Proven Method to Inventory Demand Forecasting

Natasha EasyEcom
Jun 6, 2018 · 9 min read

Disclaimer : “This Article Was Originally posted on EasyEcom’s EasyBlog. Read Full Article at http://www.easyecom.io/blog/inventory-demand-forecasting/

Inventory Demand forecasting is the process of finding values for demand in future time periods. Demand forecasting is an estimate of demand during a specified future period based on proposed marketing plan and a set of uncontrollable and competitive forces.

In this article we will discuss in details about the various advantages and methods of inventory demand forecasting.

What Is Inventory Demand Forecasting?

Let’s have a look at this graph which is a typical supply chain management lifecycle curve.

This graph explains the stock management cycle for SKU ID 100324.

After considering various factors like the Inventory Demand they have for the SKU across the geographical locations, competition, feedback, promotions, Supplier time etc. They have reached on four critical numbers.

  • Maximum Stock Level:
    This is the optimum stock level for the particular SKU.
    In this case it is 100.
  • Reorder Point:
    At this stock level, a new purchase order is generated.
    This number is arrived after considering the time taken by the supplier to deliver the goods in the warehouse.
    Here the reordering point is 42.
  • Low Stock Warning Level:
    This level is the warning level for the SKU going out of stock.
    In this case it is 15.
  • Restocking Level:
    This is the level after the reordered quantity is delivered in the warehouse.
    In this case the number is 90.

Now imagine doing this for all the SKUs across your catalogue.

In a supply chain, The above graph poses some difficult questions.

The biggest dilemma that an organisation faces is what would be the Inventory Demand for this product in future.

How much should I order/manufacture?

More importantly,

When should I start the reordering process?

Inventory Demand forecasting is the process adopted to answer these challenges efficiently.

How to estimate the Inventory Demand for new product?

Let’s take the example of a restaurant.

Imagine you’re starting a restaurant. Obviously you have no idea how much you will be selling. In order to come up with the first estimate, need to begin with an assumption.

One of the ways of doing that will be to start with a map of chairs and tables, figure out how many breakfasts, lunches, and dinners you can serve in your working hours.

Then go from there to coffees and drinks based on the number of meals.

Use realistic averages based on location, general Inventory Demand and take into account the seasonality.

That will give a good average number of sales for each meal time.

Now multiply the hours and days to get months’ worth of projection. Next multiply the number of units and prices based on a reasonable sense of forecasted prices.

That’s your sales estimate using a scientific average.

Please remember.

This is an iterative process and needs regular revisions based on assumptions testing.

Inventory Demand estimation (forecasting) may be defined as a process of predicting Inventory Demand in future time periods.

More specifically, Inventory Demand forecasting is a scientific approach to predicting sales during a specified future period based on the proposed marketing plan and a set of uncontrollable and competitive forces.

Factors affecting Inventory Demand:

There are several internal and external factors affecting sales for an organization:

  • Seasonality of Inventory Demand
  • Competition
  • Technological failures
  • Reputation
  • Labour issues
  • Supply chain related factors
  • Inflation
  • Recession
  • Change in government laws

Iterative Inventory Demand Forecasting:

So…,

When it comes to forecasting, a lot of uncertainty is involved in any organization.

To reduce the adverse effect of these uncertainties, an organization can take an iterative approach towards determining the Inventory Demand or sales prospects for its products and services in future.

For example,

A subscription-based web business ought to be able to project website views based on its organic search placement, paid advertising, email marketing, and other alliances & promotions. Project traffic by summing up your sources, then estimate the conversion rates, and that gives you unit sales.

Now as time moves on, We need to make changes in the assumption.

For instance,

If we predicted 10% month over month growth rate by spending $5000 per month on advertising.

However, Due to budget cuts, the company can only spend $3000. That will cut down the traffic by 40%.

Similarly if a competition decides to discontinue their services, conversion rate might shoot up by 30–40% hence boosting the demand.

Constant (quarterly or monthly) revisions in predicting mechanism are recommended depending on the nature of the company products and time required for a feedback cycle to complete for that particular assumption.

Advantages of Inventory Demand Forecasting:

  • Happy Customers:
    When a customer gets the product when they require it without delay, they tend to trust you more for their needs.
    This helps in repeat purchases and loyal customer base.
    Consider a scenario where the customer would require a product on a regular basis which if not procured on time may affect his business.
    If he is able to purchase the product and get this delivered on time, he will be a loyal customer for life.
  • Reduced Stockouts:
    This is a one of the most talked about yet common issues among retailers for sales loss reasons.
    Factors need to continuously optimized to get a better estimate trend.
    An accurate forecasting method not only ensures lower inventory idle time in the warehouse but also the less operational cost is required.
  • Efficient Production Cycle:
    Forecasting involves closely monitoring present Inventory Demand to understand future.
    Responding and adapting to the changes or pattern of consumption by the consumers gives a better idea how the future is going to be.
  • Lowering Safety Stock:
    When your Inventory Demand forecasting process is accurate, it increases your reorder capacity and thereby reducing the safety stock level to free up capital.
    If a business is using proper forecasting to plan then you don’t need to carry high safety stocks to manage your Inventory Demand.
  • Reduced Idle Stock:
    Obsolete inventory is a big burden on the margin of any business.
    To reduce the burden, it’s essential to identify, repurpose or removal of obsolete inventory. It decreases the volume of inventory on hand and subsequently both direct and indirect costs of keeping the obsolete inventory will be reduced.
    Having a reliable forecasting Inventory Demand method will reduce ordering any excess stock and increases net profitability.
  • Managing Manpower Better:
    When a business suddenly starts to grow, manpower requirement is also increased to handle the operations.
    So, Inventory Demand forecasting report helps the organization be better prepared for a sudden growth in future Inventory Demand with a proper manpower planning in place.
    For example,
    In the case of a subscription business, if the Inventory Demand is going to shoot up in few weeks’ time, the recruitment effort has to start immediately in order to be able to fulfill the Inventory Demand.
  • Better Pricing and Promotion Strategy:
    With a better co-ordinated and planned promotion strategy always yields better results.
    With integrated distributor-level promotions and related forecasts helps to improve the flow of goods.
    It also achieves better results in terms of availability and stock fill rates.
  • Better Supplier Negotiation:
    When you know exactly when and how much you are required to order, the negotiation becomes easier for you.
    The supplier is also aware of the kind of business he can expect from you and hence gives you a better price.
    By having a negotiation based on logic and research you are positioning yourself as a credible customer who wants to have a long-term relationship rather than one-off spot buy.
  • Plan Sales Strategies:
    Forecasting is very helpful with Product Management, Marketing, and Product Design planning.
    Decisions on promotions, pricing, and purchasing are made with data derived from Inventory Demand forecasting. This has a positive impact on the sales and profit margin.

How to Measure Inventory Demand Forecasting?

When you have data available,

Anything can be measured but to accurately forecast Demand, the focus should be on those data points which are relevant.

Before we jump into how to do Inventory forecasting, we need to consider few crucial points.

  • A Forecast Accuracy Metric That Is Objective, Quantitative, And Manageable.
  • Deciding what to forecast?
  • Level of aggregation: Individual products or product groups? Weekly, monthly, or quarterly demand?
    Units of measurement: units or dollars?
  • Determine How Often It Should Be Measured
  • Feedback cycles of each assumption used in forecasting

But one thing remember,

The management needs to understand that no matter how sophisticated the forecasting techniques you use, forecasts will never be 100% as it involves factors that are not controlled by you and a lot of uncertainty involved in it.

Techniques Used To Measure Demand Forecasting:

While forecasting Inventory Demands, there will be two sets of products.

First is for the products that have stable Demand and has past data available.

It can be forecasted more accurately.

The second type of products are items that are new, low volume and innovative products.

It is very hard to predict an accurate forecast with the considerable uncertainty involved. And to make the problem more complex, there is zero historical data and some assumptions have to be made to calculate demand.

The techniques used can be broadly divided into 4 different categories.

  • Trend forecasting:
    These are short-term forecasting methods.
    When a particular type of upward or downward trend for a particular product is involved, this method is used for short-term forecasting.
    The double exponential smoothing, regression, triple smoothing etc are few techniques popular in this category.
  • Graphical forecasting:
    When you have data and you convert them into a graphical representation, it conveys the pattern visually.
    Visual representation of data is easier to comprehend.
    This technique can give you a general trend without getting too much into understanding the data. Previous Inventory Demand exploration, trends, and patterns help you forecast easily.
  • Qualitative forecasting:
    When historical data is unavailable or irrelevant or are scarce, the forecasting is done based on an intuitive or judgmental evaluation.
    When a new product or a new innovation is launched, this scenario arises.
    Some typical qualitative techniques are based on personal insight, sales force feedback, panel consensus, market research, visionary forecasting, and the Delphi method.
  • Quantitative forecasting:
    When a historical Demand data is used to project future Demand, it becomes more accurate and relevant.
    The available data and the other relevant factors are taken into account while forecasting the future demand in these methods.
    The popular methods adopted by organizations are the Extrinsic and intrinsic techniques, time series forecasting methods (relying on past data) supplemented by qualitative judgments.

Inventory Forecasting Best Practices:

Keeping few very crucial points in mind while calculating Inventory forecasting gives maximum output.

  • Get input from various stakeholders. Take input from Sales, Marketing, and Finance.
  • Competitors sales data
  • POS data
  • Amount of obsolete stock
  • Frequency of stockouts
  • Shipments
  • Orders
  • Measure Forecast Accuracy at the SKU, Location, and Customer Planning Level
  • Adjustments Based on Feedback of Current Cycle & Focus on exceptions
  • Talk with customers
  • Review the data for trends

Conclusion:

It is always very beneficial to have a great Demand forecasting team who work on a regular interval to understand the trend and derive accurate Inventory forecasting method.

It is important to understand what kind of data is more important with respect to forecasting accuracy.

Is it the external data like competitor sales, POS data, sales team forecast or the internal data like stock-outs, shipments, orders, etc?

Apart from this,

It is also important to determine which time buckets are most suitable for forecasting. For example, whether to use monthly time buckets or weekly time buckets for planning. All these factors changes from organization to organization.

Nobody can predict 100% accurate inventory forecasting every time.

It is very essential to understand this and review the past forecast, learn from the trends and improve the accuracy.

“You should learn from your competitor, but never copy. Copy and you die.”
Jack Ma, Executive Chairman of Alibaba Group

Now Its your turn,

Let us know which method you used and worked for you in past, we’ll add them to the list very soon.

Cheers!

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade