Audience Dialogue

Scenario planning and related methods

On this page: Scenarios | Causal layered analysis | Scenario network mapping

Related pages on this site...
Peering into the future
Glossary of 120-odd terms used in futures studies
Paradoxes, riddles, and quotations about the future
Predictions that didn't turn out as expected

This page is a very quick introduction to scenario planning and two related methods: causal layered analysis and (my own method) scenario network mapping.

Scenario building

Scenario planning (as it's often called) is probably the best known way of researching the future. The usual output is a set of scenarios (2 to 7 of them) about how a particular aspect of the world might be, typically in 10 or 20 years' time. Like other futures studies methods, scenarios are not predictions, but possibilities. The hope is that if a wide variety of scenarios is chosen, they will envelope the eventual outcome. A more recent school of thought holds that the importance of scenarios lies not so much in the possible futures they portray as in the thinking of those who develop the scenarios.

The key to successful scenario work is how the scenarios are chosen. The obvious danger is of missing the point, by selecting scenarios that turn out to be irrelevant. Probably the commonest method of choosing scenarios is the "critical uncertainties" method. The US firm South Wind Design has produced a good clear manual on this method of scenario planning Window on the Future: a Scenario Planning Primer - available online as a PDF file.

Shell Oil, the world's best known user of scenario work, has a number of scenario planning documents on its website at, including Scenarios: An Explorer's guide.

Causal layered analysis (CLA)

CLA is the newest method of futures studies, founded by Professor Sohail Inayatullah in the 1990s. It is not a method that produces prediction, but one that increases understanding of the forces that create futures. Its tools are not mathematical, but verbal. CLA uses a structure of four layers of causation, of increasing complexity. The top layer is the "litany" - superficial understandings, as reported in the popular press. Underpinning that is the second layer: social causation, as measured by social scientists. Underlying that is the third layer, of discourse, worldview, and unexplored assumptions. Deepest of all is the fourth layer, myth and metaphor, unconscious emotive dimensions. Any issue can be examined in terms of these four layers, and examining the interplay between them leads to greater insight into possible futures.

Sohail Inayatullah's website has more information on CLA. The most comprehensive work yet published on CLA is the Causal Layered Analysis (CLA) Reader, edited by Sohail Inayatullah, and published in 2004 by the Tamkang University Press in Taiwan. The book has 30 chapters (including one by myself) covering CLA in all aspects and a wide range of variants.

Scenario network mapping (SNM)

Scenario network mapping is my variant of scenario planning; the development of SNM was the subject of the PhD thesis I completed in 2005. At least that's how it started out, but as the method evolved, it became less like conventional scenario planning, and more like causal layered analysis.

How does SNM and differ from conventional scenarios?

  1. Conventional scenarios look like short stories about possible futures. SNM output looks like a road map, in which the "towns" are scenarios and the "roads" are links between them.
  2. While a conventional scenario project usually produces around 3 or 4 detailed scenarios, SNM usually produces 30 or 40 minimal scenarios. With SNM, the focus is more on the links between them - how one scenario leads to another - than on the scenarios themselves. Because the SNM scenarios are minimal, they can be expanded as necessary - so what at first appears to be one "town" can, after elaboration, end up with many "streets". In 2002 I created a scenario map of the (then) possible war in Iraq. Three years later, one scenario from that map - "troublesome US colony" - has many streets indeed!
  3. Conventional scenarios focus more on outcomes than on paths, but SNM focuses much more on how things happen. As with causal layered analysis, the "road map" of SNM is only the top layer. Underneath that, deeper layers make things happen.
  4. Conventional scenarios are produced by teams of experts, while SNM maps are produced in workshops, with a wide range of stakeholders participating.
However, SNM is not as different from conventional scenario planning as the above list makes it seem; there are as many similarities as differences.

Now that the thesis is finished, I'm planning to convert it into a book about dealing with the future, written in a non-academic style. If you're interested in methods for anticipating the future, you can read my articles on Multiple pasts and alternative futures and Three maps for navigating the ocean of future. These are initial versions: the final versions have been published in the academic journals Futures and the Journal of Futures Studies, respectively. However, both articles assume that the reader already knows a lot about futures methods, so beginners may find them a bit daunting.

Related pages on this site

Peering into the future
Glossary of 120-odd terms used in futures studies
My doctoral thesis on scenario network mapping
Paradoxes, riddles, and quotations about the future
Predictions that didn't turn out as expected