Organizational decision making: An information aggregation view

Paper highlights

Organizations often possess more knowledge than any one member, but they still need a process for turning that dispersed knowledge into a decision. This paper compares four such structures: relying on one individual, delegating each problem to an expert, voting, and averaging members’ estimates.

No structure dominates. Delegation performs best when expertise is diverse, the organization can identify the right expert, and its knowledge matches the problems it faces. When those conditions fail, voting or averaging can be more accurate and more robust.

How the model works

The formal model gives individuals different domains of expertise and lets projects vary in the knowledge they require. It then evaluates how accurately each structure assesses those projects. The analysis varies the breadth of knowledge inside the organization, errors in assigning projects to experts, and changes in the external environment.

Main results

• If expertise is narrow but complementary, averaging can combine partial knowledge effectively.
• If members have broad knowledge and delegation is accurate, sending each decision to the best-matched expert works well.
• If expertise is hard to measure or delegation is noisy, voting is a safer choice.
• Radical environmental change can reverse a structure’s advantage by changing the fit between existing expertise and the problems being evaluated.

Why it matters

• Dispersed knowledge helps only when the decision rule can recover it: averaging, voting, and delegation lose different kinds of information.
• Delegation is powerful when expertise is specialized and accurately located, but brittle when problems are routed to the wrong expert.
• The same structure can succeed in one knowledge environment and fail when expertise or incoming problems change.

How to use this paper

Cite this for

• A formal comparison of individual judgment, expert delegation, majority voting, and averaging.
• Conditions under which delegation works: diverse expertise, accurate routing, and fit between expertise and incoming problems.
• A model of why environmental change can reverse the advantage of an information-aggregation structure.

Useful for teaching

• Why dispersed knowledge needs an aggregation rule before it can improve decisions.
• How expertise breadth, routing accuracy, and voting or averaging rules interact.
• Why no single decision structure should be treated as the default answer for organizations.

Careful claim

The model identifies contingencies under which delegation, voting, or averaging perform well; it does not rank any one structure as generally superior across organizations.

Abstract

Abstract We study four information aggregation structures commonly used by organizations to evaluate opportunities: individual decision making, delegation to experts, majority voting, and averaging of opinions. Using a formal mathematical model, we investigate how the performance of each of these structures is contingent upon the breadth of knowledge within the firm and changes in the environment. Our model builds on work done in the Carnegie tradition and in the group and behavioral decision making literatures. We use the model to explore when delegation is preferable to other structures, such as voting and averaging.

Our model shows that delegation is the most effective structure when there is diversity of expertise, when accurate delegation is possible, and when there is a good fit between the firm’s knowledge and the knowledge required by the environment. Otherwise, depending on the knowledge breadth of the firm, voting or averaging may be the most effective structure. Finally, we use our model to shed light on which structures are more robust to radical environmental change and when crowd-based decision-making may outperform delegation.

Other links

• SSRN preprint

Last updated 2026-06-21