduce uncertainty in stages.

Practical Steps:

• Start Broad, Then Get Specific: Begin with general questions that remove large chunks of uncertainty. Follow up with more specific questions that target the remaining possibilities.
• Ask "What’s Missing?": After answering a broad question, ask, “What key uncertainty remains?” This will help you formulate follow-up questions that focus on resolving the remaining uncertainty.
• Anticipate Possible Outcomes: For each broad question, think ahead about the possible answers and prepare follow-up questions that address those outcomes.

Approach:

• Binary Search Style: Think of each follow-up question like a step in a binary search. After a broad question eliminates half of the alternatives, ask a more specific follow-up to divide the remaining options further.
• Decision Tree Approach: This involves creating a decision tree, where each branch represents a question that eliminates some uncertainty. The tree’s branches get narrower with each question, leading to a final decision.

Formal Steps in Algorithmic Terms:

1. Define all possible alternatives (outcomes, solutions).
2. Construct an initial broad question that splits the possibilities roughly in half (inspired by binary search logic).
3. Calculate the remaining uncertainty (or number of possibilities left) after each question.
4. Prepare follow-up questions that further divide the remaining alternatives (guided by a decision tree model).

Practical Example:

Let’s say you’re preparing for a meeting to clarify "What" needs to be developed for a new software product. Here's how to structure your questions for maximum information acquisition:

1. Initial high-level question (1 bit of information): "Is the primary purpose of this software to solve an existing problem or to create a new opportunity?" (Problem/Opportunity)
2. Follow-up questions using a decision tree model:

This approach systematically reduces uncertainty, allowing you to gather the key information needed to drive effective decision-making.

How to cite:

Bakardzhiev D.V. (2024) Efficient Meetings : A Knowledge-centric approach https://docs.kedehub.io/kede-manage/kede-meetings.html

Works Cited

1. Shannon CE. (1948), A Mathematical Theory of Communication. Bell System Technical Journal. ;27(3):379-423. doi:10.1002/j.1538-7305.1948.tb01338.x

2. Bakardzhiev, D., & Vitanov, N. K. (2022). KEDE (KnowledgE Discovery Efficiency): a Measure for Quantification of the Productivity of Knowledge Workers. BGSIAM'22, p.5. Online: https://scholar.google.com/scholar?oi=bibs&hl=en&cluster=1495338308600869852