How Change Happens + 3 months: how’s it going?

February 9, 2017

A philanthropist using systems thinking to build peace

February 9, 2017

What determines whether/how an organization can learn? Interesting discussion at DFID.

February 9, 2017
empty image
empty image

I was invited along to DFID last week for a discussion on how organizations learn. There was an impressive turnout

Simple, right?

Simple, right?

of senior civil serpents – the issue has clearly got their attention. Which is great because I came away with the impression that they (and Oxfam for that matter) have a long way to go to really become a ‘learning organization’.

So please make allowances in what follows for all the warm, cuddly areas of mutual agreement – I’m going to focus on the areas of disagreement, which are inevitably the most thought-provoking.

To mean anything, learning requires a change both in ideas and behaviours. So what were the theories of change that underpinned the approaches to learning in the room? I found it hard to pin down exactly – they seemed mostly tacit – but a lot of what I heard reminded me of the discussion at Twaweza a couple of years ago. For many present, the tacit theory of change seems to be ‘knowledge → learning → changed behaviours → changed outcomes’. Yeah right.

What we realized at Twaweza was that ‘it’s all in the arrows’. You need to unpack the assumptions and think about what needs to be in place for that theory of change to have any chance of resembling what happens in practice.

And the neat linear ‘we all want to learn from the evidence’ assumptions certainly didn’t fit with my own experience of working at DFID (e.g. phoning up ODI to say ‘we need evidence of the harmful effects of EU cotton subsidies for the meeting in 3 weeks time’ – classic policy-based evidence making). Or the senior Treasury official who stomped down Whitehall to inform us that we should never question the ‘eternal truths’ about trade, namely that ‘trade liberalization leads to more trade, and more trade leads to less poverty’. No amount of knowledge, evidence etc was going to change his mind on that one. Maybe that has all changed in the decade since I left, but I would need to be convinced (preferably by some evidence!)

For my contribution, I played around with a 2×2 – how big is the new idea/piece of learning v how well aligned (or not) is it with current thinking and practice. Some 2x2s prove very useful, others get shot down in flames by FP2P readers. Let’s see which way this one goes. (By the way, does anyone have an easy way of putting together 2x2s? Doing them in Word is a total pain).

learning organization 2x2
How does learning take place in the four quadrants in an institution like DFID (Oxfam isn’t much different, I suspect)?

Small idea, aligned with organization: this is the only quadrant that fits the implicit theory of change, provided there are mechanisms in place to disseminate the new ideas, an enabling environment (incentives, leadership) to allow people to learn and put them into practice, and people who are actually curious about the world and want to learn new stuff. And even those are pretty big ifs.

Small idea, contrary to organization: uphill work. People will mutter ‘that seems counterintuitive’, ‘it goes against my priors’ or a variety of other euphemisms for putting their fingers in their ears and singing la la la. Individuals will either give up, or seek out fellow believers and form small guerrilla networks of dissidents. Whether that network grows depends on the individuals involved, but also on a permissive environment from bosses. The good news is that as part of its increased commitment to ‘knowledge for development’, DFID seems committed to

Yoda's in the bottom right

Yoda’s in the bottom right

encouraging these kinds of ‘communities of practice’.

Big idea, aligned with organization: You’d think this would be straightforward, but it isn’t. For a start, people in different disciplines and professions will have contrasting views about how important this ‘big idea’ really is. Should it hoover up lots of management attention and resources, taking them away from their own areas, or is it some fringe hobbyhorse that should not get too much airtime? You get the picture.

Big idea, contrary to organization: Step forward Thomas Kuhn and paradigm shifts. There will be huge resistance to the big idea from those whose disciplines, careers or views of the world it threatens. There will need to be a high degree of ‘unlearning’ – battering the organization with evidence, narratives and credible voices to show that the current paradigm needs revision. Large failures and critical junctures (eg financial crises) can help dislodge fixed ideas. Then a massive exercise in experimentation, accumulating evidence and followers, probably followed by a final battle with the existing paradigm.

But maybe not in the way he means it

But maybe not in the way he means it

And as so often, the elephant in the room was power. Power (visible, hidden, invisible) determines what evidence is seen, listened to, gets traction or is dismissed. To unpack those arrows, you also need to understand the nature of power in a large institution like DFID (or Oxfam, for that matter).

It will be fascinating to see how this discussion plays out, and the stakes are high. Any organization that is serious about ‘doing development differently’, ‘adaptive management’ etc has to get those arrows in place.

I’d be interested in hearing from others at the meeting – what did I miss(represent)?


  1. Duncan, two quick points. First, in your illustration of against the grain, you seem to frame it in terms of resistance to a message, but the power aspects would probably point to the bigger issue (even with a small idea) being getting onto the right agendas. That is, nobody says “that goes against my priors,” they just don’t attend the meeting where topic X is discussed because they aren’t aware that it is relevant to them or because they are too busy. Seems to me that resistance is much more likely in this sort of impersonal way – folks don’t know what they are missing – than in an intentional one.

    Second, while you may not always see eye to eye with DFID, I’m not sure you want to call them senior civil serpents :-)

    1. When I worked for DFID I loved by called a civil serpent – made me feel much more wicked and interesting! However, if anyone from DFID is traumatized by it, please let me know……

  2. Sitting in Pakistan one of the serious drawbacks with organisations like DFID is how little a time do their staff get to understand the context and talk to those on ground.They are posted out before that happens. The absence of institutional memory means that history is repeated endlessly with old tested ideas being replayed with new tunes and the results no different than last time. If organisations dont have memories how do they learn.

    1. I agree with you totally on lack of institutional knowledge preventing learning from experience. A lot of this blog is about governance etc projects, but most of the arguments also apply to infrastructure and the implementation of infrastructure projects. I was fortunate enough to work as an engineer on a project in a remote village situation for 8 years. Sadly many engineers are trained to believe in their own abilities, and to believe that an adaptation to local situations only requires a minor tweek, and are only brought in for a week or two to undertake designs. I have seen innumerable infrastructure projects fail because the engineers (and donors) have repeated the same mistakes, because either the institutional knowledge was not there or they did not seek it out.

    2. I completely agree with Masood, until the donor agencies like DFID and USAID remove the glass ceiling of what level national staff can work to and until they stop bringing foreign donor staff in for only a year to two years with little chance to get to know the context and little ability or permission to leave enclaves and capital cities then little in this regard can change.
      Local donor staff have the knowledge, the relationships and the experience to embrace the learning needed to embed DDD and adaptive management to design, fund and guide truly local led and politically smart programmes. Unfortunately, I would argue that this shift would sit in your bottom right quadrant Duncan and would take some brave senior management in the donor agencies to recognise and make that change.

  3. Keeps coming round, this question, thanks for the blog report. But don’t the the clues to answers tend to be organisation-specific? Aren’t those all-important power relationships unique to a particular time, group of people and context? So might be more useful to ask the question, ‘what determines how *this* organisation learns?”. For a review of learning and knowledge management in WaterAid UK we tried Appreciative Inquiry, identifying projects and teams generally seen as’ successful’ and investigating why they worked well ( As ever, there was as much learning in the process as the case-study products. The generalisable themes were probably less significant than the way that combinations of people found a way to make things happen within the organisation. Sharing and celebrating that was important. There must be stories of change that illustrate your matrix from within Oxfam, and for that matter, DfID. It would be interesting to hear case-studies from within

  4. Agree with Masood and Mary, on the importance of institutional memory. Too much of the talk about organisational learning is focused on acquisition of new information and knowledge and too little on retention and use of old knowledge. Unfortunately, the past is simply not sexy. Let me know when you hear of a conference coming up on the subject of how we can make better use of our own past experiences in x area (i.e. not the latest / interesting new stuff). I will make a donation to Oxfam.

  5. I use Excel for creating 2x2s. The “frame” provided by the individual cells makes it malleable. Then you can remove the grid lines and drop it into a word document as an object.

  6. I dunno Duncan – what is this particular 2 x 2 actually saying? The forms of learning that dominate or the level of resistant that is likely? Giving the 2 x 2 a title might help populate the quadrants with comparable descriptions. Good not to forget basic organisational learning theory – single, double, triple learning which you simplified as small deal/big deal (a subjective interpretation by the way as what might be a big deal for you might be a small deal for me and vice versa)? In a 2000 publication Managing for Impact, Jim and I referred to two forms of learning as ‘status quo confirming’ or ‘status quo challenging’. Your against and with organisational grain? The latter is more difficult but the former is not unimportant. Power comes into that for sure. Which kinds of learning are most fostered, what is allowed to be challenged?

    Two anecdotes don’t prove an agency’s capacity…however, this would tally with some aspects of what we discussed at the Politics of Evidence at which you said we were exaggerating the problems, if I remember well! :)

Leave a comment

Translate »