There’s been quite a lively debate at Business Week, where two contributors – and a long list of commenters – indulged in some weighty mutual executive briefcasing (handbagging just didn’t sound right) in response to the question: “Multi-dimensional organisational design (Matrix) is the best way to restructure a business. Pro or con?”

In the Pro corner, Jay Galbraith argues for the value, inherent merit and – in today’s trading environment – the inevitability of the victory of a collaborative approach over a command and control variety. In the Con corner, Guido Quelle sees matrix organisations as painfully slow, lacking clarity and clear lines of responsibility. Verbal bruisings have been administered and received on both sides but there’s been no knock-out punch: anyone hoping to see the late, grand old man, Peter Drucker holding the limp wrist of one argument aloft and counting to ten would be disappointed.

Reading between the contributors lines, it is also possible to glimpse some pretty trenchant opinions. Where these opinions to be transcribed, the gaps between their own lines might reveal something else: people whose personal working preferences are deeply engrained, and opposed to working models that lie outside them. While there’s virtue in being even-handed, it’s not unfair to say that the ‘traditionalists’ exhibit this trait more strongly, but is it too simplistic to point out that opposition to new methods is what makes you a traditionalist in the first place. There are other points here that are not unrelated, more of which later.

To mangle the metaphor further, there’s a cause and effect element. Life, the world, and the operating environment move on, and if the type of golden egg changes, doesn’t that imply it might be time to make a few tweaks to the chicken? Without adaptability and a keen ear and eye on the customer, aren’t we looking at a love of resolution that could lead to a resolute, clearly defined obsolence (the ‘Con’ argument) or an embrace of complexity that ties itself in knots while the opportunities drive past in the fast lane (the ‘Pro’ argument).

The big ‘buts’ that I saw did, however, occur to one or two those commenting:

  • Your customer doesn’t care how you’re structured: they care about product and service (interestingly, the biggest flaw in web design strategies that was pointed out for several years by usability guru, Jakob Nielsen, till more companies realised that the web surfer controls the mouse: the web is not a presentation, and the audience have to want to listen)
  • ‘One size fits all’ isn’t a mantra for organisational design: even within the matrix outlook, there are a wide variety of matrix models, some more suitable for some business than others. Proctor & Gamble’s matrix model isn’t the same as IBM, as Proctor & Gamble isn’t IBM: its product line, markets and culture are all significantly different
  • Even the best idea doesn’t work if it’s implemented poorly, or parts of the organisation are working against it: companies fail for many, many reasons – including those following organisational models that are well and truly ‘tried and tested’. The unfortunate thing about trials and tests is that failure is an option. If everyone always passes the test, it’s not a test.

There’s a larger point, which is as much about organisational change as it is about organisational design. The matrix approach is the new (or not so new, depending on your definition of matrix, be it hard or soft, a hybrid, or a project based organisation) design contender – the thing to which not just organisations but their people must adapt. And change without resistance is like an English summer without rain: delightful to contemplate, but unlikely to arrive. Although much is made of the requirement of employees to adapt to a situation in which they report to multiple managers, the impact of the matrix of managers probably gets less attention.

The role of providing focus remains a management responsibility, but managers must learn to work collaboratively and to arrive at intelligent compromises every bit as much as the teams working below them in the matrix. The comfort and satisfaction of establishing clear responsibilities, processes and rules is the other side of the coin from the tunnel vision of the silo supervisor: complexity is being managed by minimalising opportunities for it to arise, rather than by accepting that more than viewpoint, perspective or imperative exists and arriving at a balanced view. Those who enjoy a sense of control can feel not just undermined – no empire builder likes to share their throne – but challenged by the need to work in teams of their own with other managers with a very different focus from their own. In reality, authority and responsibility will still exist within the organisation, but they are now shared rather than owned.

Matrix organisations depend on relationships – not in the sense of comparative status but in the sense of interaction: people must define themselves not in relation to others but in relationship with others. The matrix requires that people give each other what the recipients need, and operate as much through dialogue as through decree. The exception to that last point comes where a multiplicity of inputs resists resolution, but that that requires is not management so much as leadership – a direction, in pursuit of which the matrix can organise its energies, rather than a dictat.

It’s a point that’s been picked up beyond the Business Week debate. Here’s John Shook, from an article called – appropriately – We’re All Connected And Nobody Is In Charge:

The fact of organizations being cross-functional in operation while being functional in structure results in a matrix which typically leaves ownership unclear, decision-making stymied, and everyone frustrated. Leadership and the management system need to facilitate a shift from debate about who owns what (authority) to a dialogue around what is the right thing to do.”

That’s why Toyota managers would avoid relying on their authority to instruct others, striving whenever possible to lead by influence and example rather than simple command. As one of my bosses in Japan told me, “Avoid telling your staff exactly what to do. Whenever you do that, you take responsibility away from them.”

He also quotes the US Marine Leadership manual: “An individual’s responsibility for leadership is not dependent on authority.”

The ‘Con’ argument has its corner fought more forcefully – although the background image of flames and the fiery rhethoric may me suspect ‘aggressively’ – by Flavio Kliger in his blog post “Come-on! The Matrix Organization Era Is Gone!”. Although the first of his 11 reasons why clears make the point that he sees agreement as a bad thing, his points deserve to be addressed. I suspect we’d have a (suitably heated) argument about innovation in particular. Here’s his tenth point:

It kills innovation as innovation is the main activity overlapping all dimensions. In the endless negotiation of budgets everyone prefer to assume innovation is someone else responsibility then no one ends-up funding long-term projects in expense of the short-term silo’s scorecards.”

It’s not just his grammar and syntax I’d quibble with here. His other points leave the impression that scorecards are essential, that performance appraisal is individual and that collaboration is unnatural. (I’m paraphrasing, but I don’t think I’m totally misrepresenting.) But as we’ve said before, there’s no real reason that performance appraisal can’t include appraising the degree to which the individual enables, supports, encourages and develops others: no-one, even in a silo, performs in total isolation and their impact on others (and others’ work) should be – and usually isn’t – appraised. If reward and recognition is linked to performance appraisal – and what else should it be meaningfully linked to – failing to appraise collaboration means failing to reward it. Failing to see it should then surely not come as a surprise. What Flavio seems to be displaying under his sulphur and brimstone tone is at least partly a line in unacknowledged legacy thinking. It’s easy to appraise on the basis of what’s easily measure, or to measure only the things we want to see. Bigger pictures can be less comfortable. (As Jay Galbraith points out in his ‘Pro’ argument: implementing a matrix saw a large turnover of senior staff in several success cases. Now that’s an interesting chicken and egg argument.)

Back to the innovation point: yes, innovation requires funding, but the funding is the egg, not the chicken. The ideas have to come from somewhere before they can be resisted (a point that you’d think wouldn’t be lost on him …). As the story MIT’s Building 20, Microsoft’s Building 99 and the recent work of write Steven Johnson (read our review of Where Good Ideas Come From) show, innovation comes from multi-disciplinary fluid teams rather than from top-down models. Just as leadership is a behaviour rather than a status, so ideas can start at any level. In organisational terms, a matrix might actually be very good for innovation. (In terms of social culture, it might not – as Jaron Lanier argued in You Are Not A Gadget – but OD wasn’t Lanier’s topic or focus.)

Unless, of course, Flavio is talking about organisational change, where the struggle to arrive at decisions where so many players need to reach resolution can have a sclerotic effect. Another blogger, Jeremy Phillips, acknowledged this problem in relation to another matrix organisation, GM, in his post: Fighting the Matrix, before concluding:

It’s as if we need one organizational structure for run of the mill, day to day activities and a completely different org structure and decision making capacity for implementing change.”

But it strikes me that just takes us back to leadership and culture (implementing change is about leading it, not managing it), and to a related element of the matrix model – matrices are supposed to be fluid and mutable. Implementing a matrix by managing it (“the matrix is in place, now will everyone get on with it”) rather than leading it (“let’s put our heads together on this one so that we can address ‘x’”) is to work against its own grain.

The argument will, of course, run and run. But I’d refer people to a book we mentioned a while back while arguing against the use of lazy sporting metaphors in business writing (even if doing is an act of wanton hypocrisy). Have a read of Brilliant Orange: the Neurotic Genius of Dutch Football – why the much-praised Dutch football team of the 1970s charmed the world with their dazzling skills and their totaalvoetbal model – the matrix with 22 feet and a ball, essentially – but kept an empty trophy cupboard. It’s a surprisingly rich and fascinating read, but I didn’t get to the end of it thinking that it was the team or the style of play that needed changing.

Add to: Facebook | Digg | | Stumbleupon | Reddit | Blinklist | Twitter | Technorati | Furl | Newsvine

Add to FacebookAdd to NewsvineAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to Furl