Organizational Structures, an explorer with five ways in, covers 41 forms in nine families: how a whole organization groups people, places authority, layers itself, and coordinates work across units.
The coordinating mechanism each family relies on
One panel for each of Mintzberg's six coordinating mechanisms; each bar counts the forms in a family whose profile names that mechanism as prime. A two-part set counts for both mechanisms, and a contested cell counts by its first view.
M, mutual adjustment
D, direct supervision
W, standardization of work processes
O, standardization of outputs
S, standardization of skills
N, standardization of norms
Show the numbers
| Family | M, mutual adjustment | D, direct supervision | W, standardization of work processes | O, standardization of outputs | S, standardization of skills | N, standardization of norms |
|---|---|---|---|---|---|---|
| Classical | 0 | 4 | 3 | 0 | 0 | 0 |
| Divisional | 0 | 0 | 1 | 5 | 0 | 0 |
| Lateral | 4 | 0 | 1 | 3 | 0 | 1 |
| Knowledge | 3 | 1 | 0 | 0 | 2 | 1 |
| Network | 3 | 0 | 0 | 4 | 0 | 2 |
| Team | 5 | 0 | 0 | 0 | 0 | 1 |
| Ideological | 0 | 0 | 0 | 0 | 0 | 1 |
| Other traditions | 3 | 2 | 1 | 0 | 0 | 2 |
| Contemporary | 3 | 0 | 3 | 1 | 0 | 1 |
Mutual adjustment is the prime mechanism of more forms than any other, 21, spread across six families. Every divisionalized form names standardization of outputs, and only two forms, both in the knowledge family, name standardization of skills.
Source: Mintzberg 1979; Mintzberg 1989.
- Set the situation. Set up to twelve contingency factors. Each factor reads Fits, Partial, Strain, or Contested, with reason and source. Twelve undisputed misfit rules (seven from Burton and Obel 2004, the most developed scholarly fit model) sink a form whose whole profile meets one; partly met or disputed rules flag it.
- Locate your own structure. Forms nearest yours, by count of matching levels on ten dimensions. Most profile levels are our judgment, and a count is a resemblance, not a verdict: configuration theory holds that a structure's elements reinforce one another. Donaldson (2001) replies that fits form a continuum, most organizations at intermediate points.
- Diagnose by symptom. Goold and Campbell's nine tests (2002), a practitioner checklist, and 36 symptoms, each linked where possible to a misfit it may signal, the forms that relieve it, and its pattern in Coordination Failures; most links rest on theory, a third on our judgment. Fifteen have disputed remedies.
- Browse by family.
- Compare two forms.
The ranking's counts use the selection form of fit (Drazin and Van de Ven 1985), matching structures to situations; they leave aside the performance a pairing adds (interaction fit) and distance from an ideal profile (systems fit). Unweighted, forms rank by count of Fits, then fewest Strains; forms still level show as a tie, alphabetically. A tie carries a note on equifinality (Gresov and Drazin 1997), that more than one design can fit a situation; a tied count alone shows no equivalence. Most cells are labeled judgments, our reading with its premise named, and a contested cell shows every view. Simulations dissent: the best design depends on interactions among choices, which one-factor counts miss (Rivkin and Siggelkow 2003).
Under every ranking, a panel gives four schools' readings: in strategic choice (Child 1972), those with power choose structure, and context constrains without determining it; in institutional theory, structures are adopted for legitimacy (Meyer and Rowan 1977) and spread by isomorphism (DiMaggio and Powell 1983); in population ecology (Hannan and Freeman 1977, 1984), structures resist change and selection does the sorting; and in configuration theory (Miller 1986; Meyer, Tsui, and Hinings 1993), structures cluster into a few coherent types. Donaldson (1995, 2001) defends contingency theory against them, a defense itself disputed.