Modern Retail

The Four Steps to Leverage Demand Forecasting Success

Forecasting

‘Forecasting’ can be a complicated term to define, and even in GSCOP’s description, it has proven to be challenging to understand exactly what GSCOP actually means in terms of what forecast data it requires retailers to share with suppliers;

  • A 6-week aggregate weekly order prediction
  • An annual prediction of the volume of every product which will be purchased from a supplier
  • An order, for delivery tomorrow
  • A 14-day daily prediction of short-term product sales, allowing for weather and promotions

Which of these should be measured against orders to determine the accuracy of a forecast? Which of these forecasts is most useful to the supplier? 

The Grocery Code Adjudicator declares that suppliers to UK grocery retailers are entitled to prompt and accurate forecasts, yet, forecasts are still proving to be one of the top three issues suppliers face. Many suppliers have reported distinct variations between forecasts and orders, receiving poor forecasts from retailers, as well as penalties if they do not meet the required service levels. 

To reform the process as it currently stands, the GCA’s Best Practice Guide makes 17 recommendations, including: 

  • Ensuring that suppliers can access adequate sales data 
  • Closer collaboration between retailers and their suppliers
  • Regularly reviewing forecast performance
  • Ensuring that suppliers are able to get access to supply chain or buying teams to share intelligence and discuss forecasts or orders 

These recommendations are all achievable through both parties being willing to share and discuss data in an easy, consistent and accessible way. Therefore, in order to ensure accurate demand forecasting to not only support retailers in complying with GSCOP and the GCA requirements, but also addressing and improving supply chain efficiency, availability and waste, there are four critical steps to consider:

1. Clear and consistent formats

Retailers and suppliers may use different ways of coding the same SKU, different names for branches or even distinct units of measure. Most suppliers will forecast at SKU level by location, but forecasting at the category level may be preferable for some retailers, which can make comparing retail data challenging. Transforming the data provided by the retailer into a format that is clear and consistent to the supplier will allow businesses to make smart and informed decisions.

2.  Humanising data through collaboration 

If goods are not provided at the right time at the right place, suppliers can be penalised for failing to meet strict criteria. However, at times, this may not be their fault if the retailer forecast is last minute or inaccurate – and then both parties have an issue. Similarly, GSCOP rules dictate that the retailer must compensate the supplier for any cost incurred by the supplier, as a result of any forecasting error. As such, having a trusted and collaborative partnership between suppliers and retailers is mutually beneficial to both sides. 

Humanising and sharing the data provided via a trusted source of common data will allow suppliers to plan and make tactical decisions on orders and stock levels, which in turn, will create a more accurate forecast, leading to less waste for both parties and improved availability levels. 

3. Granular level data 

The purpose of a forecast is to help ensure suppliers meet order expectations, so it is crucial that they receive forecast order volume with a granular level of detail. By providing specific information by product, day and depot, suppliers can ensure that the appropriate inventory is in the right location at the right time. This allows orders to be fulfilled to high-service-level targets (normally 98% or above), rather than being an estimation.

4. Rapid data interpretation 

Suppliers with thousands of products being delivered to multiple depots each day rely on accurate and timely data, as well as proactive alerts to any changes in forecasts, so they can react quickly to challenges such as predicted stock-outs, despatch issues or poorly performing promotions. 

Daily data supplied by retailers lacks analysis, trends or alerts, so consequently many suppliers have created complex spreadsheets. But this can only fix the problem as a short-term solution, as complex spreadsheets are prone to error. Instead, utilising a system whereby state-of-the-art visual analytics are presented, allows key interventions to be made with confidence and ease by anyone in the business. This way, suppliers will be proactively alerted to changes in the forecast, so that opportunities are not missed and they are able to react quickly. 

Conclusion

To achieve successful demand forecasting, we need to move away from the supplier vs. retailer concept and instead encourage wider supplier and retailer collaboration. Full visibility of SKUs and regularly updated information will allow both parties to react in a quick and informed way to any changes in the forecast, with minimal disruption. Visual analytics approaches combined with accurate supply chain and sales data can help businesses gain insights for both current and historic sales patterns, revealing trends and where demand truly lies. In turn, this results in enhancing supply chain efficiency, reducing costs, ensuring regulatory compliance and happier customers in an increasingly demanding and competitive retail environment.

Credit: Ian Hall, COO, Atheon Analytics

Media Partner