Weather Forecasting vs. Demand Forecasting: A Case of Different Expectations

This Photo by Unknown Author is licensed under CC BY-NC

Meteorologists predict what the weather will be like, whether it be tomorrow or the next few hours. 

Demand forecasters predict what customers will buy and how much, whether it be next week, next month or next year. 

When a weather forecast is wrong, we don’t hold the meteorologist accountable.  We may grumble about the inconvenience caused, but we won’t condemn him or her. 

When a demand forecast is wrong, some executives hold the forecaster accountable.  Some condemn him or her. 

We recognise meteorology as a science.  It uses data like barometric pressure, temperature, and wind speed.  A weather forecast is based on hard data and usually isn’t too far off from what really occurs.  We can excuse the meteorologist if it rains at 7:00am instead of at the predicted hour of 6:00am. 

We don’t recognise demand forecasting as a science.  We see it as a management art mixed with some mathematics using historical data.  If the forecast is off by as much as a margin of 10%, we see it as flawed.  We believe the forecaster could do better even if there is a lack of hard data. 

Weather forecasts for the next 24 hours are more likely to be accurate than predicted conditions in a week from now.    

A demand forecast of how much of an item will be sold tomorrow is no more accurate than for a prediction of how much will be sold on Wednesday next week.  The margin of error of how much customers buy tomorrow versus forecast usually is similar, if not more, than the margin of error next week. 

We pay greater attention to the forecast of a typhoon’s path in the next hour than the forecast for the next day.   We care more what’s the weather going to be like this morning than what will be happening later tomorrow.    

We pay more attention to a demand forecast at the beginning of the month than one towards the end of the month.  We care more for what we think we will sell next month than what we will sell this month, especially if this month is almost over.  We clearly can see what we will be selling this month as it nears its end so we’d find an updated forecast for the current month as not really useful. 

We accept the outcome of the weather, never mind if the forecast was right or not.  We can’t control the weather after all and meteorology looks complicated as it is.

We see demand forecasting more as speculation and something we can influence.  If demand forecasts turn out close to the outcomes we want, we celebrate and praise the forecaster.  If things turn out not what we want, we find faults in the system and in the performance of the forecaster.

Meteorologists push to improve the science of their forecasting by investing in technologies and state-of-the-art equipment such as satellites and sensors. 

Enterprise executives push demand forecasters to be better in their predictions with hardly any investment in science or technology.  The most enterprises will do to improve forecasting is to expand market research and customer surveys. 

We adapt our behaviours and plans to the weather forecast.  We prepare our raincoats and umbrellas if the forecast is rain or we wear lighter clothing if the meteorologist says it’s going to be hot.

We insist demand forecasters revise their numbers and marketers influence customer preferences if we don’t like the sales predictions.  Executives believe the adage, “we don’t predict the future, we make it,” applies to demand forecasting. 

Forecasting is an activity that should as much as possible be based on science and hard data.  But it can never be fully accurate either for weather or for market demand. 

Yet, we treat each differently. 

We see meteorology as a science and though we may be frustrated by its sometimes inaccurate moments, we adapt to what the meteorologists forecast, even if the forecast is for the next 60 minutes than for the next day. 

We see demand forecasting as an approach to be performed with high accuracy and its predictions and outcomes consistent with enterprise goals, especially those pertaining to revenue.  We expect the organisation to change market behaviour and not the other way around.

Forecasting the weather is a science.  Forecasting demand is performance.  We accept and adapt to what the weather forecast is.  We insist that the demand forecast should always be correct and in step with our objectives. 

Forecasting is about predicting what the future will bring.  It can never be 100% right.  Yet, we expect differently depending on what we forecast.  This is because there are things we think we can influence and things we know we can’t. 

We can’t change what the weather will be like tomorrow but we can change what customers will prefer. 

The trouble is we sometimes bark at the wrong tree.  We try to influence the forecaster when we should be trying to influence the market. 

About Overtimers Anonymous

Published by Ellery

Since I started blogging in 2019, I've written personal insights about supply chains, operations management, & industrial engineering. I have also delved in topics that cover how we deal with people, property, and service providers. My mission is to boost productivity via offering solutions and ideas. If you like what I write or disagree with what I say, feel free to like, dislike, comment, or if you have a lengthy discourse, email me at ; I'm also on LinkedIn:

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: