Realizing Decentralized Economies of Scale

Five years ago, a senior executive painted me a nautical picture of her large, established organization’s goal for digital transformation: to operate not like an unwieldly tanker but as a flotilla of sailboats. Via the collective actions of small cross-functional teams, the organization would sense and seize opportunities to continuously transform—just like the nimble start-ups in its exceedingly competitive industry.

Data from a recent MIT CISR survey[foot]This research briefing is based on data from the MIT CISR 2022 Decision Rights for the Digital Era Survey. Respondents (N=342) comprised organizational leaders representing 61 organizations with annual revenues of at least US$3 billion and 272 organizations with annual revenues under US$3 billion. Fifty-seven percent of the organizations represented in the survey operated outside of North America.[/foot] confirmed that it is indeed possible for large, established organizations to operate in an equally agile fashion as their smaller industry peers, particularly if the organizations decentralize their decision-making. To realize this dynamic capability,[foot]Dynamic capabilities represent the ability to integrate, build, and reconfigure internal and external competencies to address rapidly changing environments, as described in S. L. Woerner, L. Owens, and C. M. Beath, “Build Eight Dynamic Capabilities for Digital Business Model Change,” MIT CISR Research Briefing, Vol. XXI, No. 8, August 2021, https://cisr-mit-edu.ezproxy.canberra.edu.au/publication/2021_0801_DynamicCapabilities_WoernerOwensBeath.[/foot] leaders retain strategic decision rights (the authority and accountability for what the organization needs to achieve and why) but distribute operational decision rights (the authority and accountability for how to best achieve strategic goals) to teams that are closest to customers, offerings, technology, and processes. This approach, also referred to as creating autonomous, empowered, or self-managing teams, focuses on teams realizing outcomes—as opposed to leaders dictating processes and required output—through the iterative realization of solutions that are desirable, feasible, and viable.[foot]For a case example illustrating attributes of desirability, feasibility, and viability, see N. O. Fonstad, “Innovating Greater Value Faster by Taking Time to Learn,” MIT CISR Research Briefing, Vol. XX, No. 2, February 2020, https://cisr-mit-edu.ezproxy.canberra.edu.au/publication/2020_0201_InnovatingGreaterValueFaster_Fonstad. [/foot]

Today, the aforementioned organization has made great strides in decentralizing decision-making in its digital and IT teams, but leaders in other parts of the enterprise are slow to give up their operational decision rights. Our recent survey showed this is a common struggle: the leaders in our survey reported that on average 47 percent of the teams in their organization (or the part of the organization they were most familiar with) could make decentralized decisions.[foot]The decentralized decision-making questions in the survey asked about the extent to which (and what percentage of) teams were enabled to make decisions without a manager’s oversight, decide for themselves how to best solve customer- or business-focused problems, revise their solutions when conditions change, and define their own performance targets and commitments.[/foot] Yet our survey data showed that granting operational decision rights to only a select few business units or corporate functions hindered an organization’s ability to sense (i.e., identify and analyze) and seize (i.e., decide on and experiment with) opportunities, thereby limiting its innovative capacity and financial performance. This research briefing describes the extent to which decentralized decision-making can help large, established organizations achieve greater organizational agility, and which practices help to turn size into an advantage for improving organizational performance.

Organizational Agility Requires Teams’ Ideas

Organizational agility in the digital age requires organizations to proactively put their teams on the trajectory of customer needs and technological development. Doing so enables teams to imagine a bold future with new solutions that benefit the organization and its various stakeholders. As successful solutions are a function of the number of hypotheses that teams can test, organizational agility depends on how swiftly and frequently teams move through what I call the Ideas Cycle (see figure 1), which represents the process through which teams with operational decision rights identify opportunities—problems or unmet needs—then analyze, decide on, and experiment with ideas to address them. This cycle consists of four stages:

• Identify problems to solve or unmet customer or business needs to fulfill. These can be surfaced by interacting directly with customers, keeping up to date on organizational developments, frequently discussing new technologies with peers, or regularly engaging with digitally savvy organizations or institutions.
• Analyze the problems and needs to make sure teams understand them well enough to formulate adequate hypotheses regarding potential solutions. This may, for example, require teams to analyze publicly available or proprietary data, examine the successes and failures of prior solutions, or reevaluate existing work processes and practices.
• Decide which hypotheses to test, based on their potential impact and ability to provide useful insights. While teams may have the decision rights to proceed with experiments that involve relatively small, predictable, or reversible outcomes, they might need to involve other organizational stakeholders to run experiments that are more complex and high stakes—affecting the speed at which teams can proceed.
• Experiment by mobilizing the organization’s expertise, funding, and other resources to test chosen hypotheses and learn what works.

Figure 1: The Ideas Cycle

A team may ultimately need to pursue several rounds of the Ideas Cycle to develop a desirable, feasible, and viable solution. How quickly teams can move through the cycle, however, depends on the extent to which its decision-making is decentralized. We observed in our survey that when most teams had operational decision rights, large established organizations could be just as agile as their smaller peers that took the same approach (see figure 2). Teams in large decentralized organizations required less than half the time of teams in large centralized organizations to sense and seize business, customer, and technological opportunities—on average approximately 244 days versus 566 days. We also saw this difference in agility reflected in measures of organizational performance: compared to their large centralized peers, respondents in large decentralized organizations reported average net profit margins and revenue growth rates that were respectively 6.2 and 9.8 percentage points higher, and their revenues from products and services introduced in the last three years (a key measure of innovation) were 1.5 times as high.[foot]In our survey, we classified organizations as decentralized if ≥50 percent of teams had operational decision rights. The performance numbers indicated differences between centralized and decentralized organizations with at least US$3 billion in revenue. Corresponding differences between centralized and decentralized organizations on net profit margin (when adjusted for industry averages) were -4.0 percent vs. +2.2 percent; on revenue growth (industry adjusted) were -4.8 percent vs. +5.0 percent; and on percent of revenues from products and services introduced in the last three years were 19.4 percent vs. 28.8 percent.[/foot]

Guide Decentralized Decision-Making with Minimum Viable Policy

Despite the potential benefits of decentralized decision-making, managers may be hesitant to relinquish their control over teams, concerned that decentralization might lead to inefficiency or misalignment with the organization’s strategic objectives. To identify ways to mitigate such reluctance, I assessed fifty practices from the survey data that make up different categories of decision rights guardrails[foot]For a discussion of decision rights guardrails and description of guardrail categories, see N. van der Meulen, “Decision Rights Guardrails to Empower Teams and Drive Company Performance,” MIT CISR Research Briefing, Vol. XX, No. 8, August 2020, https://cisr-mit-edu.ezproxy.canberra.edu.au/publication/2020_0801_DecisionRights_Meulen.[/foot]: enabling constraints by which teams can operate with greater meaning, competence, direction, or impact. While nearly all the practices related to higher measures of organizational performance, there was one category of eight practices—those making up the minimum viable policy guardrail—that stood out as particularly effective. Minimum viable policy practices reduce bad complexity and promote the reuse of existing solutions, which enables teams to move through the Ideas Cycle more quickly (as illustrated by the bottom arrow of figure 2) and thereby improves organizational performance (as outlined in figure 3).

Practices to Reduce Bad Complexity

The aim of the minimum viable policy guardrail is to guide teams and safeguard business continuity with the least amount of policy required. Organizations committed to this aim pursue the power of subtraction: their leaders eliminate impediments to team progress, such as excessive or complex policies that increase bad complexity, and instead implement high-level foundational principles that define collective constraints and guide teams’ decision-making. This cuts the average time to make complex or high-stakes decisions in half. Teams, in turn, can challenge existing policies and deviate from established norms—so long as they share their motivations for deviating to surface opportunities for further policy minimization. This motivates teams to identify new opportunities and problems to solve; in our survey, teams from organizations with minimum viable policy practices that were more developed than average did this at three times the rate of teams from organizations with practices that were less developed than average.

Practices to Reuse Existing Solutions

Another way leaders can minimize policies is by fostering habits of reuse, by setting expectations for teams to consolidate solutions in a centralized repository (preferably a digital platform connected to the organization’s operational backbone) and to reuse their peers’ solutions when practical. Our survey data showed that when reuse becomes an organization-wide habit, teams can analyze opportunities and experiment with solutions more than twice as quickly as teams in organizations without this habit. While designing for reuse requires more time, and extra effort from certain teams, this approach enables other teams to focus on solving fresh problems while taking advantage of solutions that already work well. It is therefore important that leaders make the creation and support of reusable solutions attractive to teams, and that they appoint a dedicated group or function (such as enterprise architects) that can help teams to find and promote existing solutions.

Decentralized Decision-Making Needs Different Leadership

While decentralized decision-making may evoke visions of less leadership, I found that what it needs is different leadership—focused on providing context rather than exacting control. Leaders in organizations driven by decentralized decision-making should support teams by removing impediments to team success and by making the reuse of solutions a team habit. The result is an organization where as many teams as possible benefit from enterprise scale, enabling them to rapidly sense and seize opportunities that help their organizations thrive in ever-changing environments.

© 2023 MIT Center for Information Systems Research, Van der Meulen. MIT CISR Research Briefings are published monthly to update the center's patrons and sponsors on current research projects.