Seasonality3 Sb 4Pipeline & Seasonality Analyzer

BUSINESS PROBLEM

Manufacturers are at the volatile end of the supply chain. Slight changes in demand at the patient level reverberate through the supply chain, gaining intensity. Pharmacies order more because they see more patient demand. Wholesalers and distributors order even more than the pharmacies, because they need to both fill customer demand and replenish their inventories. What the manufacturer sees is a much larger change than would be expected based on the patients’ increased demand. An interesting example of this behavior can be found in a simulation from MIT called the Beer Game.

Another factor is speculative buying, where some customers buy more product before a price increase is announced and then sell this product at the higher price after the price increase. For example, wholesalers make some of their profits this way. The net result is buying patterns at the manufacturer’s level, phenomena we call buy-in and buy-out

Buy-in results when customers temporarily buy too much and sales are higher than true demand. 

Buy-out results when customers temporarily buy too little and sales are lower than true demand. 

These fluctuations make it difficult to determine the underlying true demand for a product. Separating true demand from changes in sales due to buying patterns is important because:

• True demand reflects the true usage of a product and the product’s true success in the marketplace.
• True demand will provide much better trends for forecasts than raw, noisy data.
• Current buying patterns can be explained and future patterns can be predicted. For example, this month’s high sales of $80 million due to a customer buy-in may result in next month’s low sales of $30 million due to a buy-out. Management needs to know this.
Another complication in estimating product demand is that some products are seasonal. That is, their demand predictably rises or falls during certain times of the year. For example, allergy products sell more in the spring and fall. Some products have seasonal sales swings for other reasons, such as regular marketing programs, price increases, or data sampling patterns.

TYPICAL SOLUTION

Companies frequently resort to a multifocal analysis, comparing current sales to recent averages, talking to wholesalers, and using judgment.

LIMITATIONS ON THE TYPICAL SOLUTION

The process takes time and requires extensive knowledge of the market for each product, as well as the proper analytical tools for the task. The employee's time required could be devoted to higher-value tasks.

OBJECTIVE INSIGHTS' SOLUTION

Objective Insights has developed Pipeline & Seasonality Analyzer to help your company better understand cyclical issues with product demand and the supply chain.

The Pipeline Analyzer component of the tool uses numerical smoothing techniques to remove the short-term buying pattern fluctuations from the raw data, while retaining information about true changes in demand. Once buying patterns are removed, what remains is true demand. By removing true demand from the original, raw data, the Pipeline Analyzer component can determine buying patterns over time.

The Seasonality Analyzer component of the tool automatically analyzes the seasonal sales data for a given product and calculates the expected monthly demand and the seasonality fluctuations. The tool presents results in two ways:

• Time Indices: Taking monthly data as an example, March may be 80% of the expected baseline, while April is 120% of expected.
• Monthly Contributions: For example, March accounts for 7% of annual sales while April accounts for 14%.

Pipeline & Seasonality Analyzer is built in Microsoft Excel and runs completely automatically. Contact Objective Insights for a complimentary version of Pipeline Analyzer.

Slides 2

For more information, please download the Pipeline & Seasonality Analyzer Overview (pdf).

 

INFORMATION REQUIREMENTS

Required Inputs

• Product ex-factory sales (units or revenues) by month
• Requires a minimum of 12 months of historical data; 24 or more months of data are recommended
• No maximum limit on the amount of data accepted by the program

 

Outputs

• Demand units and revenues by month and by subproduct
• Ex-factory shipping units and revenues by month and by sub product (includes pipeline filling)