Business Analytics Blog

Business Analytics Blog

Opinions expressed on this blog reflect the writer’s views and not the position of the Capgemini Group

Demand Sensing by Yigit Gungor, Hark Atwal and Nigel Lewis

Improving existing operations to reduce risk, increase margins and yield competitive advantage is more important in the current market than ever – and we believe that analytics can provide the answer. This week Capgemini’s Business Analytics team are releasing a daily blog on some of the key use cases of Operational Analytics – today’s topic is Supply Chain Analytics

Demand sensing – creating a win-win situation for both consumer product and retail companies

Last autumn two clothing companies, Next and M&S, reported lower sales due to unusually warm weather. Predicting demand accurately has a direct impact on the performance of a company, though being able to do so depends on many internal and, sometimes unpredictable, external factors. Although it is almost impossible to plan for unpredictable events like weather, if necessary steps are taken, companies can react to changing consumer trends and demand signals much faster. In this post we are going to explore the benefits of collaboration between consumer product companies and retailers, and how demand sensing can help both in using demand signals in a challenging environment.

What is demand sensing?

Demand sensing can help improve organisations to be more demand driven. We can summarise demand sensing as incorporating downstream data with minimal latency to use both consumer and channel data, hence increasing organisations’ responsiveness to demand signals. Collaboration between CPs and retailers is critical to allow integration of downstream data such as ePOS. In terms of methodologies applied, demand sensing differs from traditional demand planning; instead of relying on time series models, more unstructured data is processed using predictive analytics and pattern recognition algorithms. Demand sensing will have an impact on the short term and is not a substitute of the demand planning systems, but rather a complementary tool to help make better and faster decisions.

Certain industries are more suited than others for a demand sensing implementation; let’s consider the fashion industry. With few exceptions, most companies do place their orders 3-6 months in advance in preparation of a new season. By the time demand signals are observed, it’s generally too late for the company to adjust inventory levels. However, there are still some potential benefits that companies with long lead times can achieve; such as store level assortment optimisation and out-of-stock reduction. Conversely, FMCG lead times vary between 4-12 weeks, which enables companies to react to demand much faster and adjust inventory levels based on observed demand signals.

Why change now?

Demand sensing itself is not a new concept and it’s been around for more than a decade. Some large CPG companies have already implemented and started benefiting from demand sensing solutions.  Some of the early criticism of demand sensing points to usage of only historical data but not incorporating useful and forward looking data points like promotion plans, weather and supply constraints.

With the advancement in technologies and processing power, more and unstructured data can now be handled, and thanks to improved capabilities of machine learning, with minimal supervision. The new developments enable additional datasets to be analysed, previously very time consuming and cumbersome, but which can now be analysed in real-time. Incorporating datasets like promotions, weather, and social conversations/sentiment now addresses the earlier criticism and enables predictive demand signals to be uncovered.

What’s in it for CP companies and retailers?

To enable a demand driven approach, collaboration and data sharing (e.g. ePOS, promotion plans) is critical between retailers and CP companies. Such data exchanges will lead to better planning and improved forecasting accuracy which is a mutual benefit to both CPs and retailers. With improved forecasting accuracy, retailers keep their shelves stocked with in-demand products. As a result, both CPs and retailers will not lose out on opportunities to maximise sales and revenue. Similarly, due to accurately predicting demand and effectively managing inventory levels and production cycles, CPs and retailers reduce inventory and warehousing costs. Another direct benefit is consumer satisfaction: with the right products stocked at the right store at the right time, customer service levels will remain high, resulting in improved customer satisfaction and may lead to higher customer retention rates. 

Benefits

Figure 1: Benefits for different business units



The scale of the benefits can truly transform companies. A global beverage maker collaborated with its retailers and exchanged valuable information such as ePOS, promotion plans, and inventory levels.

Incorporating 300+ variables, a demand-driven value chain including state of the art demand sensing system was built, and as a result increased +26 weeks store level SKU forecast accuracy to 95% from previous levels of 60%.

To extract the full benefits out of demand sensing, companies have to ensure their value chain is reactive enough to turn demand signals into action; without a consolidated and reactive value chain even the most sophisticated forecasting model would be useless. Demand sensing should be considered as part of a wider company transformation to make all parts of the company demand-driven involving significant changes to people, culture and process.

Capability architecture

Figure 2: Capability architecture



As part of annual business planning, a long term strategy is agreed for demand planning, assortment planning and promotion planning. Long-term strategy is then executed and reviewed with the feedback (POS) weekly/monthly/quarterly. By the time relevant teams review the feedback, there will be a lag between demand signals and action. Processes like demand sensing, assortment optimisation, out-of-stock reduction and trade promotion optimisation can help turn insights into action and reduce the latency in the process.

Transformation

Figure 3: For a successful implementation, People, Process and Culture needs to change



On top of process changes, the people and culture of the organisation should also be changed; instead of silo forecasting, a more agile and collaborative approach must be adopted, keeping communication as the highest priority, thereby enabling a consensus-based forecasting process. Additionally, there should be open communication of data and information between retailers and CP companies, where trust is at the centre stage of the relationship. Finally, internal divisions within CP companies should also share insights; demand signals identified by one team should be leveraged by other teams to increase overall business performance.

Quick wins and long-term rewards

Demand sensing goes well beyond traditional approaches to demand planning and forecasting.  A change in data, analytics, people, processes and technology is fundamental to success, and poses a significant challenge to CP companies and retailers alike.  However, an agile and comprehensive approach will help companies to gain early benefits, while building the required capabilities and harvesting the rewards over time.   

If you would like to know more about Supply Chain Analytics, our Operational Analytics offer or where Capgemini has delivered this capability before – please feel free to contact one of the authors.

About the author

Nigel Lewis
Nigel Lewis
Nigel leads the Capgemini Consulting’s 35 strong Business Analytics team, which delivers analytical, operational and strategic modelling solutions to clients. He has 18 years consultancy experience as well as 8 years experience in the UK gas industry. Nigel has successfully managed complex projects in both the public and private sector, including capacity modelling, simulating supply chain operations, strategic business modelling to support future policy decisions, and implementing complex demand forecasting systems. Nigel is currently focussing on the development of Capgemini’s customer analytics and analytics advisory services.

Leave a comment

Your email address will not be published. Required fields are marked *.