On hierarchical agility and fractal stability

By Siegfried Lautenbacher Feb 23, 2016
On hierarchical agility and fractal stability

Translation of the Original Article by Siegfried Lautenbacher 

Fractal images are often beautiful to look at. In a fractal (fractus = Latin for "broken", " from frangere = (“break") in pieces is a "self-similar structure." Fractals are not just for complexity, but also for the formation of structures in nature. Many agile instruments in companies also operate on the principle of "self-similarity". This makes them capable of offering structures and stability in an otherwise very dynamic environment. A feature that distinguishes them as the backbone of future organizations.

Picture: tommietheturtle - Fractal Art Series 1 on Flickr (here). Under the terms of the Creative Commons (0), that is as public domain file.


Fractals appear to be well suited to bring you the topic of this blog a little closer. As described on the image, in companies many agile instruments such as Scrum function according to the principle of "self-similarity". This makes them capable to offer structures and stability in an otherwise very dynamic environment. A study by McKinsey December 2015 (here) wants to show that this is no conflict in not only the arts, but also the prerequisite for companies to remain "healthy" in the long term, and thus survive. What does this mean? The way this study processed this subject, is my view not a sustainable Undertaking. How far can we take the analogy with the fractal structure formation? I will use the example of the debate about the new blueprint of organizations. But, let's start with the first point of the study and their statements.

Agility rhymes with stability

"To master this paradox, companies must design structures, governance arrangements, and processes with a relatively unchanging set of core elements - a fixed backbone. At the same time, they must also create looser, more dynamic elements that can be adapted quickly to new challenges and opportunities" (McKinsey 2015, S. 1).


The study of Wouter, de Smet and Weerda from last December (2015), hereafter referred to as simply McKinsey study, is extremely interesting for our context. It shows that the understanding of the core of a Social Business has not really arrived everywhere. The authors postulate already at the beginning of the study that many companies would accept a false trade-off between flexibility and stability. Because of this (wrong in the eyes of the authors) assumption they then remained on the "safe" side of the stability to utilize economies of scale in the best possible way - at the expense of flexibility. On the other hand, there is much management literature on agility and flexibility, which in turn unilaterally neglects the issue of stability in the eyes of the authors. At this point, the study lists self-organizing teams within the context of  Holocracy models. In their own words, the study is about bringing both in balance, because for them only companies with both, namely "speed" and "stability" (p 2), have a 70% chance, to be listed in the top quartile of McKinsey index for organizational health. If you disregard from the outset that speed is not the same as flexibility and agility. In order to support that "stability" and "dynamic" (pay attention to the repeated displacement of terms) account for the core organizational agility, the study tried a nice analogy: A smartphone. This comprises of - again in the eyes of the authors - a largely stable hardware and a fixed operating system (company management and structures) as the backbone, as well as extremely flexible Apps (methods, processes) as "Instruments of agility". They have not chosen this analogy for nothing.


The Power of and!


It is completely clear for the authors of the study that both has its justification. It is about corporate management and agile project management as well as corporate structures and flexible process design. As well as balancing along each of three levels of organization:

  1. Along the structure level at which the resources are distributed,
  2. in the area of the management level, within which decisions are made and
  3. at the level of the processes that are critical for the concrete implementation of the decisions taken!

They try to prove this positively and negatively with examples known to them of actually existing companies. Although the result in this is a "and" as “as well as", but not in a really convincing sense. Why? Because on the one hand they do not really meet the character of agile processes and organizational forms, and on the other hand that the hierarchical decision-making structures that are mainly affected thereof are not called into question. Therefore constantly move the reconnection to (classical) Structural Constellations (hierarchy) of companies in the foreground, even if they explicitly reject a mechanistic conception (p.6).


Scrum is considered the agile method par excellence. To guarantee the necessary stability for agility, the method is exactly regulated in detail. The ScrumMasters have to pay particular attention to the timings and compliance but also to eliminate emerging obstacles. The structural formation and stability comes through the continual iteration of Scrum cycles, rather than via the hierarchical organization. Picture: Oliver Tacke - Scrum Cycle on Flickr (here) under the terms of the Creative Commons (BY - attribution).


This can be best seen in the section on the processes and their transformation (P. 10 - 12), which is why I want to outline this briefly now. For the study's authors mainly the key processes need to represent a stable backbone for the processes. These are to be designed accordingly – by being standardized. "In other words, everyone must speak the same standardized language" (as above pg. 10). If this type of standardization is missing, then the agility is suffering according to the authors. Consequently, there are two new roles in terms of jobs: A “Business-Process-Owner” and an “Integrator”, who is supposed to lead and integrate the processes of cross-functional teams (pg.11). If that is for no further expansion of hierarchy. In another example, processes are controlled based on extended KPIs (Key-Performance-Indicators). What is striking: Teams as a designer of processes, and communities as the fundamental structures are not found in this paper. This is in complete contrast to how agile companies actually work, even in the eyes of the authors. "Agile companies regularly rethink and, if necessary, redesign their structures, governance mechanisms, and processes to strike a balance between speed and stability" (McKinsey 2015, pg. 12). They do so, however, in the context of projects, teams and groups, not on the level of hierarchies!



Process and Structure?

It is actually very strange, imagining concretely the proposal from the McKinsey study - beyond the implementation of agile projects and interdisciplinary groups for new topics in the hierarchy. In this respect, the chosen analogy of smartphones actually fits quite well: Around the stable hierarchical core as backbone, for the purposes of the study there is a circle of agile project teams. But is this indeed what a (really) agile company is? Also with respect to the employees or the relationship to them, they have a peculiar understanding. When it comes to the question of Group Power anyway: "Employees spent too much time on internal discussions about best practices, methodologies, and process frameworks and not enough on actively improving their own ways of working" (as cited above, pg. 10). The last statement about the "norms of behavior" in organizations somehow does not fit in the picture, either. Most striking, however, is that the question of a collaborative cooperation in terms of social collaboration does not appear at any point of the study. For us an indication that the authors of the study have not really understood this aspect of flexibility and agility. This we would like to remedy, of course. For this, a short small trip to the question of structures and processes.

Looking at so many agile concepts such as Scrum, Holocracy and Wirearchy or even Working out loud (WoL), then this is a completely different understanding of the functioning of an organization. This also means relativization and adaptation of their structures. In principle applies here that of process of providing service is meant (and provided) relatively exclusively from the competent groups, teams or communities forth. The term Scrum, literally translated as crowding, for example, describes the situation that the team members continually and in close contact - but autonomous and independent of the hierarchy - vote on how they develop the project or product further. Next, they consult, also autonomously, with the customer. The same virtually applies to the concept of Holocracy, only transferred to the entire organization. Explicitly the decision-making powers and the authority are here placed with the self-organizing teams (the Holons). Even Working out Loud is about that employees connect in circles regardless of the hierarchy and learn from each other in a self-determined manner, thereby improving the work and to achieve greater efficiency overall.


For most businesses, the search for a new and agile blueprint has only just begun. However, it becomes clear that the processes are the new backbone of organizations. And likely replace the hierarchy, but at least significantly reduce it.

Picture: Tobias Kämpf - Presentation image of the lecture of the scientific results within the Wing Conference "The digital working world of tomorrow needs the people". Used with permission.


The procedures - different not only in detail - of how these concepts work (and are partially. described in great detail) replace what the McKinsey study calls (and presupposes) a hierarchical structure. One might say, the structure of the agile process provides the stability in these concepts. In the words of empirical research: These "processes create structure and become the backbone of the organization" (Tobias Kämpf 2015 here). In this respect, all concepts attach great importance to the fact that they are adhered to in the correct sequence and implemented iteratingly (repetitively). It is very conspicuous that this other type, to see structures and stability through agility and processes, does not occur in the McKinsey study. Probably not without reason. Only here is where agility and stability would actually rhyme in terms of a content identification - and not only literally.




Total fractal through collaboration


"Instead of needing permission to act or innovate, Holacracy gives blanket authority to take any action needed to perform the work of the roles, unless it is restricted via policies in governance or it involves spending some assets of the organization" (Wikipedia: Holacracy).

In agile companies, it is not about restructuring for restructuring’s sake. Instead, a need for change of the structures usually results from the fact that changing conditions should be responded to swiftly. Different iterated procedures to the action are proposed in order to still obtain stability and reliability in the individual processes. And at last here we come back to our Fractal: Agile methods are themselves related processes that are spelled out throughout an organization. For example, the concept of Holocracy on the macro level of the organization believes that self-organized circles ensure the necessary stability at all levels. In these circles, the roles are defined at a lower level, which can - and must - then each be filled in by different people. Here, the circles also determine independently their respective management function and decision-making processes.

Quite different, but certainly also in self-similar structures, does Scrum work at the micro level of the teams and projects. There are three central roles: The ScrumMaster (Process Manager), the Product Owner (technical expert) and the Team (the actual developers). A fourth role are the stakeholders (observer and consultant). Scrum is at least at two levels self-similar, and therefore fractal:

  1. On the level of rules and procedures, it can be described as small organization within the organization that is very oriented to the project schedule. The respective team designs the corresponding key process autonomously.
  2. On the level of iteration cycles (Sprints), a new Sprint starts after 30 days. This one has the same duration as the previous one, but the content has changed.  During the Sprint, there is a self-similar daily scrum meeting.  

Agile methods - according to Niklas Luhmann – approach the conclusion that one can adequately cope with the complexity of the environment of an undertaking by a similar complexity internally. A quasi self-similarity of the procedural complexity.

Particularly worth mentioning in our view is that we believe that the success of agile methods is another self-similarity, even more important to us: That people encounter each other again on equal footing in the circles, groups, teams and communities. Employees who are themselves self-similar as people, because they assume roles, although within the different concepts, but in addition will always have the freedom to actively connect and perform those actions that are necessary to reach the goal. With an enormous expandability through digital platforms. If you like, you will be totally fractal through a (social) collaboration. At least in a metaphorical sense.


Wouter Aghina, Aaron De Smet and Kirsten Weerda (2015): Agility. It rhymes with stability. In: McKinsey Quarterly, Dec. 2015. Available at: http://www.mckinsey.com/insights/organization/agility_it_rhymes_with_stability