Sensitivity analysis – how wrong will your forecast be?
The one thing you can be sure of when setting your budgets and forecasts is that they will be wrong. The skill of the accountant is making sure that they land within an acceptable range.
So how can owners and managers understand when they need to be worried and when they can relax?
One method is to subject your forecasts to a sensitivity analysis.
A sensitivity analysis is a way of working out how far certain key drivers can move before it becomes an issue for the business.
By analysing theoretical changes in these key drivers, managers can understand not only their importance but also plan for when things don’t go as expected.
A sensitivity analysis isn’t just for budgeting time though. They can be used for any major change in the business such as when considering a new project or bidding for contracts. Conducting a sensitivity analysis at these times will give a clear indication of where problems in planning may exist.
So how do you start with a sensitivity analysis?
The first requirement is to have an initial forecast or budget that you wish to work on.
Although there are some excellent products available to help, an analysis can be carried out in a fairly simple spreadsheet and this makes a convenient system to use.
It’s important to build in as much automation into the model as possible as it will be used as the basis of a series of forecasts through the process.
A choice will need to be made as to what variables are going to be flexed.
Although it’s possible to conduct a sensitivity analysis that flexes everything, this is not very desirable as it tends to make the whole exercise into a cottage industry.
Instead, managers can make the best use of their time by only re-forecasting variables that are key to the project or business. It is also important to choose measures that are material to results, for instance, it is pointless for a company with a turnover of £60m per year to be re-forecasting a £100 phone bill.
The best way to approach the analysis is to see it as a ‘what if?’ exercise.
What if we lose our biggest customer? What if the price of electricity rises? What if the price of our raw materials doubles?
By identifying these key drivers the accountant can build in functionality that will then flex any dependent variables based on change in the key variables.
So for example, if the price of raw materials rises for a manufacturer then their cost of sales naturally also rises and as a consequence of this, they may choose to raise their selling price.
Once the independent and dependent variables have been identified and built into the model, managers can then start to flex these to find where pinch points arise.
Typically the approach taken will be to either flex until a breaking point is reached or to overlay scenarios to model out what the future may look like.
In the first case, the model could be used to change a variable until the point where the company makes zero net profit.
Managers would be able to see that if the cost of raw materials rises then the selling price will rise and consequently sales may reduce. They can also see the maximum amount that their raw materials could increase before the company starts to face difficulties.
In scenario planning managers invent a future where a series of changes all combine to make the future different to the one predicted in the original budget.
This may include a rise in fuel prices, raw materials, wages, taxes or any number of variables. Although each of the changes may be small the combined picture might produce a very different outcome to that predicted by the budget.
We can see that not only does the scenario approach identify where the company may face a reduced profit but it also helps managers understand what the biggest drivers to profitability may be and also how all of the variables interlink in the company to produce the final outcome.
Sensitivity analyses are formed from a process of setting a budget, identifying key drivers, setting a materiality level, flexing the key drivers and overlaying scenarios.
Using sensitivity analysis in this way allows owners and managers to understand how a series of alternative futures could look for a business and allows a clearer sight of the way that the organisation operates.