The onset of COVID and the rapidly changing environment of a pandemic-hit world has meant demand planning has become more difficult – and as a result, more critical – than ever.
In this article, we do a deep dive into demand planning and look at what steps, strategies and KPIs can be used to develop best-practice demand planning and forecasting.
What is demand planning?
Demand planning is the process of predicting what customer demand will be for a certain product. This feeds into effective inventory management and supply chain planning, and ensures customers receive their goods in a way that is cost-effective, timely and satisfactory for their requirements.
Demand planning plays a critical role in optimising replenishment and maximising customer satisfaction.
The demand planning process can be extremely complex as it requires the analysis of numerous factors, including historic and current datasets and internal analytics.
Demand planning also considers factors such as:
- Your marketing strategy
- Seasonality
- Market trends
Demand planning teams need to constantly be alert to technology advances such as in Artificial Intelligence or software tools which can be utilised to their advantage.
They must draw on all these factors to create a ‘demand plan’. This enables the company to manufacture and distribute its product in a way that, ideally, efficiently meets consumer demand. In turn, that ensures the customer will be satisfied and likely to return and/or endorse the business to others.
Why is demand planning important?
Demand planning should be a key plank of your organisation’s business strategy. Efficient demand planning ensures the business can balance its supply chain requirements and inventory planning with its customer demands.
However, demand planning is a complex process, with numerous moving parts. It requires insights from across the supply chain and customer base, and well-managed coordination from within the business.
Doing it well creates significant advantages for the business. It ensures inventory is shipped in and held at the right level to match customer demand, even as those demands shift for various reasons.
The flow-in benefits of this include:
- Supply chain costs are more efficiently managed.
- Inventory is stocked in a way that maximises space utilisation.
- It lessens the risk of being left with excess, low- or no-value stock that the business then needs to sell or dispose of.
- The business has the products available to meet the demands of consumers.
- Customers are satisfied, and more likely to give a positive review or recommend the business to others.
Effective demand planning should also help avoid the so-called ‘bullwhip’ effect, which occurs when there are unexpected fluctuations in retail demand, which then ricochets down the supply chain and potentially creates major flow-on problems.
The demand planning and forecasting process
The process of demand planning and forecasting can be extremely difficult, as it involves multiple teams across an organisation and data points from both internal and external sources. However, ensuring demand forecasts are as accurate as possible creates huge benefits for your business.
Below are 5 key elements of the demand planning process.
1. Build your demand planning team
Develop a team of effective communicators, analysts and forecasters. The need for strong collaboration underscores the need for a team with strong communication skills.
Beyond these ‘soft skills’, demand planning requires the ability to analyse data including internal records, manufacturer predictions, consumer trends and behaviours, and market shifts. It also requires the ability to know how to leverage that data and put the learnings into practice.
2. Review data from relevant sources
This step requires an understanding of what data will be useful, and what should be left out. For example, a decision will need to be made on what time frame is most useful for analysis.
In a fast-moving industry such as consumer goods, a shorter timeframe may be preferred. Conversely, an industry with slower-moving trends may indicate a longer time frame is preferable.
The data should be enough to show a clear picture of customer demands over time, availability of products, ability to warehouse inventory and safety stock, turnover rates, and an understanding of how promotions or market changes influence changing demand.
3. Collect external data
External data sets are also key to effective demand planning. Third parties such as manufacturers, distributors and suppliers can feed in external data to refine knowledge of customer behaviour.
Macro influences such as inflation, mortgage interest rates, and political environments should also be considered as they affect consumer behaviour.
4. Test your findings
Check the data against the reality of stock being ordered and warehoused and consumer demand. This should surface any supply chain or transportation issues, or a misunderstanding of consumer behaviour and demands. The demand forecasts can then be refined.
5. Determine which KPIs to track
Identify key performance indicators to keep demand planning and forecasting on track. We cover this in more detail later, but in brief, they should include sales rates, the total cost of goods, and forecast accuracy.
Demand planning must include a determination and recording of key metrics to track throughout the year.
4 effective demand planning methods and strategies
As mentioned above, the COVID pandemic challenged many traditional demand planning tools and strategies.
Uncertainty has been the norm in recent years and issues remain with supply chain delays, labour shortages and a rapidly shifting macro-economic environment. However, there are certain methods and strategies which can help, including:
Benchmarking
Benchmarking is the process of considering your business’s performance against others in the industry, and therefore being able to build realistic performance metrics.
This ensures demand planning KPIs are achievable, particularly given the unstable current macro environment.
Demand sensing
Demand sensing draws on modern software systems and AI to analyse real-time data and deliver forecasts which are more closely aligned with current trends.
This method can also be used in conjunction with the more traditional analysis of historical data to find trends.
Segmentation
Segmentation, or splitting customers into segmented groups, can help hugely when dealing with numerous data points.
Customers who show similar buying patterns can be segmented into groups by the demand planning team, with the sales and marketing teams then targeting them with certain promotions appealing to the segment’s particular attributes.
This allows you to bulk purchase certain products to meet that segment’s needs. It means the demand planning team can group the order together in the demand plan, thus saving time and cost.
Rationalisation
Rationalisation is the process of identifying, then deleting or minimising underperforming product lines.
While this may have traditionally been a time-consuming exercise, AI tools can now use data sets to quickly identify products which are underperforming and should be considered for removal.
A properly established demand plan will ensure you’ve always got sufficient stock on hand to meet consumer demand.
How to optimise the demand planning process
As with any business workflow or process, there is likely to be a benefit in auditing the demand planning process to ensure it is operating as efficiently as it can, with up-to-date technology and current data sets.
There are several steps to look at when considering the process:
- Ensure the business is analysing data over a correct time frame. Generally, sales data of between two and five years will most accurately identify trends and seasonal shifts in demand.
- Segment customers appropriately to maximise insights into certain behaviours. To maximise efficiencies, customers can be split into segments depending on their behaviours.
- Utilise data from others in the supply chain. Gathering information and forecasts from others, such as manufacturers, will help refine the data used.
- Ensure seasonal data is updated and refined on the back of changing patterns. Trends can change over time, particularly as weather patterns and temperatures change. Major weather events can prompt major shifts in consumer purchasing patterns, as can the increasing significance of special marketing events such as Black Friday and Cyber Monday.
- Consider if there are constraints to the business which need to be considered. Again, these can change over time – some constraints may be overcome with the help of, for example, improved technology. Conversely, new constraints or issues may arise as market environments change.
Benefits of demand planning software
There are numerous demand planning software options in the market, including Unleashed. Analysing data sets will tell you how much is required to be purchased, and what the replenishment strategy should be.
You can generate replenishment suggestions and create purchase order plans. Using this automated system means the business avoids over- or under-ordering and does not face inventory losses or excess stock.
It also enables the correct calculations of safety stock in case of a fluctuation or error in the forecast of expected orders.
Software with demand planning features will empower your decision-making with empirical data, giving a more calculated shot at success.
Demand planning KPIs and metrics
Key Performance Indicators (KPIs) are an important way of assessing if demand planning and forecasting are accurate.
Actual sales vs forecasts
This metric is also known as ‘forecast error’. It’s the key reason for demand planning.
It’s also simple to assess, given the forecasts should be in place, and the actuals will be available to compare. If there are frequent and/or large forecasting errors the demand planning team will need to consider if and where mistakes are being made.
The KPI can be set monthly, six-monthly, or annually as suits the business.
Forecast error is found by using the formula:
Actual Sales – Forecast Sales = Forecast Error
Forecast bias
A forecast with a consistent bias is known as a mean forecast error (MFE).
It is relatively common for forecasts to tend toward a certain direction, and that can cause issues with ordering, stocking and sales. Biases can be caused by issues including human error, and a desire for sales to reach a certain level which they ultimately may not.
It is important to identify and correct bias before it creates ongoing issues for the business including an oversupply of goods, which in turn creates obsolete or unsaleable stock.
The formula for bias is:
(Sum of Observed Forecast Errors of Multiple Periods) / Number of Observed Periods = Bias
The Mean Absolute Error (MAE)
This KPI measures the deviation from a forecast by considering its variations from actuality over a period of time.
It takes the forecasting errors – for both positive and negative skews – and averages them out, to reach the MAE figure.
The formula for this is:
(Sum of Observed Absolute Forecast Errors Over Multiple Periods) / Number of Periods
