How behavioural economics changes the rules of strategy development

The most profound challenge that behavioural economics raises for company managers and leaders concerns how business strategies are developed.  Through enhancing our understanding of how bias can insert itself into what we assume to be a rational process, unhealthy the insights from behavioural science should cause a fundamental re-think about the validity of the prevailing paradigm of strategy development.

Bias thrives on uncertainty; and the greater the uncertainty, cheap the greater the scope for distorted thinking.  Uncertainty is greatest with decisions about the long term – the strategic decisions that may take a significant time to execute and will impact the business’ fortunes for several years thereafter.

At the heart of the biases permeating the current approach to strategy development is a desire to pretend uncertainty doesn’t exist.  The Old Rules of Strategy pander to this need for certitude, creating a false but seductive impression that the future can be predicted so decisions can be made on what seems to be hard data when they are nothing more than imputed numbers generated by pliable assumptions.  As such, the old rules encourage the overconfident over-optimism and other distortions that lie at the core of poor strategic decision-making.

To counter these biases, a new approach to strategy development is required – one that is founded on an acceptance of uncertainty rather than the delusion that it can be assumed away.  The first step, however, is to outline the Old Rules of Strategy and the assumptions that underlie them.

The Old Rules of strategy

The prevailing approach to strategy development can be summarized as follows.

Firstly there is one right strategy for any one company at any one point in time to optimize the returns to shareholders over a 5-10 year time span.

This correct strategy can be accurately specified at the beginning of the period it will cover – all requisite insights are available at the time the strategy is devised.

These insights are sufficiently formed and evidentially supported so that they can be uncovered by an objective method – detailed research with customers about their emerging needs, for example.

Using these insights as inputs, the strategy can then be deduced through analysis of external factors (markets and competitors) and internal factors (offerings, competencies and financials), as guided by textbook strategy frameworks.

One consequence of the first four rules is that strategy is about ’what?’ rather than ‘how?’  The focus is on what markets the business should focus on, what positioning it should take versus competitors, what high level capabilities are needed and what key performance indicators should be tracked.

How performance targets are achieved falls to operational teams.  As a result design skills – critical for delivering effective propositions and business models – are less important than analytical ability in determining suitability for strategy roles.

A further consequence is that strategy development can be done equally well by people unconnected with the company (e.g. consultants), as by its own managers.

Even if undertaken internally, the inputs to the strategy process tend to come from a few like-minded individuals. Strategy is typically the preserve of the CEO with analytical support provided by strategy experts (MBA graduates or alumni of consulting firms).

Those in strategic planning roles do not require deep operational understanding.  This means they often lack the detailed understanding of how the business actually works to challenge the insights (and uncover the de facto assumptions underlying them) of the senior executives who own the strategy.  Informed (and potentially testable) intuitions are therefore less likely to be separated from personal value judgments that are more susceptible to bias and self-interest.

Next, the main process for testing strategy is to generate a financial forecast of the returns the strategy will produce.  This is done bottom-up – assumptions are made about the likely market share, price points and margins that the strategy will deliver.  Forecasts of the evolution of market growth, inflation rates and input costs are then loaded on top to provide a facsimile of the wider economic context within which the strategy will operate.  A button is pushed and a Net Present Value (NPV) calculation is generated.  The strategy with the highest NPV is deemed optimal.

A few different political and economic scenarios may be presented to test the sensitivity of the financial projections, but typically these are limited to a few stereotypical scenarios (such as “Euro collapse” or “Oil price at $200”).  Rarely is the strategy scrutinized for its essential flexibility or its potential to gain or suffer from “black swan” events.

The financials are modeled at a much greater level of detail than merited by the accuracy of the hard data available for the model to be based upon.  For example, a key variable in any such forecast is market size – but the concept of the market is always more complex, with many blurred lines, than such analyses can handle.

In any standard NPV analysis typically only 20% or less of the calculated financial value of the strategy is generated in the first five years – 80% or more comes in the period that begins five hence.  Yet companies struggle to budget accurately for just one year out.

It is not the purpose of this detailed financial forecasting to add insight or test the original intuition.  Typically it is only the ingenuity of the financial modelers that is tested.

Despite awareness of the deficiencies in the forecasting method, such forecasts are still the principal means of justifying a particular strategy.  Comfort is gained from the semblance of rigour given to the process of turning a rough hunch – if it sounds about right to others – into a hard financial forecast that can be tweaked to pass hurdle rates.

The positive result is deemed evidence of the strategy’s correctness.   As a result, the forecasting exercise dangerously adds confidence to those championing the suggested strategy, reducing the likelihood of criticisms being acknowledged and alternatives being considered.

The need for New Rules

As the above section highlights, strategy development is typically seen as an analytical exercise.  The web sites of leading consultancies all eulogize about fact-based decision-making or rigorous analysis.  And business schools educate students by teaching them to apply analytical techniques in case-based (i.e. analogical) contexts.

On the face of it, what strategists can learn from behavioural economics would appear to be supportive of this paradigm.  Bias thrives when there are no facts, just myths, views and beliefs.  What could be more valuable than collecting data and analyzing it so that judgments are made on evidence and not opinion?

To some degree this is the case.  Strategy development needs to start with a complete understanding of the current situation and analyses can deliver that.  But strategy is about the future and the future is unwritten – there are no facts for analyses to be built on, merely assumptions.  And, analysis-based (and, for that matter, analogy-based) predictions are limited in their accuracy.  But the supposed rigour with which they have been assembled blinds people to this weakness.  The result is far greater confidence in the prediction than there is validity.

We believe a far more important learning from behavioural economics is that the analytical paradigm of strategy development has very significant limitations and these limitations are not as widely acknowledged as they should be.

This is not to say that structured analytical approaches have no part in the strategy development process (over and above the value of delivering a thorough understanding of the truth about the past, as described above).  But it requires a different mindset because it is the process that is valuable rather than the answer generated.  If the focus is on the answer, the likelihood is that analysis is shaped to justify intuitions.  If the focus is on the process, there is a far greater chance that intuitions become informed by the analysis.

Ultimately all decisions where there is uncertainty rely on intuition.  It is far better for those intuitions to develop through a process of discovery than for them to be pre-conceived and justified by the flexibility inherent in assumptions about variables that the analytical process requires to produce an answer.

Rather than using analysis to pretend that uncertainty doesn’t exist, we believe a more fruitful approach is to accept it and seek to reduce it – make uncertainty management a key plank of strategy development.  This involves techniques such as scenario planning and assumption management.    Such an approach also requires a ‘crawl, walk, run’ mentality to entering new areas (where uncertainty is greatest) rather than attempting to sprint from the off to generate returns as quickly as possible.  Central to this are stage gate processes for investment approval and Monte Carlo simulations to show the uncertainty inherent at each stage.  And these are particularly important when looking for the next big thing, a discipline that businesses need to avoid the downward spiral introduced by omission bias.

Furthermore flexibility needs to be incorporated into strategy development.  Strategy is about placing bets and it is not possible to delay placing them until certainty prevails (because it never will).  But there needs to be an acceptance that the decisions made will often be sub-optimal.  And rather than going ‘all-in’ organizations need to retain the capacity to change direction by keeping in mind alternative (and potentially conflicting) courses of action to that being followed, all while giving the existing strategy the commitment required to give it the best chance of success.

This requires comfort with ambiguity – not typically an attribute sought in business leaders – that is manifested in a willingness to experiment with different ideas.  Moreover there need to be sufficient experiments for alternative options to be evaluated – not just one to test the preferred option.  Such an approach reduces the risk of a major strategic miss-move.

The second problem created by the over-emphasis on analysis is that the other mental processes are ignored, notably design thinking.  Design is arguably the most critical element of strategy development and value creation.  Both compelling value propositions and their underlying business models are essentially the products of design thinking – for example, what elements to include, how to combine them and, critically, what to exclude.  This underweighting of design stems in large part because rather than focus on value creation for customers and other stakeholders, strategists prefer to focus on value extraction – which customers are most profitable and should be a focus for retention, which have most growth potential, how can less profitable customers be made more profitable, etc.  This is essentially an analytical discipline.

Not that there is anything wrong with such thinking, but you can only extract value if you are creating it.  But innate egocentricity often results in the creation of value being assumed with exclusive focus on extraction.  And it is not just customers where value creation needs to be considered before value extraction, the same thinking needs to be applied to all stakeholder groups.

This egocentricity is partly imbued by the use of over-simplified analogies for strategy, such as war, chess and sport.  The direct nature of competition and the ‘I win, you lose’ nature risks too much emphasis being placed on beating competitors with the destructive consequences that can have.  In a business context such thinking confuses an outcome (generating superior returns than competitors) with an objective (developing a more compelling value proposition for customers and a more economically viable business model for delivering it).  And spoiling tactics serves neither company’s long term profitability.

If strategists are to look outside business for models to support decision making, a better area to look at is poker.  The multi-player nature of poker and the consolidation of financial wealth make it a far more analogous competitive environment (though it still represents a simplification of the complexity businesses face given all the different stakeholders they have).  But most importantly poker illuminates the key challenge that strategists face of having to make bets (strategic investments) in the face of uncertainty.

But perhaps the biggest learning that we have from behavioural economics is simply the extent of bias – the sheer number of cognitive distortions that impede our ability to reason rationally.  Everyone is susceptible and the starting point for any management team serious about trying to reduce the damage these cause is to diagnose their susceptibility.  Based on the culture of the organization and the sub-culture of senior decision-makers, what are the most likely blind spots?  What distortions are introduced by how key objectives and challenges are articulated?   What framing biases are likely given the frameworks used?

Such a diagnosis will reveal what checks and balances need to be put in place.  And there are a number of process improvements that strategists can implement – pre-mortems being one – to ensure some of the most disruptive biases (such as overconfident over-optimism) are detected.

There are also organizational changes that can be made, for example instituting the role of Devil’s Advocate.  Encouraging a culture that embraces contrarianism also helps counter bias (particularly groupthink) by ensuring that all decisions are robustly challenged.  Ultimately a senior executive will need to make a choice.  But when challenge comes from multiple sides, the likelihood of all angles being seriously considered is much greater.

All these arguments will be expanded upon in future posts, what we have called The New Rules of Strategy – a manifesto for a more behavioural approach to strategy development.  We feel that up to now discussions about the business relevance of behavioural economics have focused on a narrow range of biases – anchoring, over-optimism, confirmation bias, etc.  These decision biases are important, but in the process much of what behavioural economics can teach business executives has been lost.  That needs to change.

Share the knowledge:
  • Twitter
  • Tumblr
  • Digg
  • StumbleUpon
  • Reddit
  • Slashdot
  • Facebook
  • LinkedIn
  • Yahoo! Buzz
  • Google Buzz
  • Google Bookmarks
  • Print

Related Posts:

About Jack Springman & Ken Whitton

Leave a Comment