forecasting

Learning to Speak the Language of Demand

by Lora Cecere on January 21, 2014 · 9 comments

New shoes feel awkward. Blisters appear. Feet hurt. The shoes are worn for short periods. Often we shelve them to allow our feet to recover. However, over time, they slowly feel comfortable. They become a part of our wardrobe.

Learning to speak a new language is similar. Conversations are strained. Mistakes are made. Pauses are awkward. Confusion reigns. Communication is stilted. It takes time. Slowly the words take definition in everyday speech.

Nine out of ten supply chains are stuck. Growth has slowed. Complexity has increased. Companies are stuck at the intersection of inventory turns and operating margin. They are unable to drive improvements in both. The secret to unsticking the supply chain is to redesign processes to be outside-in. The supply chain processes need to be designed from the market back.

This a step change, not an evolution. Why? Most companies have designed supply-centric processes from the inside-out. The first step to making the shift is learning a new language.

Step Up and Learn the Language of Demand

In companies, there is no standard model for demand processes. It is evolving. New forms of analytics make new capabilities possible. In the traditional organization, some demand processes are sales-driven. Others are marketing-driven. However, sales-driven and marketing-driven processes are quite different from market-driven processes. <In fact, so much so that I wrote a book about it.>

Unfortunately, companies have invested money in traditional forecasting processes believing that if they make the forecast better that corporate performance will improve. Improving forecasting is not sufficient. It is about much more than conventional forecasting. While we need forecasting and we need to improve the processes, we also need to teach teams how to use new forms of demand data and adopt demand processes.

Why is this important? Supply chain leaders are fluent in the language of supply. They don’t know the language of demand. To become demand driven (or market driven), they need to learn how to speak a new language. In this process, they slowly learn that the customer order is a poor representation of demand.

Tonight, I am stuck at a New York airport in a snow storm. I have been at client’s for the last two days helping them to make this transition. So tonight, instead of making snow angels, I thought I would help readers to get started in speaking the language of demand.

New Terms to Know

The concepts of demand driven are now vogue. Many supply chain consultants will quickly rattle off case studies and proof points, but the smart supply chain leader will ground the discussion with clear definitions. Let’s start with these:

  • Demand Sensing: The reduction of time to sense purchase and channel takeaway. Demand sensing is a process, automated by technology, that reduces demand latency.
  • Demand Latency: The latency of demand signal due to demand translation of a customer purchase through the supply chain to an order for a trading partner. The time is different in each supply chain based on product sales velocity and the technologies used. For example, in a hospital, it is the translation of usage in a procedure to hospital order to a distributor and the translation of that usage to an order for a manufacturer. This time lapse varies by product and by channel. For the purchase of Tide at Walmart to translate to an order at P&G, the time is 5-7 days. For the translation of a purchase of Aleve at a retail outlet store to Bayer, the manufacturer is 60 days. As the long tail (small orders shipped with low-frequency) of the supply chain grows, demand latency increases and there is a greater need for demand sensing technologies.
  • Independent Demand. The purchase of a product by a customer in the channel.
  • Dependent Demand. The translation of this demand signal from a channel demand signal to a manufacturer or a distributor through a bill of material or a transportation or manufacturing routing.
  • Demand Translation. The translation of demand by role within the organization. Each role–customer service, sales, procurement, manufacturing–have a different need/definition for the demand signal.
  • Demand Shaping. The use of demand tactics –price, sales incentives, marketing programs, new product launch, promotions, and assortment– to increase baseline forecasting.
  • Demand Shifting. The shifting of demand from one period to another (examples include pre-shipments at the end of the quarter, stuffing the channel to get rid of stock, or shipping early) increases supply chain costs and distorts the demand signal. Try to minimize demand shifting and maximize the value of demand shaping. Get clear on the difference.
  • Forecastability. The mathematical determination of ease of forecasting (the determination of the probability of demand).  Many technologies include this in the base software package.
  • Forecast Value-Add (FVA): A methodology for continuous improvement of the demand plan where steps of the process are evaluated and the question is asked, “Did this change improve the forecast (bias and error) as compared to the naive forecast?” (For more on this topic check out the book, The Business Forecasting Deal.)
  • Naive Forecast. The historic forecast using prior month shipments.
  • Downstream Data: Use of channel data (Point of Sale (POS) and Warehouse Withdrawal) to sense channel demand.
  • Demand Synchronization. The demand signal must be connected from node to node in the supply chain and then synchronized and mapped. The most frequently mapped data elements are product hierarchies, time/calendars, and locations. In this mapping, the data granularity and frequency must be harmonized.
  • Demand Visibility. The translation of demand by role across the organization and across tiers and nodes of the supply chain.
  • Demand Consumption. The translation of the demand signal across planning horizons. In early planning products this was accomplished through rules-based consumption. New and more advanced technologies are using optimization and cognitive learning techniques to consume the forecast across planning horizons.
  • Integration. Close coupling of the data elements to use the data into software. Integration without synchronization and harmonization does little for the demand signal.
  • Harmonization. Data harmonization enables data of differing granularity and data structures to be harmonized into a common database.

Conclusion

Did I miss any? Just let me know.
And, please let me know if you have any great tips to share for the application of these concepts.
Also, if you want to practice speaking the language of demand face-to-face, it looks like I may be at the Marriott Airport Hotel in LaGuardia for a loooong time.  Flights are canceled for at least 48 hours. Look for me at the bar….

 

Things Have Changed: What Do We Do NOW?

by Lora Cecere on November 10, 2013 · 0 comments

This week I interviewed Robert Byrne, Founder of Terra Technology, on the results of their fourth benchmarking study on forecasting excellence.  For those that do not follow this work, let me give a preamble. The work done by Terra Technology, in my opinion, is one of two accurate sources of benchmark data on forecasting in the industry. The other is Chainalytics demand benchmarking.

There are many forecast benchmark studies in the industry, but most have a tragic flaw. The issue with most forecasting benchmarking is that the data is self-reported.  Demand planning processes lack standardization and self-reported data is suspect.

Background on the Study

Eleven multinational consumer products companies participated in the study. They are large and significant, representing a total of $230 billion in annual sales.

Complexity Escalates. Companies want to grow. Success in new product launch and trade promotions is critical to accomplish this goal. However, the increase in new products and trade promotions makes the task of forecasting tougher than it was four years ago. In the study, new products represent 17% of total cases shipped. New product shipments increased 10% over the last three years.  This was coupled by an increase in seasonality and promoted items. Traditional forecasting processes support the forecasting of turn volume, or baseline products, well, but are not well-suited for new, seasonal and promoted products. New products and promoted items had 4-5X the bias of turn volume. Products in the long tail of the supply chain have an average error of 70% MAPE and a 15% bias.

Figure 1.

  • Tougher in Europe. Both bias and error are higher in European supply chains. The average MAPE for North America was 36% while the European average MAPE was 45%. The average bias of European forecasts versus North America had 2% more bias.
  • Process Excellence Helps. Forecast Value Add (FVA) analysis has increased in popularity. The use of this continuous improvement process had a significant major impact on the bias and error of leaders.  The average MAPE for top performers in the study is 46% and the use of FVA and other techniques reduced bias from 7% to 2%.
  • Technologies Need to Change. In addition, the use of statistics to replace rules-based consumption (often termed “demand sensing”) reduced demand error of the forecast at the warehouse level by 33% as shown in figure 1.

My Take:

If growth is important to your business, you cannot manage demand planning processes like you did ten years ago.  My recommendation is to:

  • Use FVA: Aggressively implement Forecast Value Add (FVA) processes.
  • Focus Outside-in. Get serious about demand modeling. Many of the forecasting systems in the market just do not have the depth to do the type of modeling that is required in the face of this complexity. Reimplement traditional forecasting systems to model “what is to be sold” using attribute-based modeling. Aggressively integrate multiple demand streams (downstream data, warehouse withdrawal data, and market intelligence).
  • Flexibly Manage Attributes to Help Modeling. Manage history based on market attributes and aggressively move from ”SKU-based modeling” to an “attribute-based model” based on attribute-based views of history. Synchronize and harmonize downstream data using an attribute-based model.
  • Build Global Excellence. Carefully define the role of the region and the role of the global team in the reduction of bias and error. Actively use FVA to improve and align global modeling.
  • Implement Demand Sensing. Companies that have successfully implemented demand sensing to improve the forecast at the warehouse DC have reduced inventory on the balance sheet by 10% within two years. I do not see the same results from the multitier inventory projects.

I look forward to getting your thoughts.