What can would-be African lions learn from the Asian tigers? It’s all about how urban elites see farmers, according to ODI.

March 13, 2014

Research → Policy; understanding NGO failures and trying to be funny on inequality: conversations with students

March 13, 2014

Why scenario planning is a waste of time – focus on better understanding the past and present instead

March 13, 2014
empty image
empty image

Time past and time future

What might have been and what has been

Point to one end, which is always present.

TS Eliot, Burnt Norton

A few years ago, I used to rock up for the occasional UK government-convened scenario planning exercise (I know, exciting life or what?). They were usually run by ex Shell or BP ‘foresight people’ turned consultants and boy, were they disappointing. Asked to identify trends, participants regurgitated what they had read that week in The Economist or Financial Times. We then spent a few hours discussing and clustering them into possible futures, and inordinate time trying to come up with catchy names for the inevitable 2×2 (see below).

Why? Well partly, there is always huge attraction in thinking we are somehow able to predict the future (the source of the magnetic attraction over politicians by the more unscrupulous economists). Scenario types always stress that there workshops are not about prediction, just mental gymnastics to prepare us better for the future. But I haven’t seen much evidence of that.

I was reminded of this by a couple of conversations last week – a UNICEF foresights person dropped in to pick my brains, and I listened


Really, what is the point of this?

Really, what is the point of this?

to a Shell scenario-ista do apparently random thought association at a conference (those guys haven’t changed). Before we break out the flipcharts and weird hexagonal post-its (beloved of scenario planners) to grapple with time future, I think we need to get much better at time past and time present (using Elliot’s language – if you haven’t read the Four Quartets, you’re in for a treat).

Time past: our ability to capture how change really happens is pretty patchy. All too often, we succumb to the temptation to airbrush out all the interesting stuff: the accidents, unexpected twists and turns, the failures and how we respond to them. The pressure to produce shiny, smooth, positive evaluations for funders or bosses undermines our ability to truly learn from the past. I often say that in Oxfam we have two kinds of ‘time past’ narrative. A dumbed down ‘thank you Oxfam for my goat, now my children are in school’ fund raiser and the 200 page largely unintelligible evaluation. What we miss is the middle ground – ‘stories for grown ups’ that welcome messiness, doubts over attribution etc, and capture the enthralling complexity of change.

Time present: but our work on time past sometimes looks stellar compared to time present. Working in complex systems where change is intrinsically unpredictable and non-linear means above all, having fast feedback loops so that you notice when the system is changing, and respond to it. This is really hard for large organizations that try to maintain coherence and direction through a hierarchy of plans (organizational, departmental, team and individual). If, after spending months agreeing these plans, something changes in the context that suggests a new direction, it is far easier to ignore it than rip up the plan and start all over again.

Even if individuals in their personal lives are aware of big changes in the context (as we call real life in the development business), they don’t bring that into their work. While he was at ActionAid, my friend Matthew Lockwood used to say that all his interesting discussions about politics and context took place in evenings, usually in restaurants or bars, but never moved into the office. When food prices took off worldwide in 2008, I’m sure Oxfam staff around the world were noticing the impact in their shopping baskets, but the first we research wonks in Oxford knew of this major disruption in the world food system was when journalists called us asking for a comment. The feedback loops just weren’t in place.

crystal ballThe exception to this is humanitarian work, designed precisely to respond rapidly to shocks. When the Arab Spring took off in early 2011, for several months it was only our emergencies teams who discussed it as a potential refugee crisis. Only later did the penny drop with the rest of us that this was a massive advocacy opportunity, and we duly began putting Oxfam’s weight into advocacy partnerships at the Arab League and elsewhere.

Conclusion? If we want to pursue agility, flexibility and all those other words beloved of the managerial classes, the key is getting better at picking up the seismic tremors of social or economic change as early as possible, and learning to tell nuanced, intelligent, warts-and-all stories about change on the ground. Only then should we think about devoting a few hours on scenario planning. Otherwise, I won’t be bothering.


  1. Great post Duncan. One technique that I have seen work well is to project yourself to the end of a project, ideally with a group of people who also have an interest in it from different perspectives. Then you talk about why the project worked or didn’t work and why. It sounds gimmicky, but the advantage is it allows you to discuss openly issues that otherwise might be buried (e.g. people lose interest in implementation once the fancy concepts have been outlined), and – ideally – address them. Anyway – keep the posts coming!

    1. “One technique that I have seen work well is to project yourself to the end of a project” Nice, thanks will give that a pop

  2. We do the same thing in our advocacy work when pulling together strategies – ie; focus on predicting what will come up and how this will realise our model of change.

    It’s usually the unexpected events that open up opportunities. So why don’t we just say, here are 4 opportunities to make change happen that we know are there, we’ll try a bunch of interventions that probably won’t work, but we’ll focus on being ready to capitalise on any opportunities that arise unexpectedly. This should include the opportunities that arise as a result of our ‘likely to fail’ advocacy interventions that surprise us.

    I can’t remember the last time change happened without a surprising shift in context that threw planning out of whack. I agree, we can’t predict the future, so let’s focus on being agile and responsive when the unexpected happens.

  3. Great post – wish I could write so well. Two points – feedback loops are not always fast and they work at the micro-level – so changes at the more macro level may be fast or slow and still ’caused’ by many factors working together or against each other – but still worth a try. Second point, not formal scenarios maybe but scenarios do give a language for looking forwards, anticipating – some of future possibilities can be quite concrete (dams or fences being built, end of oil reserves) and even when possibilities are less concrete, foresighting can help alert us to what to look out for, have contingencies for. So I think: look past, look around (systemically), look forwards.

  4. Straw man argument Duncan; you show a very poor example of a scenario matrix. And it’s not all about the future – scenarios can help a body of people decide where they are now and where they have come from and what are some of the forces to be considered, if they are done properly.

    Scenarios are just one tool for thought – can be useful if you know how to use them – can be a useful stepping stone between narratives and models.

    Finally, are you suggesting that the IPCC scenarios were a waste of time?

  5. Great post, Duncan,

    1. Scenario thinking facilitates exploration of the “adjacent possible”, so don’t knock it all too much.
    2. We have to get out of the prediction business, and focus on “silent transformations” that happen under our very nose (when did you begin aging?)
    3.We need to learn how to change mentalities, the mindset that at the deepest level unconsciously drive our actions (UK colonial policy was based on the Eton mentality). How to do so? A new way is deep videogames(SimCity is an example), which teach to explore events, rather than explain them in a few words. Videogames convey much more than a linear narrative, and can highlight the hidden connections that get lost in linearization. Also videogames are non-confrontational, and do not challenge identities. However, they teach participants to feel comfortable with alternative ways of approaching things. And the mentality change you get with videogames is FAST, for one can do many iterations in an afternoon. Bill & Mellissa Gates have bought into this trend already, for general education. It should be a piece of cake to adapt them. See SciAm Febr 2014

  6. Yesterday I was idly reading an FCO ‘state of the world’ briefing from last December (prepared for a worthy committee I sit on)before filing it away and realised that there was no mention of Ukraine! So much for predictive power!

    If you are a VC investor you know that what you are investing in is a high level outcome (a future valuable company that has made its market) and the team (that may have to change over time) whereas ‘the business plan’ is a necessary fiction. In development we seem to invert this, so what is important is the ‘plan’ (as safe, contained and fitting to the current context, rather than future outcome, as possible) and the agility and judgement of the team is secondary. But then, of course, VC investors are happy if one or two (out of ten) investments truly succeed! A ‘success’ rate that development funders could never tolerate, even if those successes were such as to transform the landscape!

  7. Provocative title and discussion but of course depends on the definition of scenario planning. In the humanitarian context it used to be called contingency planning but that IS now an unpopular buzzword so we all call it something else. At a recent meeting where there were five humanitarian agencies in the room we had four different names for this “what if” thinking. It was called contingency planning, scenario based planning, emergency preparedness planning and hazard event planning. As humanitarians we should and do spend some time thinking about “what if” xxx events, what might happen and how we as humanitarians should be prepared to respond to such disaster events? But this type of “scenario” planning is not perfect in the humanitarian world either!

    Research indicates that contingency planning by humanitarian agencies tends to be unproductive and fails to adequately raise level of preparedness in the country office. Choularton (2007) found that contingency plans are “lost, forgotten or ignored, and the relationships and common understanding about what to do in a new emergency never develop”. Levine, Crosskey and Abdinoor (2011) noted that contingency plans were “long documents with a great deal of information, but they rarely include the elements that would actually help speed up response if the contingency did occur”. Similarly a World Food Programme (2008) evaluation found that “contingency planning as a separate activity had in itself little impact on response… and the plans themselves were almost never used”. The WFP evaluation concluded that contingency planning was a technical exercise, undertaken by a small technical group with the sole purpose of generating a plan. The emphasis was on the production of a highly detailed plan rather than on the preparedness process (WFP 2008). Therefore contingency planning is failing to increase the level of emergency preparedness in the country office.

    Research indicates that the emergency preparedness process is often inadequate, poorly implemented, and even missing important components. Choularton (2007) found that country offices are often caught in the “scenario trap” where contingency plans are nothing more than summaries of scenarios. Similarly Hellmuth et al (2011) in a study on emergency preparedness and the use of scenario based planning observed that the lack of scenario based planning with linked preparedness actions as one of the key stumbling blocks in emergency preparedness. Suarez and Tall (2010) found that there was a need for preparedness actions to be developed against specific early warning indicators. Levine et al (2011) observed that emergency preparedness processes lacked clear decision-making trigger indicators, seldom used seasonal or crisis calendars and ignored the time lag between the decision to act and the implementation of that decision. More critically Levine et al found that responsibility and accountability for preparedness actions were seldom if ever assigned to individuals within an agency. Levine et al (2011) concluded that, “Agencies were not used to the idea that preparedness should be quantified, or that anyone should or could be held accountable for managing it”.

    Our research indicates that generally country offices and national partners continue to be inadequately prepared to respond effectively immediately after a disaster strikes. All humanitarian agencies we interviewed were in the process of reviewing or were interested in improving their emergency preparedness process. The literature indicates that there are serious defects in the emergency preparedness process and this is recognised by many of the mainstream humanitarian agencies. The WFP evaluation emphasised the need to “re-conceptualize contingency planning from being a stand-alone operational planning activity to an element in an integrated strategic problem-solving process conducted within an overall inter-agency framework” i.e. integrated into the normal routine of the country office. Choularton concluded that achieving and sustaining dynamic emergency preparedness processes remains a major challenge for humanitarians who are typically beset by competing demands, limited staff time and constrained resources. This is a major shortfall of the emergency preparedness process. ‘Agencies should carefully consider how to develop emergency preparedness systems and contingency planning processes that reinforce each other, and thus improve the effectiveness of their humanitarian response’ (Choularton 2007). Levine et al (2011) concluded that, “Preparedness could be given a huge boost if everyone took it more seriously.”

    As humanitarian agencies we must take emergency preparedness planning seriously. However we must improve the preparedness processes so that our country offices can increase their preparedness level. The ALERT consortium, which consists of six humanitarian agencies, is currently working with a broad spectrum of stakeholders including our country offices and national partners to develop a more coherent, logical and harmonised emergency preparedness process (irrespective if we call it scenario planning, contingency planning or any other “buzz phrase”) so that our humanitarian responses are faster, appropriate and more effective.

    1. Really interesting Andrew, makes me wonder if we should be a bit more direct. If the purpose of the exercise is to prepare staff for the unexpected, maybe it would be better to just run regular simulation games at regional, even national level, which would both build capacity, and identify shortfalls in terms of logistics etc. Who in the humanitarian sector does this (can’t believe no-one has thought of it!)?

  8. Yes I think simulations are “later” part of the emergency preparedness process, but you can’t do simulations in isolation either. For example Disaster preparedness is defined by FEMA (United States Federal Emergency Management Agency) as “a continuous cycle of planning, organizing, training, equipping, exercising, evaluating, and taking corrective action in an effort to ensure effective coordination” during a disaster response. So in this definition we see that preparedness is a process, which includes planning (dare I say scenario planning), organising, exercising (simulations) etc. The biggest problem we face in the humanitarian sector is that as Levine et al (2011) puts it, “preparedness could be given a huge boost if everyone took it more seriously.” Much as we would like to do simulations there is seldom money for training, planning etc. much less simulation exercises. So maybe humanitarian agency CEOs should take note of Levine et al comments and take preparedness seriously. They must enable humanitarian departments and country offices to get prepared by allocating adequate and appropriate levels of funding for disaster preparedness.

  9. Frustrating to read this post having just returned from the launch of a research programme which is testing the use of transformational scenario planning as a tool for informing the development of climate change adaptation strategies that do more than fix current risk and take seriously the possibility of 2-4+ degrees increase in average global warming and all the changes that that will bring!

    To be honest – I felt myself reacting along the lines ‘how nice to bash something you’ve had the privilege to learn about’,

    The programme I refer to above is using transformational scenario planning to understand past, present and future. The process is being led by REOS, whose founder used the process to develop scenarios for South Africa’s post apartheid government and was introduced by a REOS partner – an extraordinarily grounded, intelligent woman who didn’t carry any of the ‘superficial-over-excited-with-own-enormous-genius’ energy that Shell scenario-ista’s and their like may carry. She emphasized social learning, relationship building and ‘seeing the whole system’ aspects of scenario planning. There is good literature – including the SREX report – that argues these are needed to address climate and other complex, dynamic change.

    And I’ve used scenario planning in my previous work in environmental planning and seen it move people working for NGOs in Belize into tighter relationships of sharing and collaborating generating the feedback loops in the present that enable the future to change.

    Time future IS here in time present – that is what scenario thinking can help us see more clearly and I personally find it empowering and a necessary inquiry.

    1. So maybe a lighter, less process heavy approach works better than the ‘turn the levers and produce your 2×2′ Shell version?

    2. Helen – very interesting what you write, I’m working on CC adaptation too, is there a link to the project you mention?

  10. Way to write a blog post with a catchy title. Very much like “[insert your -ism of the day] is dead!” blog headlines that are so popular these days. Especially with those who care about web traffic more than substance.
    Let’s start with the fact that anything, and I mean anything can be a waste of time; math, economics, medicine, or even international aid. It all depends on context. Anything you do out of context can be brought to a level of complete irrelevance. But if the title of this piece was “Scenario Planning Should Be Done In Context of XYZ” it wouldn’t make a catchy headline would it?
    Secondly, the article is internally in consistent and ends with “only then should we think about devoting a few hours to scenario planning.” So is it a waste of time or not? It would be a lot stronger (albeit still inaccurate) if the author stuck with his position rather than flip flopping.
    Pieces like this are very distracting and damaging to thoughtful discourse, especially when they are written by influential people. Very disappointing indeed!

  11. Good post Duncan. In a complex world it is far more important to increase foresight of what might happen then trying to forecast that which is essentially impossible.

    Foresight is built upon present insights – increasing awareness of what’s really happening – and understanding the origins and dynamics of change lets us be better prepared for eventualities by noticing weak signals of change before they become too big or costly to respond to.

    Apologies in advance (poor etiquette linking on someone else’s blog page maybe?) but here is a link to a very short presentation demonstrating how this might be viable (and has been used extensively in development).


  12. Duncan,

    I made a comment to this post – it was amusing to see that neither you nor the others even acknowledged it. Those who “know” the prevailing paradigm do not take kindly to being suggested to that they may explore its “adjacent possible” – the first point of my comment. It is well known that those who know the truth no longer seek it – I’ve been to too many meetings where establishing the pecking order had replaced the quest for understanding the context.

    I juxtaposed the concept of “silent transformations” to that of predictions. It is one of the fundamental concepts in Chinese culture. What lies ahead of us are “silent transformations”, not goals or targets. It replaces the dual thinking “right – wrong” with an adaptive and evolutionary concept. It takes some time to do the switch, so much we are keen to “know” the way ahead.

    The paradigm of truth and rationality, and consensus, underlay much of the discussion. As an alternative I proposed to focus on “change in mentality” as the desirable outcome. Mentalities are ways of doing things, not projects, or goals. Mentalities get changed by trial and error, not reasoning, or check-lists.

    In the XXI century we are blessed with totally new tools, in particular we can replace the linear reading or arguing process (the copper wire of thought transmission) with multidimensional games (the fiberglass connection). Games allow us to become comfortable with possible outcomes, networks, and what not. Deep videogames are replacing rhetoric, and I’m sure, much of education soon, because they replace explanation (the current paradigm) with exploration. They bring together the theory and the context in a ludic and non-confrontational way (which is a prerequisite for changing a mentality). Games like SimCity and their avatars can teach us how the system as a whole functions and you are engaging all the senses and emotions, rather than the plodding word.

  13. Really enjoyed reading this discussion. I want to pick up the comments made by Erinch and Nicholas C. There’s lots of clever stuff here about what planning is and what it looks like. But surely the point really is that a ‘plan’ (scenario-based, foresight or otherwise)is indeed a necessary fiction – a carefully constructed image of something that almost certainly will not happen. It is the decisions and actions of the people who are working to a plan that determine the impact we have. A critical justification for the plan must be to identify the people we need and the structures and support they might need in order to react and act to the sorts of changes/events that will happen around (almost certainly not according to) the plan.
    Yet our conversation almost always seem to go back to how good the plan is. Perhaps this is partly because, unlike investors, development types are often pursuing objectives that are themselves very complicated and so need to refer back to something that says what we want to achieve for now, as well as how we plan to get there.
    This reminds me of discussions about the recent demise of the England cricket team. The succeeded for a while by having rigid plans, specific tactics and people who can execute them by the book. But ultimately they started losing to people who knew what they wanted (to win) and knew what the possible twists and turns of a game looked like (messy, unpredictable, but with certain critical shifts and turning points), and who critically had internalised the planning bit and were ready to react to what is in front of them… to know that however good the plan is, the most important thing may be when to throw it out of the window and do what the situation demands to exploit a ‘game-changing’ opportunity.
    Like sports teams, we risking coaching that instinctive and creative problem-solving out of people if the plans (and the planners) are held in such high esteem. Perhaps no co-incidence England lost the plot around the time when their most talented and instinctive game-changer became an outcast in his own team.

    1. Rob – you write about ‘a’ plan. Yes you are correct that any one plan of the future is almost certainly going to be wrong.

      But the whole point of scenarios is to develop several visions of the future, that depend on unpredictable events.

      In that way you are primed about a range of possibilities and can much more easily spot change when it happens and even have a plan to deal with it.

      Scenarios are not rigid plans; a good football manager will change things at half-time if the team are doing badly – s/he will have mapped out several possibilities and will therefore stand a better chance of spotting the problem and fixing it. Mourinho I believe is a good example of this.

Leave a comment