The flattening of hierarchies transforms the workforce.
Published on:Mit Sloan Management Review Italy, August/September/October 2025. Year 4. Number 4.
Today's managers are aware of the importance that corporate organization has for company performance, thereforefrequently modify the company structure in order to improve results. However, the effect such changes can have on the workforce itself is less understood.
Although research has shown that employees are a diverse group and that attracting and retaining talent requires a combination of incentives, we know less about how various types of organizational structures attract different workers and whether those structures tie them to their employer or push them to leave the company (Raman and Rosenblum, 2023).
The more radical changes senior management intends to implement, the more critical this question becomes. Among the most significant transformations observed in today's corporate landscape is the transition from a traditional and hierarchical organization to flatter structures withfewer hierarchical levels. These structures offer greater autonomy, but also place the burden of self-organization on employees (Greer and Klotz, 2019; Laloux, 2014).
Top management must seriously reflect on which structural changes to implement and which complementary measures to adopt to adapt the composition of its workforce to its needs. But first and foremost, it needs to better understand how the shift from a hierarchical to a more self-organized work structure can impact the composition of the workforce. It needs to understand what types of people will be attracted to the company, which ones will be likely to leave, and how individual circumstances may affect the outcome.
To develop someinitial guidelinesto implement a flatter structure, we conducted a large-scale empirical study using data from over5,500 companiesof the US financial services industry that had significantly reduced hierarchical levels between 2000 and 2022. We found that, on average, one year after simplifying the corporate structure, the workforce has a higher percentage of conscientious, affable and open individuals. We then conducted in-depth studies on two relevant companies to better understand what drives the differences between the organizations.
Because “structure follows strategy” is too simplistic
Many high-level strategists agree with the scholar's influential maximAlfred Chandler, dating back to 1962: “If the structure does not follow the strategy, inefficiency results”. In practice, this means that a leader must consider their customers, offerings, and value delivery before dividing work among their people. And only after having distributed all the tasks among the employees, the leader will know how to reintegrate all the contributions of the individuals, channel their communication, give them decision-making rights and, finally, assign managerial roles. In other words, only then will the leader understand how to structure the company to achieve collaboration (March and Simon, 1958).
It is precisely this type of contingent thinking that has led to a rather standardized approach to dealing with structural issues, according to which high-level managers can choose between a number of stereotyped organizational models, depending on the context in which they find themselves. Many leaders, with disturbing automatism, still tend to implement stereotyped organizational forms, depending on the circumstances. As their companies get larger, they move to a divisional structure; when downsizing, they prefer a functional structure. And when everything becomes complex and dynamic, they may choose to move to the so-calledM shape, with moresemi-autonomous divisions.
However, this approach is at odds with today's business reality, where objectives other than strategy execution have gained importance. Some leaders may seek to experiment with different structures in order to promote the learning and adaptation needed to identify new opportunities. At the end of the 1980s, at the Danish hearing aid manufacturer Oticon, theCEO Lars Kolindintroduced the term “organizing spaghetti” in an attempt to revive the fortunes of the company. The non-hierarchical, non-divisional, not function-based, but project-based form of working of a spaghetti organization was intended to discover and identify new viable business ideas (DeFillippi and Lehrer, 2011). Much of what Bill Anderson, the current CEO of Bayer, is doing now in his organization is reminiscent of that approach (Cutter, 2024).
In any case, we have seen that organizations pursuing similar goals can differ significantly when it comes to structuring work. A myriad of contemporary management approaches, such as (scalable) agility, loscrum of scrumsor Haier's RenDanHeYi, are based on an organization with fewer levels of managerial authority and agreater autonomy of the workforce. These different structures appear to work for very different people, regardless of the strategy companies pursue (Reitzig, 2022).
Explore employee reactions to company structure
The fact that people react differently to different company structures is no surprise: each of us is different, and employees value and expect different things from their workplace. Research shows that employees perform best in jobs that best match their abilities and that they prefer to work in organizations with similar values to their own. Importantly, the literature clearly demonstrates thatcreating a better fit between people, their jobs and their organizations not only reduces involuntary turnover, but also helps employers attract the most sought-after talent. However, this line of research has not yet addressed the question we aim to address here, namely which people might be attracted to which type of organizational structure. This is relevant for strategists who are currently experiencing the move to less hierarchical structures and who need to understand the implications for talent.Will the transition from more hierarchical structures to less hierarchical structures attract new people to the company or motivate current employees to leave? And if there is such an effect, will workers naturally transition to the management structures best suited to them?
Our results are clear. First, by controlling for firm-specific effects that remain constant over time, as well as by controlling for firm size and particular year effects (macroeconomic events), we can demonstrate that the workforce actually changes as a result of reducing hierarchical levels[1]. Secondly, we can prove that,on average, employees become more conscientious, affable and open; that is, they score higher on three of the five main personality traits that many HR professionals use to evaluate their employees (the other two are extraversion and neuroticism). To quantify the effects: when a company with a five-level hierarchy eliminates a managerial level in a given year, in the following year theaverage score of the workforcefor conscientiousness will increase by2.0%; theaverage score for affabilitywill increase by3.6%; and theaverage score for the openingwill increase by3.4%. These findings are relevant in absolute terms: many organizations compete to hire talent with these characteristics and this has a direct impact on performance. But they also indicate that reducing hierarchical levels leads to the self-selection of workers into different types of structures, which should help increase productivity. Here's why: When companies reduce the depth of their hierarchies, the scope of control of the remaining managers expands and, as a result, they can no longer micromanage their direct subordinates, but must delegate more decision-making to workers. These employees should therefore be able, and indeed must, organize their work to a greater extent. Individuals who particularly value this new autonomy and thrive in flatter environments are typically very conscientious, show astrong need for achievementand promote new initiatives and projects.
At the same time, personable people help ensure that self-organization does not lead teams to choose only the tasks that excite them and avoid those that do not specifically benefit or appeal to them. In other words, a team must include people willing to carry out support tasks on everyone's behalf, for example organizing the Christmas party.
It is important to underline thatthe overall level of qualification of the workforce does not change following the restructuring: it neither increases nor decreases. This underlines our previous point:there is not necessarily a single optimal structurethat encourages an organization's most skilled workers to stay; Among talented staff members, different structures attract different individuals[2].
These findings raise the question of whether the net effects we have described are the result of the arrival of new employees or the departure of workforce members. Given the transition to a less hierarchical structure, what types of individuals will be attracted to the company and which will be inclined to leave?
On this point too, our results are unequivocal: the overall changes in the workforce resulting from the structural change were attributable to employees who had decided to leave their employer, not to those who had just joined. In other words, the structure affects workforce retention or attrition, but does not appear to attract newcomers. This may be because the structure and the effect it has on employees' perceptions of autonomy can be better understood through personal experience rather than through the description of, for example, a hiring manager. In this way, it has an effect on current employees, but not necessarily on potential candidates.
In-depth analysis of working contexts
Like any large-scale empirical research, our study also presentsadvantages and disadvantages. While our findings are based on LinkedIn data on episodes of hierarchical de-stratification in the financial sector over a 20-year period, this dataset does not provide detailed insights into individual cases like a small-scale interview study might. Therefore, to complement our findings, identify potential boundary conditions, and provide further insights, we conducted two in-depth case studies with companies that have moved towards creating less hierarchical self-organizing environments. We listened to their experiences and discussed their interpretations of the data related to changes in their companies' workforce. The goal was to shed light on the extent to which individual case conditions can influence the overall picture.
Our first reference company is a major US bank holding company that undertook a series of restructuring initiatives from 2000 to 2020. The latest reorganization during that period led the entire company to adopt a team-based working approach, following the scalable agile methodology. We interviewed a former master agile coach at the company, who told us that the aspiration was to deliver a better product in a more timely manner,”doing more with less”. The coach was tasked with helping people understand agility and work across different businesses, teams and project portfolios. The change involved moving from a very traditional hierarchical structure based on command and control to one where autonomy and self-organization would become the new expectation. As we found in our industry-wide analysis, average employee conscientiousness steadily increased throughout the transformation, not surprising to the master agile coach.
The coach interpreted this change as reflecting the fact that people with a strong need for affirmation, particularly programmers, could better showcase their work. For example, more people, even senior-level people, started attending programmers' presentations. Likewise, friendliness and openness among employees steadily increased during the transformation, much to the satisfaction of our interviewee. “One of the hardest things to break down is the 'my idea is the only idea' mentality. Agile transformation attracted people with an open attitude,” he said. Firing people with an outdated mindset probably also contributed to some of the overall effect, but was probably less important.
Two other findings are interesting, although they differ from our sector-level analysis. First, the average skill level of the company's workforce has increased. However, this did not happen spontaneously. In parallel with the agile transformation, the company sought to hire people with different skills. “Before the transformation, we didn't need one person who could cover the entire stack. Everyone was a specialist.", our interviewee told us. Because the company wanted to be the first in its industry to fully exploit the resources of Cloud computing, it changed its hiring scheme to prioritize recruiting people with the skills needed to work with what was then a disruptive technology.The percentage of extroverted employees has also increased and other research has shown that these people tend to feel happier and be more productive in non-hierarchical contexts compared to more rigid environments.
Our second company is a major payment processing company that implemented significant company-wide work restructuring from 2014 to 2017. Like the bank holding company, this organization also transitioned to team-based working and followed the scalable agile methodology. According to our interlocutor, a former director of the company's agile transformation program, the initiative was aimed at countering some of the challenges the company was facing at the time, including employee turnover, low morale and sub-optimal productivity. A total of3 thousand employees, approximately three-quarters of the workforce at the time, were brought together into agile teams and encouraged to grow in a productive mode of self-organized work. The preparation lasted approximately18 monthsand included team formation and role assignment, as well as self-coordination training. According to our interviewee, the initiative has borne fruit with an increase in productivity: in 2016 and 2017, the company carried out the same volume of work as before, but with fewer people. However, as time passed, patterns in workforce composition began to look very different than what was suggested by both our industry-wide study and the company's initial analysis. Data from LinkedIn, which covers about half of the company's workforce, indicated a decline in the average skill level of employees.There was also a decline in average friendliness, conscientiousness and open-mindedness.
Our conversations with the transformation director suggest some possible explanations. Our interviewee admitted that, in hindsight, “focusing on the exploitation of the mass of people may have resulted in a lack of attention towards some highly qualified experts”, creating unintentional collateral damage.Some highly qualified experts felt sidelinedbecause they no longer felt like the reference "problem solvers" and therefore they left the company; they were not encouraged to leave. The same was true for some of the potential top performers who scored high on conscientiousness.
The two cases, both involving financial sector companies that were pursuing very similar approaches to reducing hierarchical control, provide useful insights that complement our large-scale analysis. Above all, they demonstrate that transformations from a hierarchical structure to more self-organized work do not always lead to more efficient workforce selection.It is important to consider individual career goals: Although some employees value autonomy more than others, they still want their work to be valued and highlighted, and it is important to provide them with positive feedback and visibility. Self-employed employees need more than just guidance on the path they should take (Greer and Klotz, 2019). The route must also be attractive to them. The cases clearly demonstrate that simply adjusting the organizational structure does not change the average level of competence.
Several factors can make a big difference in getting the most out of an organizational transformation. Many of them can be company-specific. Our case comparison suggests that two underlying factors can often be crucial to the outcome of the trial. One is prior learning and experimentation. While Company 1 had been experimenting with agile work environments for some time before undertaking the company-wide transformation, Company 2 took the plunge all at once. The second factor is the coordination of top management and integrated communication. Company 1 had the full support of the CEO and CIO, knowing that business and delivery would need to work hand in hand to ensure the success of the transformation. Initiatives in Company 2 were led exclusively by the CIO.
In summary, our large-scale analysis and case work demonstrate that organizational restructurings are likely to change the workforce, as individuals move to the environments they prefer, albeit not always efficiently.Restructuring will not change the skills of the workforce, but in some cases it may help attract better talent.
Implications for managers
CEOs develop strategies. Chief Operating Officers (COOs) implement the structures. Chief Human Resources Officers (CHROs) are responsible for the workforce. However, only a few HR leaders have a say when CEOs and COOs rethink their organization's business model and reorganize work. This is probably an error.
As our study shows, people's preferences for different work environments vary. That's why designing a business structure without thinking about its potential effects on your workforce can have unintended consequences.
Flattening hierarchies to become more agile can increase the percentage of conscientious, open and affable individuals in the total workforce – precisely the people needed to work in such an environment. While this is good news for companies aiming for greater self-organization of their teams, our results also show that the effects are only moderate and are mainly driven by attrition.
To fully reap the benefits of a flatter structure, CHROs should be involved early in corporate restructuring initiatives. CHROs can help supplement the natural effects of turnover with additional recruiting efforts aimed at attracting workers best suited to the new structure. They can createroadmapof Human Resources strategy linked to organizational objectives and characterize the profiles of skills and experiences suitable for teams that work with significant autonomy (Bosset al., 2023). Theseroadmapcan draw on performance evaluations and psychological assessments of the current workforce to identify who is able to work more autonomously and coordinate independently.
But CHROs can and should also coordinate with COOs in redefining roles as part of a transformation initiative. The reduction of hierarchical levels involves the revocation of some decision-making powers of managers to guarantee greater autonomy to their (former) subordinates. Redefining positions to retain such individuals requires leaders to think about non-monetary compensation mechanisms. Deeply understanding the motivations of these individuals and creating new roles for them outside of the traditional hierarchy is a task that COOs and CHROs can best accomplish together. For example, is it possible to create a specialist role for the former mid-level IT manager stripped of hierarchical power following a streamlining initiative?
To have a say in these decisions and be trusted partners in making strategic and organizational decisions, CHROs must look at the organization through a broader lens than traditional HR considerations. That's because CEOs, CFOs, and COOs think about structure in terms of hierarchical depth, divisional and functional specialization, and breadth of control. Understanding how these macro concerns translate into micro job and role decisions, and vice versa, will be critical to CHROs' ability to have useful conversations with the leadership team.
Finally, a restructuring requires all senior managers to adjust the way they coordinate and supervise subordinates. This last aspect has many facets, but a few points stand out in particular: Repressing the impulse to micromanage more autonomous employees does not mean that managers should not provide guidelines for the work. In fact, the greater the autonomy granted by managers to their subordinates, the greater the need to design workflows and processes in order to facilitate employee self-coordination.
In this context, the creation of transparent, modular and project-based structures that enable efficient self-selection of workers into projects, avoid duplication of team efforts and enable smooth cooperation between rather independent groups becomes crucial to maintaining company productivity. Instead of continuing to personally direct every single initiative, successful leaders will increasingly transform themselves into designers of workplaces that meet both the company's goals and the preferences of their employees.
But no matter how well designed these workplaces are, they too will generate friction. And then it will be the responsibility of leadership to help resolve them, providing support structures for peer conflict resolution when possible, and making themselves available to employees when things get (too) difficult. Clearly, this is a team effort that requires all the skills and experience that a top-tier management team has to offer. True human-centered leadership must merge strategic considerations, structural reflections and employee needs from the beginning, because not only does structure follow strategy, but people also follow structure.
Notes and bibliography
[1] We also checked company performance, where such data was available, in a series of subsample tests. The results remained constant.
[2] We have no reason to believe that our results are affected by forced turnover (layoffs) or changes in individual salaries. From further studies we have conducted, we know that particularly conscientious and open employees voluntarily stay longer in flatter hierarchies and vice versa.
A. Raman and E. Rosenblum, “Focus on Skills to Grow Your Workforce,” MIT Sloan Management Review, June 21, 2023, https://sloanreview.mit.edu.
L. Greer and F. Klotz, “Teams Still Need Leaders,” MIT Sloan Management Review 61, no. 1 (Fall 2019): 15-17; and F. Laloux, “Reinventing Organizations: A Guide to Creating Organizations Inspired by the Next Stage of Human Consciousness” (Brussels: Nelson Parker, 2014).
A.D. Chandler, “Strategy and Structure: Chapters in the History of the Industrial Empire” (Cambridge, Mass.: MIT Press, 1962).
J.G. March and H.A. Simon, “Organizations” (Hoboken, NJ: John Wiley & Sons, 1958).
R. DeFillippi and M. Lehrer, “Temporary Modes of Project-Based Organization within Evolving Organizational Forms: Insights from Oticon's Experiment with the Spaghetti Organization,” in Research in the Sociology of Organizations (Advances in Strategic Management, Vol. 28),” edited by L. Frederiksen, F. Täube, S. Ferriani, et al. (Leeds, England: Emerald Group Publishing, 2011), 61-82.
C. Cutter, “A CEO's Radical Solution to Corporate Problems: Eliminate the Bosses,” The Wall Street Journal, March 22, 2024, www.wsj.com.
M. Reitzig, “Get Better at Flatter: A Guide to Shaping and Leading Organizations With Less Hierarchy” (Cham, Switzerland: Palgrave Macmillan, 2022), chap. 4.
Greer and Klotz, “Why Teams Still Need Leaders,” 15-17.
V. Boss, L. Dahlander, C. Ihl, et al., “Organizing Entrepreneurial Teams: A Field Experiment on Autonomy Over Choosing Teams and Ideas,” Organization Science 34, no. 6 (November-December 2023): 2097-2118.
BOX. The research
The authors created a dataset from publicly available information on LinkedIn in 2022 provided by Bright Data. It included all U.S. financial services companies on the platform, as well as the educational and career paths of their employees over the past 20 years.
Based on observations of 5,500 companies that had flattened their hierarchies, they empirically modeled changes in workforce composition (average education, age, affability, conscientiousness, extroversion, and open-mindedness) as a function of company structure and critical control variables.
This research approach, although validated in previous studies, is influenced by employee coverage rate on LinkedIn and limited to public information. Therefore, the authors supplemented their large-scale study with two in-depth interviews with companies in the financial services sector that had restructured to move towards greater self-organization.