Escaping the Prison of Forecasting

Retail and CPG businesses are trapped by the disconnect between today’s digital customers and long-established demand forecasting and supply-chain processes. Find out more.

Paul Taylor
Paul Taylor
2022年8月10日 4 分で読める
Demand forecasting in the digital age

Retail and CPG businesses are trapped by the disconnect between today’s digital customers and long-established demand forecasting and supply-chain processes. Businesses in all sectors must adapt to the desire for instant gratification and the need to serve customers through their channels of choice at their convenience, 24/7. But that’s not the only challenge. Price sensitivity is increasing especially as fears over inflation, interest rates and potential recession bite. And, finally, ethical consumerism demands that all of this is delivered through transparently sustainable and fair operations. Retailers and CPG firms, increasingly challenged by digital-first new entrants, find themselves at the sharp-end of this trend. 

Yet, amidst these tightening pressures and demands, many find themselves constrained by fragile and inflexible planning processes. Billions of dollars and years of development have been invested in lean, just-in-time, globalised supply chain that did an amazing job of increasing choice and lowering costs in relatively stable times. However, disruptions, which were once occasional exceptions, have become the norm. Volatility and unpredictability have rendered many demand forecasting processes too slow and inflexible to meet the requirements of the marketplace. Plans that presume tomorrow will look somewhat like today have become a straitjacket that constrains retail and CPG business scope to act.

Planning and forecasting are an essential element in any business. But planning processes can quickly restrict opportunity and limit agility. A faster, more flexible approach to planning that leverages more granular data is needed. Quarterly, monthly even weekly forecasting cycles can miss important changes. Working with Teradata, leading retail and CPG companies are now aspiring to intra-day forecasts – using data from the last few hours to predict demand for the next few hours. 

The scale and breadth of data to be considered has also exploded. Seasonality and other simple analytics have limited predictive power in a digital world with trends driven by social media. Accurate forecasts need up-to-the-minute data on what shoppers are doing, ‘liking’ and clicking on now. Collecting, integrating and analysing data, at individual product and customer level from multiple sources in real time is necessary to spot opportunity and risk early. 

Finally, whatever forecast or plan you have, you must be ready to reassess and reconfigure it based on the next set of new data. Your demand forecast must refresh in tandem with actual demand and provide impetus to execute at scale and automate across physical and digital channels. 

If granular, up-to-the-second data from a diverse set of sources is the key to escape the straitjacket of traditional forecasting then the next step is to leverage it to support quick, clear thinking and action. Currently the standard approaches to overcome inaccurate forecasting and clumsy supply chains include stockpiling, second sourcing and near-shoring. However, all are partial, blunt and expensive solutions. Teradata’s scalability, speed, parallel processing and low cost per query all make it possible to analyse vast and complex data from multiple sources in near real time. This capability transforms forecasting from an occasional ‘heavy-lift’ to an ‘always on’ automated activity. Predictive models developed to run on live data spot new demand signals fast. They allow retail and CPG businesses to shift from retrospectively sifting through transactions to spotting patterns in customer behaviours as they happen. More detailed and timely triggers can be shared with supply chain partners to match orders more precisely with actual sales and support real-time and on-demand product delivery. 

To accomplish this, leading retailers and CPGs are bringing data from across their businesses and supply chains, combining it with valuable 3rd party data (everything from weather forecasts to social ‘likes’, traffic data, economic indicators), and refreshing it at high frequency. Using Teradata’s patented Retail/CPG data model to organise it they get a short-cut to insights. 

Working with Teradata retail and CPG businesses are already escaping the prison of inflexible planning. One European general retailer saw 2.2% increase in sales as it leveraged multi-layer modelling to unearth and identify new and unexpected demand signals. In an uncertain world it was able to increase the accuracy of its forecasting from 66% to 77%. 

Using real-time sales data at a per-department, per-store level, a UK major grocery store was able to create intraday forecasts that saw a 12% uplift in sales of baked goods. Knowing precisely what was selling, where and when in real time, enabled the business to share precise demand forecasts with bakery units within individual stores so that they always knew exactly what to bake and when. Customers had a ready supply of fresh croissants, and the store minimised wastage. 

Using scalable, cloud-first data analytics to drive business critical insights can modernise demand forecasting to meet customers’ needs. Creating precise, timely demand insights to share with operations systems and partners up and down the supply chain is helping Retail and CPG companies adjust to the new reality of today’s marketplace. If you are interested in how Teradata can help you unlock the full potential of your supply chain with dynamic, accurate and reliable demand forecasting, please get in touch


Paul Taylor について

Paul has over 20 years of experience in the aerospace and automotive industries. His career started at Rolls-Royce Aero Engines in manufacturing and engineering, broadening his experience through a series of customer facing, programme management and business orientated roles before moving into supply chain. Most recently Paul has worked with Jaguar Land Rover to manage their Connected Supply Chain programme. This programme aimed to drive a step change in how the OEM collaborates with its suppliers, to optimise inventory levels and to provide visibility, tracking and issue alerting for long distance supply chains. 

Paul has a degree in Aerospace Manufacturing Engineering from the University of the West of England. 
He has a passion for motorsport, in particular MotoGP, and has built a replica AC Cobra kit car which he now enjoys driving whenever time and the weather allow.

Paul Taylorの投稿一覧はこちら



テラデータはソリューションやセミナーに関する最新情報をメールにてご案内する場合があります。 なお、お送りするメールにあるリンクからいつでも配信停止できます。 以上をご理解・ご同意いただける場合には「はい」を選択ください。

テラデータはお客様の個人情報を、Teradata Global Privacy Policyに従って適切に管理します。