McKinsey talked what they called one of the ugliest and most common charts in strategy at the start of the year – the hairy back.
At the start of any process – a startup pitch, a sales plan, a production forecast, you need to show a chart where, from a standing start, forecasts go confidently upwards – the “Hockey stick” graph.
In reality, the forecasts are not achieved. Year after year, as actual performance stays flat or even drops, the forecasts create a series of lines that go up – and the hockey stick turns into a “hairy back” – like the ones cows have…
Why does this happen? Why do smart people get it wrong year after year?
There are a number of reasons, and they all come down to the psychology of business.
Targets are set that are not connected to the underlying business
A target is easy to set. 50% growth. 20% increase in production. 80% reduction in staff turnover.
If someone important in a business sets a target, then everyone else starts to work to try and meet that target.
Proposals are worked on to make sure the numbers reach the target rate, especially if everyone is competing for resources and the only way to get your budget for the year is to make sure you reach the target.
This is called “gaming” and a big part of good strategic planning is minimising the opportunities for gaming behaviour in pursuit of a target.
As Warren Buffet writes, if your business is based on “making the numbers”, you can well end up with a situation where you “make up” the numbers.
Targets should be based on a realistic assessment of the capacity of the business and the resources it has in place and how they create value.
Human biases get in the way
Once someone decides that a particular approach is a good idea, confirmation bias kicks in.
That person now looks for information and data that confirms their point of view, and discounts or ignores information that disagrees with it.
People believe that all you need is a goal and optimism and you will do anything with whatever you have.
In reality, what you achieve is often determined by mundane things like whether you have the resources and time to do fairly basic tasks well day after day.
If it eventually turns out that the idea doesn’t work, attribution bias helps the person explain it away by blaming whatever seems convenient.
Ultimately, although forecast setting seems very scientific, there is an emotional dimension to almost everything we do, and that causes us to use shortcuts, see patterns where none exist, tend to believe what we would like to be true, explain everything and feel like we operate more logically than we really do.
How to create better forecasts
There are two ways to improve forecasting.
First, put better decision making processes in place. Use techniques to generate ideas, encourage dissenting views and create real discussion, debate and challenge about options and what they mean for you.
Second, listen to the voice of the business. Targets need to be connected to the business, and there will be lots of data that shows you what is happening.
The trick is to convert this data into insight, separate out what is signal from the noise and connect decisions to data.