What is demand forecasting?
Demand forecasting uses historical trends, market knowledge, current activity and expert predictions to determine the potential demand for specific products during certain times.
For example, a jeweller may know from experience that their busiest seasons are around major holidays. They also know they’ll have a big advertising push for a specific line of necklaces before Mother’s Day, so they expect high demand.
Why is demand forecasting important?
Demand forecasting is crucial for businesses that sell goods, especially if there could be unexpected supply chain delays.
Let’s look at the benefits of demand forecasting:
Prevent out-of-stock products
This is perhaps the most significant reason for demand forecasting, especially from the customer’s perspective.
Demand forecasting can help you assess which products are most likely to be in demand at different times of the year and determine what quantities you’ll need in stock. When you do this accurately, you can reduce the risk of running out of stock.
The last thing you want is plenty of eager buyers and insufficient stock.
Prevent over-ordering
Just as you don’t want to run out of in-demand items, you also don’t want to find yourself with a stock room full of slow-moving products you can’t sell.
Over-ordering can erode your profit margins. You may need to sell the items at a steep discount (or even risk not selling them), to free up valuable inventory space.
Enhance customer service
When you have what customers are looking for in stock and ready to go, you’re delivering a great customer experience.
They’re more likely to return to your business in the future, which can mean more sales and increased profitability down the track.
Improve decision making
When you can determine the demand for certain products in advance — and accurately estimate seasonal fluctuations — your business is at an enormous advantage.
For example, you may realise that with the tourist season comes increased demand, so plan for this ahead of time in your procurement of stock and inventory and in your staff resourcing.
Increase profitability
You can allocate resources better and reduce inventory costs by accurately predicting customer demand. This helps drive efficiency, which in turn, helps increase profitability.
Types of demand forecasting
Active and passive demand forecasting
The main difference between active and passive demand forecasting is that active demand forecasting incorporates different data points to assess potential demand, including:
economic trends
the company’s growth or advertising campaigns
historical trends
changes in the market, such as in the manufacturing industry.
Passive demand forecasting relies entirely on a business’ historical data.
Internal demand forecasting
This type of forecasting assesses how well your existing team and resources can meet current and projected demand.
For example, you may realise you need to:
move your inventory to a larger warehouse
upgrade to a bigger storefront
hire more team members.
Long-term and short-term demand forecasting
The main difference between short-term demand forecasting and long-term demand forecasting is that short-term demand forecasting makes predictions within the next year.
Long-term forecasting looks at periods over a year, which can help you understand and plan for broader market trends as well as adjust for annual and seasonal patterns.
Macro and micro
The main difference between macro and micro-demand forecasting is that micro-demand forecasting looks at a specific industry or customer segments to make projections.
For example, an international ecommerce store selling outdoor goods might see long-term weather forecasts for increased rainfall along Asia’s coast, but a dryer-than-average year in Australia. They’ll prepare for increased demand for raincoats and umbrellas for the Asian market, and they’ll be ready for their Australian base to order sun hats and sunscreen.
Macro-level forecasting, however, looks at broader and more external market influences. It assesses how the economy, consumer trends and competition may affect demand holistically.
Qualitative vs. quantitative demand forecast models
Qualitative
Qualitative demand forecast models use experiences, opinions and estimates.
Expert knowledge plays a huge role in the predictions you develop in qualitative forecasting and may involve internal or external experts (or both). There are three commonly used models in qualitative demand forecasting:
Delphi method
The Delphi method involves sending a questionnaire to a panel of relevant experts and then analysing the feedback. Businesses may need to pay these experts for their time, so a more affordable alternative may be to to survey customers or business peers via social media
In-house experts
This involves working with a panel of “expert” employees who function almost like a focus group to review data and offer predictions based on their experience.
Market research
Market research may include a variety of approaches: reading reviews, studying competitors, and monitoring customer social media engagement, for example.
Quantitative
Often called “statistical demand forecasting,” quantitative demand forecasting is all about the numbers.
Quantitative demand forecasting combines historical data with algorithms and mathematical formulas to assess potential future demand.
Causal
Causal quantitative models provide explanations based on past historical data, looking for a cause-and-effect relationship so you can better predict demand.
There are two causal models:
econometric, which looks at historical sales data and economic trends and changes
regression, which looks at multiple factors inside and outside the business to determine what may reliably affect demand.
Time series
Time series quantitative models assess historical data to establish patterns over different periods. These models are helpful because most businesses have natural fluctuations with somewhat predictable patterns.
Different models include:
irregular, which may assess demand data in a specific period that may not follow seasonal trends (like the pandemic)
seasonal, which considers specific seasonal changes that recur year-over-year
trends, which may assess changes in demand based on events that may recur, like a recession.
How to forecast demand
1. Gather the data
Collect accurate, reliable data that details:
market research for past trends and future guesses at market changes
how your current year is tracking compared to previous years
sales forecasting data
past years of data from your business, particularly data showing seasonal fluctuations.
2. Choose your model and input the parameters
Choose the model most relevant to your business, and assign parameters for data analysis accordingly.
3. Analyse the data
Get a “numbers person” on your team to analyse the data. Better still, use software to pull off reports and generate visual displays of what it all means.
Get control of your stock with MYOB
With MYOB’s inventory management software, you can get control of your stock, respond to seasonal fluctuations, and better forecast demand. You can:
Review your inventory information to identify trends and keep track of your business
Run inventory reports to dive deeper into inventory data, stock levels and pricing
Reconcile your inventory to help identify errors
Export your inventory data to Microsoft Excel and more.
At MYOB, we have you covered.
Disclaimer: Information provided in this article is of a general nature and does not consider your personal situation. It does not constitute legal, financial, or other professional advice and should not be relied upon as a statement of law, policy or advice. You should consider whether this information is appropriate to your needs and, if necessary, seek independent advice. This information is only accurate at the time of publication. Although every effort has been made to verify the accuracy of the information contained on this webpage, MYOB disclaims, to the extent permitted by law, all liability for the information contained on this webpage or any loss or damage suffered by any person directly or indirectly through relying on this information.