They say the world is flat and supported on the back of four elephants who themselves stand on the back of a giant turtle.

They say that the elephants, being such huge beasts, have bones of rock and iron, and nerves of gold for better conductivity over long distances. (Not rock and iron in their dead form, as they are now, but living rock and iron.)

Terry Pratchett (1999: p. 9; emphasis as in original)

On the way to a flatter world?

Judging from the flurry of books on the subjects, one may believe that the world is on the way to be seen as a flat disc once again—the world of organizations, at least. While these much hyped ‘decentralized’ and ‘low hierarchy’ organizational designs carry a variety of different names (see, e.g., Bernstein et al. 2016; Ketkar and Workiewicz 2022; Martela 2019 for wonderful overviews), the key claim with which they are all associated is similar: moving to flatter organization designs with fewer (or even zero) levels of hierarchy between top managers and frontline employees implies performance improvements for the organizations choosing to do so. Prominent examples include firms, such as the highly physical tomato processor Morning Star as well as the highly digital online retailer Zappos.

Questions aside about whether flat organization designs are truly new (Puranam et al. 2014; Romme 2019)—for example, the idea of autonomous teams self-determining how to execute to achieve a pre-specified goal can be traced back at least to king Gustav II Adolph of 17th-century Sweden—claims of general superiority should always be met with a healthy dose of academic skepticism. First, as with the efficacy of any organization design being highly contingent on the firm’s strategy and competitive environment (e.g., Burns and Stalker 1961; Donaldson 2001; Lawrence and Lorsch 1967), we should expect that flat organizations should not simply always work, but be tools and templates that promise higher levels of efficiency or efficacy only in specific settings. Second, even if flat organizations were always better, there may still be variance in their effects: some forms should work better rather than others, given a specific situation and purpose. And third and finally, even if flat was always good, and potential performance differences between different forms thereof negligible, we would still expect that varying firms—given they start from difference resource endowments (Barney 1991; Penrose 1959)—should face idiosyncratic struggles in attaining the benefits of (various forms of) flat.

Markus Reitzig’s “how to get better at flatter designs”

These key questions—(1) what the promise of ‘flat’ truly entails, (2) when and why various approaches to ‘flat’ may bring benefits to firms, and (3) how firms may go about transitioning toward a flatter organization design—are central to Markus Reitzig’s book Getting Better at Flatter: A guide to shaping and leading organizations with less hierarchy.

In a balanced manner, he begins by critically assessing the costs and benefits of a move to flat. He sharply observes (Chapters 1–2) that while a move to a flatter hierarchy may eliminate ([some] middle) managers from the org chart, the tasks traditionally associated with managementFootnote 1 cannot miraculously disappear from the actual flow chart of activities an organization needs to perform (also see Puranam et al. 2014). Given the cognitive limits of top management—how should they possibly do all the additional work that a lower level of management had done previously?—and absent near-all-powerful support tools, such as artificial intelligence directing the firm, Reitzig persuasively argues that these managerial tasks will need to be distributed among employees. Put differently, moving to a flatter organization forces executives to grant employees more autonomy—the discretion to make ‘managerial’ decisions about their own work.

In turn, the promise of such autonomy needs to be sufficiently luring to a large-enough share of the workforce to accept this responsibility on top of their day jobs (Chapter 3). Specifically, summarizing insights from management, economics, and psychology, Reitzig carves out how varying aspects of the promise of autonomy (e.g., ‘can I pick my own work’ or ‘can I decide how to conduct my work’) not only map differently on the former managerial tasks, but entail different motivational and performative responses across individuals, setting the stage to answer the crucial ‘it depends’ question: conditional on what improvements executives hope to attain with a flatter organization design (e.g., ‘faster time-to-market’ or ‘employees more strongly identifying with the firm’), executives may find that they should opt for quite different variants of flat, through a well-balanced configuration of managerial tasks they want to delegate to the workforce.

Reitzig’s analysis up to this point is clear and convincing—possibly, the discussion of existing and well-known types of flat organizational forms (such as holacracy) that he has reserved for Chapter 12 could have already been included at this point to give academically inclined and organization-fad-informed readers a chance to link his arguments more directly to some prototypical flat models.

Understandably though, given Reitzig targets executives actively considering making the change to flat, he goes on to lay out how executives should think about facilitating this change, by looking at how to align employees with the transition (Chapter 4), the kind of management (support) still needed in flat organizations (Chapter 5), and the organizational structure and workflow executives may design (Chapter 6). As before, Reitzig continuously manages to show impressive breadth across a variety of literatures to substantiate his argument and leverages this insight to lay out a broad portfolio of design choices executives considering flat need to consider. In turn, by consistently applying the same logic of which managerial tasks need to be performed somehow in the organization (task division and allocation; information and reward provision; exception management) to shed light on the various flat design choices, and illustrated by clear examples, Reitzig helps the reader not to get lost in the broad configurational space he lays out.

Across Chapters 7–9, Reitzig brings the previous points together to highlight when (Chapter 7) and until when (Chapter 8) flat promises particular advantages, and provides detailed guidance on finding the sweet spot (Chapter 9). Drawing on the most suitable application areas for flat organizations (i.e., settings demanding more creativity and innovation, or higher identification with the organization), Reitzig highlights how executives need to determine how various approaches to (less) flat organizations differently affect the organization’s ability to collect, assess, and enact information. As such, a smart flat organization should not just consider which managerial tasks it delegates to employees, but also which support structures to create to prevent potential coordination challenges that a hierarchical design might address by default. Finally, Chapters 10 and 11 contain two detailed cases on the two challenging high-level processes coming with becoming flatter: the actual transition to flat (Chapter 10) and growing the flat organization (Chapter 11). The chapters nicely complement the helpful examples Reitzig gives throughout the book of companies that have successfully embodied some elements of flat organization through a deep dive. Finally, the aforementioned chapter 12 reflects on approaches to flat organizing broadly.

Overall, Reitzig easily manages to find the much-needed balance between an academically rigorous discussion of the boundary conditions determining the potential use of flat organization designs and a managerially relevant explanation and instruction set of how to do so, when, and why. Beyond Reitzig’s own expertise on the subject, his broad reading of a variety of literatures from different disciplines leave the reader with a feeling of substance—that the book’s recommendations are both evidence-based and up to date. The configurational perspective shining through Reitzig’s argument is particularly laudable: going flat is more than a few isolated choices, but that the success of such a decision crucially depends on aligning said choices with each other, with what the firm wants to achieve, and with the employees and resources the firm has at its disposal. In addition, the various examples and clear instructions help prevent said content from ever becoming dry. In sum, even if, as Reitzig admits, there will still be a lot of customization to be done by any executive, the book will provide an excellent starting point and holistic structure on how to be better at flatter: applying Reitzig's insights will allow managers evaluate more objectively what aspects of flat organizing a firm may choose when, implement these aspects more effectively, and manage to resulting organization more comprehensively.

What’s next for flat organizations?

Reitzig’s treatise on how executives may transform existing hierarchical organizations toward flatter design—that is, making an existing firm flat(ter) while keeping the goal of the firm and its members largely fixed—is timely, clear, and useful. In turn, the focus on organization that already have reasonably well-specified goals and existing employees working toward those is fully reasonable given the focus of his book: an academically appealing, yet ultimately practically applicable introduction on when and how to move an existing organization and its members to flatter models of organizing.

At the same time, advanced readers thinking about further blurring the limits that clear organizational goals and membership rules represent may see how Reitzig’s ideas could also help pave the way to understand the potential use and usefulness of flatter organization designs more broadly. Specifically, I will argue that we may be able to generalize Reitzig’s points broadly: rather than ‘only’ using a flatter organization design to solve existing and known problems more effectively (e.g., Simon 1973, 1996), I believe that Reitzig’s insights highlight how some approaches to flat organizing may even be conducive to tackle settings in which we do not know what the “right problems” are in the first place (e.g., March 1978, 1987, 1988).

Reitzig focusing on organizations that have a sufficiently clear goal actually sets up his argument in a setting in which large firms would usually still be considered as (at least) reasonably competitive. The intriguing ‘it depends’ Reitzig points out is that, even is this assumption relaxed—meaning the organization would compete in a context characterized by uncertainty, so that it may not reasonably specify clear goals a priori (March et al. 1993)—flatter still need not always be better. To be more precise, I am not suggesting that large, centralized firms would always need specific, all-encompassing goals to function (see also Cyert and March 1963). Yet, the less there is clarity about what the firm’s goal should be—as when the firm is facing uncertainty—the more difficult it should become for the central planners traditionally assumed at the helm of large corporates to define crucial tasks in a way that they are independent of each other as well as accurately matching employees’ skill now and in the future (Burns and Stalker 1961; Lawrence and Lorsch 1967). For such settings, in line with Reitzig’s call for flatter designs, organizational scholars have long recommended establishing additional coordination and communication channels to increase the organization’s ability to filter and process information, and to delegate decision make power to lower tiers of the hierarchy to react this information more speedily and appropriately (e.g., Galbraith 1994; Malone and Crowston 1994).

Yet, as Reitzig notes, the crux is that as uncertainty gets higher, more and more communication and coordination would be required to ensure that the information these empowered agents gather is shared sufficiently widely and interpreted best by the organization, in light of all other information available. Rather, with increasing uncertainty, there may be a level at which too high a degree of decentralizing information processing may lead to the costs of coordination and communication required to assess and act upon information rise faster than any potential benefits of better information gathering. In short: the system may simply become clogged up with too much information and coordination. In line with this argument, a burgeoning literature on new ways of organization innovative activity has noted how contexts characterized by extreme levels of uncertainty, such as ‘new to the world’ technologies or ‘grand societal challenges,’ may often not be addressed cost-effectively by flat organizations (e.g., Afuah and Tucci 2012; Alexy e al. 2012; Felin and Zenger 2014; Knudsen and Srikanth 2014; Piezunka and Dahlander 2015; Wallin et al. 2018; Zenger and Felin 2020). In this view, fundamentally uncertain contexts may possibly be better left to integrated firms. While those may not be as good as information gathering, they should be more effective at processing newly incoming information, and more quickly able to form an internally consistent view of the uncertain world and derive strategy recommendation increasing the odds of the firm’s survival (Argote and Greve 2007; Gavetti et al. 2012; Grant 1996; Kogut and Zander 1992).

However, when also relaxing the second assumption—that a flat organization would (roughly) follow the same boundaries for membership in the organization as its hierarchical predecessor—there may be new perspectives emerging on just what flat organizations can do and how much information they may process effectively (see also Baldwin and von Hippel 2011; Benkler 2016; Choudhury et al. 2020; Hsieh et al. 2018; Majchrzak et al. 2018; Moffett et al. 2021). Specifically, as Reitzig notes himself, a key factor toward the success of decentralized organizations is having excess resources—so-called slack (see also Cyert and March 1963; Galbraith 1973). And what better way to have a lot of slack than to design an organization that may draw, in essence, on a limitless number of additional volunteers from outside of the organization to contribute to the organization’s efforts? Without going into too much detail here, one organization that comes to mind is Hyperloop Transportation Technologies (Majchrzak et al. 2018), HTT, an organizing hoping to deploy hyperloop systems around the world, though, at least at the time of its inception, without knowing what kind of hyperloop it should build precisely, where, how, and with what business model. To tackle this fundamental challenge, HTT decided to draw on thousands of voluntary, self-organizing contributors. This bottom-up approach leads to volunteers, in essence, exploring parallel, possibly conflicting, and potentially even redundant technological trajectories to collect a series of plausible ideas for aspects of hyperloop implementations that may be technologically feasible and economically viable. That means that HTT does not bet in advance on only pre-specified component technologies, system designs, or routes derived from highly speculative projections on how an eventually deployed hyperloop may change current transportation systems. Rather, HTT can continuously catalyze the information produced by its volunteers to see how various component technologies are maturing, what technological implementations they may sustain, and how those correspond to emerging market opportunities—until eventually striking when a promising configuration of technology, capabilities, and market crystalizes.

As such, I suggest that by also relaxing the boundary of a flat organization, firms may identify new ways of tackling settings of fundamental uncertainty, in which a clear way of proceeding does not exist initially, as in the case of HTT. In line with Reitzig’s points on how increasing the amount and complexity of information gathering posits new challenges for assessing and enacting said information, I expect that such flat organizations will find ways to delay the need to commit to interpret information through a specific and deterministic lens (and decide on a specific course of action with that) until their environment helps them identify a particularly promising one (e.g., Burgelman 2002; Dattée et al. 2018; Mintzberg 1979; Sull 2007; Wiltbank et al. 2006).

In turn, to design such organization, one of Reitzig’s many insights points toward yet another great opportunity for research on the future of flat organizations: if technologies such as artificial intelligence (Iansiti and Lakhani 2020; Puranam 2021; Raj and Seamans 2019) or blockchain (Hsieh et al. 2018) may increasingly help solve the coordination challenges inherent in going flatter (see also Altman et al. 2014; Bloom et al. 2014), we may even begin to see fully decentralized organizations that are not bound by the limits to scaling up that Reitzig finds in his book (and which are also reminiscent of stories of holacracy-following organizations, such as Zappos). Such technology-enabled organization might not stop at ‘just’ being flat. Possibly, we may also see a rise in fully autonomous individuals (not employed by any firm) collaborating spontaneously via platforms or mediated by markets to solve even the most challenging organizational problems (e.g., Boudreau et al. 2016; Boudreau et al. 2011; Boudreau and Lakhani 2015; Riedl and Seidel 2018). Eventually, such developments may possibly even culminate in a shift in what organizations of the future are, when we need them, and for what. While, again, this call is certainly not novel, the organizational innovation Reitzig lays out, combined with the technological innovations I allude to herein, may point to some of the puzzle pieces that had previously been missing to describe and theorize some of the promising future organizations to come.