Four steps to achieving successful demand forecasting


The Grocery Code Adjudicator states that suppliers to UK grocery retailers deserve to receive timely and accurate forecasts. Its GSCOP code of practice defines what is expected, and its latest annual report[1] states that all regulated retailers achieved compliance.

Despite this, forecasting remains one of the top three issues suppliers face. The report highlights suppliers reporting poor forecasts from retailers, significant variations between forecasts and orders, and penalties for failing to meet service levels.

In part, the problem lies in the lack of definition of ‘a forecast’; even GSCOP does not make clear what it means. Consider the following forecasts:

  •       An annual prediction of the volume of every product which will be purchased from a supplier (a joint business plan?)
  •       A 3-month prediction of weekly product sales in store (a seasonal forecast?)
  •       A 6-week aggregate weekly order prediction
  •       A 14-day daily prediction of short-term product sales, allowing for weather and promotions
  •       A 7-day demand forecast (not the same as a sales forecast, and more important to a supplier). Is this by depot?
  •       An order, for delivery tomorrow

Which of these is the ‘forecast’ that GSCOP requires retailers to share with suppliers? Which of these is the most useful to the supplier? Which should be measured against actual orders to determine ‘forecast accuracy’?

Furthermore, the GCA June 2018 publication states that:

“It was found that retailers adopted a range of approaches, and used the word “forecast” in a variety of ways. Some made a clear distinction between a forecast and an order; others did not see forecasting as a discrete activity but rather, as an integral part of supply chain management, often proceeding close to real time;”


So how can retailers and suppliers move forward?

The GCA’s Best Practice Guide [2] makes 17 recommendations to improve the current process. These include:

  •       Closer collaboration between retailers and their suppliers
  •       Regularly reviewing forecasting performance
  •       Ensuring that suppliers are able to get access to supply chain or buying teams to share intelligence and discuss forecasts or orders
  •       Ensuring that retailers have adequate systems and processes which learn from and take account of known or past issues
  •       Ensuring that suppliers are able to access adequate sales data

The 17 recommendations in the GCA Best Practice Guide are all achievable – only if both sides can make better use of available data, and are willing to share and discuss insights in an easily accessible way.

As such, there are four critical elements to accurate demand forecasting that will help retailers not only comply with GSCOP and the GCA recommendations but, as a result, help transform supply chain efficiency, availability and waste.

1.   Data at a granular level

A forecast’s purpose is to help ensure suppliers meet order expectations, therefore it is important they receive forecast order volume at a granular level of detail:

  •       By product (not by category, and not a total)
  •       By day (important for short-life goods and/or just-in-time logistics)
  •       By depot (essential for appropriate inventory in the right part of the country)

Without this, any other ‘forecast’ is at best a guide to what might be required and when, but won’t help the supplier ensure that appropriate inventory is in the right location at the right time so that orders can be fulfilled to high service-level targets (typically 98% or above).

2.  The ability to quickly interpret data

Some suppliers have hundreds (even thousands) of products going into multiple depots each day. Not only do they need accurate data, but they need to be proactively alerted to changes in forecast.

Analysing changes is time consuming and opportunities can easily be missed.  Instead, suppliers need to be able to react quickly to rectify predicted stock-outs, poorly performing promotions and despatch issues.

The data that is supplied by retailers on a daily basis does not include alerts, analysis or trends.  As a result, many suppliers have developed complex spreadsheets to extract the data and provide some analysis, but generally this relies on the knowledge of one or a few people at the supplier.  Those complex spreadsheets, by their very nature, are open to error.

Data warehouse or BI software provides an alternative to spreadsheets, but the views of the data have to be specified, built and maintained, and often the knowledge is – again – held by one or a select few people in the supplier’s business. A system in place whereby state-of -the art visual analytics are presented, daily for anyone in the business to see and interact with, allows key interventions to be made with confidence and ease.

3. Consistency of definition

Forecasting at SKU level by location is critical to most suppliers, but forecasting at the category level may be enough for some retailers.  Even those who forecast to SKU level often don’t distinguish between locations or even channels, so a forecast to a retail buyer may simply be a quantity over time, whereas a “good” forecast for suppliers is much more granular.

Retailers and suppliers may use different coding for the same SKU.  Suppliers may supply in different units of measure to that used at the retail end. (Cartons instead of units, for example). Suppliers and retailers could even give different names to the delivery depots and branches. All of which make comparing retailer data difficult – both across retailers, and also with your own internal metrics and reports.

It’s really important, therefore, to be able to transform the data provided by retailer into a format that is clear to the supplier.  It is only then that the business can make smart and informed decisions that will impact their business.

4.  The ability to challenge forecasts and collaborate

Suppliers can be subject to penalties for not providing the right goods at the right time to the right place.  But if the retailer forecast is so short-term or inaccurate that the supplier cannot react, then both parties have an issue. It is imperative that suppliers and retailers forge trusted partnerships which encourage mutually beneficial collaboration.

Similarly, under GSCOP rules, the “retailer must fully compensate the supplier for any cost incurred by the Supplier as a result of any forecasting error in relation to Grocery products”.


So, how should that collaboration take place?

In order for any discussions to take place between retailers and suppliers, there first has to be a trusted source of common data from which to work. In addition, that data should be “humanised” – in other words be presented in a format that is easily understood by both parties. Finally, there has to be a means of sharing that data online as the retail buyer and supplier are likely to be in different locations, which would ensure both parties are ‘singing from the same hymn sheet’.


With this shared view of the data, the supplier can help the retailer to:

  •       Make tactical proactive decisions and interventions on orders and stock levels
  •       Plan and optimise promotions
  •       Review forecasts based on accurate analysis


The net result of this can be seen as:

  •       Fewer stock-outs, improved sales for both parties
  •       More accurate forecasting leading to less waste for both (particularly fresh produce)
  •       Improved service levels and availability



When it comes to achieving successful demand forecasting, we need to eradicate the concept of the supplier vs the retailer. Taking a step back and looking at the bigger picture, both need to realise that it is only through fostering a mentality of genuine collaboration, that everyone can benefit. A platform which both supplier and retailer have full visibility of SKUs and access to key daily information with ease is a must. In doing so, genuine relationships can be built and changes in forecast reacted to with minimal disruption.

By using best-practice visual analytics tools and techniques, supply-chain and sales data can be blended to highlight indicators for demand (such as sales spikes, low-depot stocks etc.), and provide an easily understood picture of recent and historic sales patterns.

In an increasingly competitive retail environment, those retailers that can provide accurate forecasts to their suppliers, will benefit from more efficient supply chains, lower costs, better availability, and happier customers (and a happier Grocery Code Adjudicator).