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.
- 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.
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.
It is Monday morning. As the sun rises, I find myself on the 6:00 AM train drinking coffee. I am giving thanks that I am able to do what I do.
There is nothing like a cup of coffee at this time of morning. As I hold the warm ceramic mug in my hands, the horizon rolls forward with the rhythmic sounds of the train on the track. I love the sounds of the train. I am lost in thought about the client that I am going to spend the day with. It is the end of a long project, and I am excited to share their data. There is such power in being able to pull together quantitative data with financial benchmarking analysis and qualitative interviews to help them see new insights. It is great to pull back the covers and help companies see the new trends and insights on supply chain excellence through research methods.
In work with clients, I find that they have good intentions and they want to be more outside-in and demand driven, but they get caught in a traps, because they have not changed the conversation. This will be a primary focus of my session today.
Volatility is rising, supply chains are becoming more important and complexity is making resiliency tougher. All are good reasons to have the conversation….
Here are the sticking points that I see:
- Focus Less on Perfect Numbers. Embrace Demand Error. Demand volatility is increasing and the technologies to manage demand are maturing. In this transition, it is more critical to learn to use demand data than to make the demand number perfect. As a result, the discussion needs to be less about the “demand forecast number” and more about the probability of demand. Companies need to try to reduce demand error to the extent possible, but realize that demand error is a reality of managing a supply chain. As a result, leaders need to drive the effort to embrace demand error and design the network to drive the same cost, quality and customer service levels given the level of demand error. This requires using new forms of analytics for inventory optimization and network design and doing less on spreadsheets.
- Help others to Understand the Impact of Complexity. Nine out of ten companies are stuck in their ability to make progress on operating margin and inventory turns. To understand this, a good place to start is the measurement of the forecastability of the products in the demand plan and understand how this is changing. Track the impact of rising complexity on forecastability and the impact on the inventory plan.
- Reduce Bias and Error. If only companies could sell what they forecast. Most companies have a large, and positive bias. To counteract this, actively use Forecast Value Add techniques (FVA) to reduce bias and error.. Communicate progress on a monthly basis. Push to help leaders understand the impact of demand bias on customer service, safety stock and slow and obsolete inventory.
- Help others to see the Options. Actively Design the Network. As you do, focus less on the levels of inventory and more on the trends and right sizing of the forms and function of inventory. (The form of inventory is the state of inventory and includes decisions for raw, semi-finished goods and finished goods. The function of inventory is the role that the inventory plays in driving the right supply chain response. The function of inventory includes cycle stock, in-transit stock, promotional stock, safety stock, seasonal stock, etc.) Actively model and help peers to understand the impact of rising complexity on the form and function of inventory. As you design the network, build push/pull decoupling points and buffers.
- Focus Forward. Finance and accounting use largely backward measurements. Push the executive team to focus forward in the design of measurement systems. Lead teams to focus on forward-looking business flows through the channel. Align the flows to maximize customer service taking ownership for sell through the channel not just sell-into the channel. Don’t stumble and get hung up on only measuring backward-looking measures.
Any others that you would put on the list?
This week, I am at the Consumer Goods Technology (CGT) conference. I hope to see you in my travels.