Thursday, September 24, 2009

What's Missing from VPEC-T?

#vpect is a deceptively simple systems thinking lens developed by a couple of friends of mine (Carl Bate and Nigel Green), and described in their book Lost in Translation. The letters stand for Values, Policies, Events, Content and Trust. For a good brief description, see VPEC-T the 5D lens by @RoyGrubb.

There have been several discussions in the VPEC-T Google Group identifying possible additions to the VPEC-T lens. Some months ago, partly in order to find out what kind of lens VPEC-T actually was, but partly because I actually think it's important, I suggested adding the letter M for Meaning, and now Roy Grubb has just started a thread suggesting the letter O for Outcomes. Previously, Sally Bean asked about Roles/People/Community. (Have I missed any other suggestions?)

One response to this has been to claim that any of these are somehow already implicit in the original five letters. I don't actually think this is a very good response. I might as well argue that Trust is implicit in the Kipling-Zachman lens (in other words, implicit in the use of the lens by those practitioners I regard as using it "properly") and nobody could ever prove me wrong. The point of calling it VPEC-T is to draw attention to these five concepts in particular (or six if we count the dash as a concept).

The next response is to challenge the notion of completeness. Obviously there is an indefinite number of concepts we might want to pay attention to. There are many other lenses (Kipling-Zachman, CATWOE, SWOT, and so on), and each of these focuses attention on a different set of concepts. I don't think it makes sense to try and produce a single all-purpose lens that covers everything, and I don't see why adding lots of other stuff to VPEC-T would be an improvement. We can agree with Sally that Roles/People/Community are important without necessarily concluding that these concepts must somehow be bundled into VPEC-T. A better solution might be to select another lens that provides particular insight into Roles/People/Community, and use this lens in combination with VPEC-T.

However, I think it still makes sense to talk about ways to improve the VPEC-T lens, provided this doesn't just mean making it more complicated and more likely to lure people into a false sense of completeness. One of the reasons I like VPEC-T is that it draws our attention to the complexity associated with each of the five letters. Many traditional methodologies have a place for a simple unified statement of value (goal or objective); but VPEC-T tells us about many interacting and often conflicting value systems (V), many interacting and inconsistent policies (P), and so on. One reason I wanted to add M for Meaning (and the same argument might also apply to Sally's or Roy's suggestion) was that I see Meaning as the same kind of concept - we are faced not with a single unified meaning emerging from the analysis (which I think was Nigel Green's argument) but with a multiplicity of meanings - each event or policy can mean something different to each Role/Person/Community.

To take the Dinner Party example from the Lost in Translation book, each of the things that happens at a dinner party can be interpreted and experienced in different ways. Policy dictates whether the guest should offer to help with the washing up, but Meaning tells you whether people actually feel grateful or threatened when this occurs.

I also think Meaning is a key element of change management. We frame change proposals so that they can mean something positive for each of the stakeholders. (I hate all this talk about "overcoming resistance", but that's another discussion.) And we create Meaning by telling stories. So one way or another, I am determined to have a lens that helps me think about Meaning.

As far as I can see, I have three possible options. My first option is to use a completely different lens for understanding Meaning. My second option is to create my own idiosyncratic version of VPEC-T, which I might call VPEC-TM. (Meanwhile, Roy creates his own version, and the VPEC-T community fragments before it has even really formed.) My third option is to persuade the VPEC-T community to make a collective change to VPEC-T.

This is of course raising a question about the VPEC-T community. Now Sally, what lens do you think we should use to think about this question of Role/People/Community?

1 comment:

  1. Richard,
    Here are some candidate lenses. I'm thinking more here of domain analysis/general problem solving that may well be broader in scope than information system analysis.

    Firstly, you've already alluded to CATWOE/SSM, which identifies customer (beneficiary), actor and owner roles. This is enriched in the 'developed form' of the SSM, described in Checkland and Scholes book of 1990, which includes something called Analysis 1/2/3:
    Analysis 1 is the analysis of an intervention itself - who caused the intervention to happen, who is conducting the intervention and who is affected by the outcome of the intervention.
    Analysis 2 covers Roles, Norms and Values, which overlaps with VPEC-T. (Interestingly roles could include things like 'trouble-maker' and 'joker' as well as functional roles)
    Analysis 3 covers Power relationships,

    Secondly you could draw Rich Pictures or Influence diagrams
    See here for examples.

    Thirdly, there is Verna Allee's Value Network Analysis, which I've not actually used myself, but am interested in learning more about.

    When I've discussed roles with Nigel Green, he assures me that VPEC-T analysis does cover roles adequately, but they are not a primary feature. I can also see that there is a danger that you may perpetuate current organisational structures that might be better redesigned, if you take them as a starting point of your analysis. However, I usually like to start an investigation by looking at the stakeholders involved, if only to understand the politics and make sure that the right people are consulted and communicated with.