Scenario Planning 2/4 – Model Design for better decision-making.

Posted by:

Scenario Planning 2/4 – Model Design for better decision-making.

In this series of blogs we will be looking at how scenario planning can help organisations produce better and more agile plans and forecasts. In the last blog we looked at the differences between planning and forecasting. In this blog we look at how models are built.


Scenario Planning: Dealing with uncertainty

Scenario Planning is a planning method that makes an assumption that the future cannot be predicted with any degree of accuracy because of things that are unknown.   What it does is to allow management to envisage a range of market developments and business relationships, from which predictions can be assessed before a decision is made. It can also be used to produce a range forecasts for different, but likely, business environments, again for management review so that an informed decision can be made.

Scenario planning starts by dividing business knowledge into two broad domains:

  1. Things that the organisation believes they know to be true (e.g. the amount of raw materials and energy required to make a product)
  2. Things that they consider to be uncertain or unknowable. (e.g. sales conversion rate in relation to advertising spend, or future interest rates)

The art of scenario planning lies in building a model of the organisation that describes how the organisation operates, i.e. that shows how resources and workload are linked to business processes. For example, to produce a certain level of product requires a defined amount of materials, staffing, energy and time. These amounts can have step-changes or limits, e.g. once the volume of product exceeds a certain level, overtime working will be required; or that production machines will be at their maximum capacity. These relationships can typically be hardcoded as rules. E.g.

  • If product.volume < 5000 then staff.cost = product.volume * staff.rate
  • If product.volume > 5000 then staff.cost = (5000*staff.rate) + ((product.volume – 5000) * (staff.rate + overtime rate))

In the above, the rules take into account that staff costs increase once volume goes above 5000 units

To this model, uncertain factors that impact specific business processes can be set up as variables whose values can be changed. In the above example, staff.rate and overtime.rate are variables that can be changed for different scenarios. Other variable examples could be things like interest rates, which is then used by the model to recalculate profitability. In this case the model could be programmed so that rate increases above a certain level would cause the volume of sales to be reduced, which could be redressed by decreasing prices that in turn affects product contribution and hence overall profitability.

By setting up a model in this way, management can then run multiple scenarios of the future by changing both the rules that relate resources, workload and business processes, as well as the variables that denote uncertainty. The output is a range of potential future outcomes that can be discussed strategically and decisions made accordingly.

In our next blog we will consider what to look for in a scenario planning solution. 



About the Author: