In May this year, I began a survey on Agile and Culture, covering the three big methods: Scrum, XP, and Lean-Kanban. I solicited participation on several major lists (scrumdevelopment, leandevelopment, and extremeprogramming) and from a couple of organizational clients. Approximately 120 people responded.
The results were confirming on the one hand, surprising on the other. As a cultural meme, Agile is fundamentally patterned on the Collaboration culture type (not surprising). A strong second preference, however, is the Cultivation culture type (surprising). On the overall level, there are only slight differences culturally between the 3 Agile methods studied (a bit surprising). However, on the level of specific culture levers (things like Power & Influence, Decision-making, etc.) the results were either different from the overall pattern (e.g., Collaboration followed closely by Cultivation) or showed a different pattern between the 3 methods.
Fundamentally, organization culture is to organizations what personality is to people. Culture combines many things: work practices, values, how processes and other systems are carried out, styles of leadership, decision making and thinking about organizational challenges and solutions. I have been using a specific culture typology for working with Agile transformation efforts for many years, and that approach is the basis of this research.
In The Reengineering Alternative: A Plan to Make Your Current Culture Work (1994), William Schneider outlines his researched-based model of organizational culture (Bill is a great guy, and one of our partners–see our Partner page). Schneider outlines four core cultures. Similar to a person taking the Myers-Briggs type inventory, there is no “right” answer or “better” culture. Any culture (like a personality) can be adaptive to its environment or not, in balance or not, and authentic or not. Determining and aligning a management approach (such as Agile) that is being implemented with the existing organizational culture is the difference between flourishing success or abject failure. Not all ideas are good ones, depending on their fit with the organization’s culture. Here’s a brief description of the four cultures.
If a culture could be said to have a quest, in the Control culture that quest is for certainty and predictability. Not surprisingly, the Control culture loves data and objective analysis. It strives for market share dominance with customers and to be the ‘only game in town.’ Managers in a Control culture tend to be directive and authoritative. Jobs are focused on functional need, even functional loyalty. The archetype of the Control culture is the military, where a strict chain of command is followed and rank means everything. The climate in such a culture is serious, formal, at times even secretive. The underlying psychological motive here is power. A potential misunderstanding is that a Control culture is inherently ‘controlling.’ The urge of the culture is for certainty—the kind of certainty needed in a nuclear power plant, for instance—which is not necessarily controlling, but rather orderly and procedural. When the culture is overly controlling, that represents an out of balance situation.
The quest of the Competence culture is for freedom, distinction and uniqueness. A consistent product strategy in such a culture is striving to be the best, innovative, one of a kind, cutting edge. In contrast to a Control culture, the role of employees here is to become an expert within one’s specialty. The culture is oriented towards learning and development in service of becoming the best. The climate is intense and competitive, with a tendency towards being Spartan and prideful. Power comes not through position per se, but through prowess in one’s field, a meritocracy. Organization structure tends towards the matrix or an adhocracy. The underling archetype is the traditional University, where people pursue being the best. Achievement is the driving motivation in the Competence culture. Many engineering organizations and specialty product companies are Competence cultures, as are many IT organizations.
The quest in the Collaboration culture is for unity and connectedness. The relationship with customers is synergistic and oriented towards partnering. The natural organizational form that goes along with this intent is the cluster, often a cluster of teams. Leadership in the Collaborative culture is participative and collegial, focusing on team building and developing trust. Employees are encouraged to be generalists, to honor diversity and utilize others as resources. There is an atmosphere of informality, of “let’s try it and see what happens,” of on the job training and learning. The climate is harmonious, trusting, spontaneous and egalitarian with a ‘can do’ philosophy. (Parallels to the Agile philosophy are perhaps obvious.) A Collaboration culture is motivated primarily by affiliation. The archetype is the family or sports team. Collaboration is the favored culture of many consulting companies and other highly collaborative service providers.
The final of the four core cultures is called Cultivation. Its quest is for meaning, for making a contribution. The relationship to its customer (or constituent) is their growth, the realization of their highest potential. Leaders in the Cultivation culture are catalysts, cultivators and stewards of human potential. The role of employees can vary from functionalist to generalist to specialist, depending on organizational need and personal inclination. Mentoring, sponsoring and a fervor to learn and grow are common. The climate of such an organization is lively, magnetic, committed, emotional and giving. The organizational structure is unconventional such as a wheel or lattice. Cultivation is the ultimate ‘values-driven’ organization. It is the least common type in the for-profit world, but quite prevalent in non-profits and religious and spiritual organizations, which provide the underlying archetype. Self-actualization is the primary motivator in a Cultivation culture.
The four core cultures are generally depicted by Schneider on a 2×2 matrix, where the horizontal axis represents the Personal cultures on the left and Impersonal ones on the right. Likewise, the vertical axis represents an Actuality culture on top, a Possibility one on the bottom. Collaboration and Cultivation are Personal cultures, Control and Competence are Impersonal ones, etc. The matrix also represents the fact that Control and Cultivation are opposite culture types, as are Competence and Collaboration.
The following diagram represents results from the Agile culture survey, combined across Scrum, XP and Lean-Kanban:
The digram shows Collaboration to be the strongest culture preference (47%) for the ideal Agile team, as judged by the 120+ respondents to the survey. Cultivation is a strong second at 41%. Competence shows up a distant 3rd at 9%, while Control (predictably) is a meager 3%. What this confirms is that Agile is clearly a strong culture meme (it is not, for instance, spread somewhat evenly across the four types) and it is decidedly a Personal culture. Further, if you are implementing Agile into a Competence or (especially) a Control culture, beware. (There are ways to mitigate this risk, but that is beyond the scope of this blog).
A further detailing of the results is revealed by examining each of the 10 culture levers measured in the survey (overall, Schneider identified 20; I choose the most salient 10 for this research).
The following diagram shows results for each of the 10 culture levers, again summarized across all 3 methods:
At this greater level of detail, some new patterns emerge. First, in only four of the ten levers does Collaboration have the strongest preference, while the other six have Cultivation as a preference. In general, Collaboration and Cultivation are number one and two. In three instances, however, the Competence culture is the second strongest. These three levers are Approach with Customers, Power & Influence, and Key Norms.
Two cautions: the results d not represent the study of actual Agile teams, but rather the ‘ideal’ preference for a good Agile team as expressed by practitioners. Second, when Schneider measures an organizations culture, he does it with a much more extensive (and statistically validated) instrument. These results may be incomplete due to this limitation of the survey.
I hope to publish a full report on this work later in the year. I will also be providing further detail on my analysis during my Agile 2010 tutorial, Blueprint for an Agile Enterprise. Hope you will stop by and say hello!