Exploring the true cost of unsold food in retail

Food waste reduction is a priority for all food retailers, in all parts of the world. This study was commissioned to bring new insights as to the cost of managing unsold food in the store and later its exit from the business if it cannot be sold.

As the report highlights, there may be hidden costs to food retailers of an estimated 1.8% of revenue on the management of unsold food and its ethical exit from the business.

The report provides insight into different choices and strategies for food retailers to consider, and as such we believe makes a valuable contribution to the industry. We urge all retailers to reflect on the findings.

I would like to thank Professor Lisa Jack and her colleagues for carrying out this research and the retailers who helped contribute to the thinking and the findings presented in this report.

As with all the research undertaken on behalf of ECR Retail Loss, it would not be possible without the active support and involvement of the retail community and the many employees who generously gave their time to participate in the most meaningful of ways. Thank you for taking the time to share your thoughts and experiences – by working together we are much more likely to Sell More and Waste Less!

Finally, I encourage you to not only read and share this study, but also take part in the work of ECR Retail Loss.

Further details can be found at: www.ecrloss.com.

Our objectives for the project were to:

a. Map the exit routes for a sample of fresh, short expiry categories across a sample of food retailers in Europe, identifying the principle causes of food exit.

b. Identify the costs, both visible and hidden, of each of the different exit routes, from donation to people, through to land fill and everything in between.

c. Develop an interactive model that can calculate the true cost of food waste, at an item level, considering the costs to exit the product, and the potential costs of prevention, such as lost sales and reduced gross margin.

We carried out interviews and observations with four food retailers1, two in the UK and two in Europe. We attended seminars and discussion group meetings with ECR Retail Loss to strengthen the mapping process and the development of the true cost of unsold food model. We also carried out a detailed review of previous academic and industry projects, that supported our findings from the case studies on most wasted foods and exit routes.

Our Conclusions

We concluded that a full calculation of the true cost of food waste would include the cost to serve of items wasted or the profit foregone, in addition to the cost of handling food not sold at full price, the additional costs attached to different exit routes and consequential losses.

Detailed data was not available from retailers to enable this full calculation but we do demonstrate the following key points:

• There are hidden costs in the handling of unsold food that decrease the contribution to margin of a food item stocked. These hidden costs could be as much as 1.8% of sales revenue.

• Unrecorded/unexplained losses matter but there are costs/trade-offs associated with the time to record a missing item and to provide an explanation.

• Decisions to buy and stock inventory based on gross margin and contribution to overheads need to allow for the possibility that any item purchased might remain unsold or not sold at full cost. This is to account for the costs of having the system in place to identify and remove items from shelf.

The Key Learning Points

1. All areas of the business need to share in the responsibility to prevent both socially unacceptable food waste and mitigate risks to the bottom line. One solution would be to have cross-functional teams that include the CSR and loss prevention colleagues, but also data analysts, buyers, store managers, customer service and others.
2. Using this report, food retailers could develop external communications with consumers, suppliers and other stakeholders that set out the cost of handling surplus and unsold foods, and that they are willing to share the cost of meeting societies’ goals to reduce global food waste. This would help to correct widespread misconceptions or myths that it costs nothing to dispose of unsold food and that retailers sell less than the actual figure of around 97% of food stocked.
3. Food retail businesses can use the information in this report to make clearer decisions concerning the improvement of forecasting and inventory management, and about exit route choices. For example, our work indicates that donating food to staff carries less cost than sending unsold food to be converted into animal feed, subject to any scheme being used fairly across the business.
4. Our work also highlights the need to choose key performance indicators (KPIs) that do not create internal conflicts. Food waste KPIs or Total Loss KPIs in the Loss prevention Function should be aligned with KPIs to optimise gross profits in Product Procurement Functions, for example.
5. Importantly, our research supports the case to invest in sound record keeping procedures for unsold food in order to support better decision making. This is how retailers can separate known shrink (food to be `given to employees or donated) from unknown shrink (damaged, theft or unknown outcome). There are two key risks. The first is that to reduce the staff cost of handling unsold food, items may remain unreported or unexplained. This means that the cost of handling unsold food is underestimated, and the communication concerning social impact of the business’ action to reduce food waste is diluted. The second risk is that where tax incentives are available for redistribution of food for human consumption, items may be reported incorrectly, leading to overestimation of cost and social impact.
6. The driver for tackling food waste in recent years has been the environmental targets put in place on an international level. The UN Sustainable Development Goal 12, which seeks to ensure sustainable consumption and production patterns, contains the imperative in 12.3.1 to ‘reduce capita global food waste at the retail and consumer levels and reduce food losses along production and supply chains, including post-harvest losses’. Although most retailers are working towards the 50% reduction target by 2030 for in-store waste based on SDG 12.3, there is still work to do throughout the food supply chain.

Definitions used

We follow the food waste hierarchy proposed by the EU2 and classify unsold food into the following categories: Food surplus: Products that are near or past their expiration dates but still safe to consume, or excess food from overbuying, over forecasting, or overproduction.

We include:

• Unsold food sold at a mark down in store

• Unsold food sold in surplus bags/boxes (Sundry food items sold in bags or boxes at a nominal price per bag)

• Unsold food redistributed for human consumption directly to staff, or through charities and community groups

• Unsold food redistributed for animal feed

Food waste: The definition of food waste differs among retailers. Some consider only food going to landfills as waste, while others also include recycled or even donated food.

We include:

• Unsold food reused or recycled into other edible and saleable products such as beers created from surplus bread

• Unsold food recycled for use in production through anaerobic digestion or composting

• Unsold food sent for incineration - ‘Energy from Waste’

• Unsold food sent to landfill

Recorded Food Surplus/Waste is used to indicate the inventory items identified on the retailer’s system as being sent to an identified exit route. Unrecorded Food Surplus/Waste are items that are not identified as being sent to an identified exit route due to omission or inventory error, or because they have been stolen or mislaid.

A note on taxation:

In some countries there are tax incentives available to retailers who redistribute food for human consumption and prevent unsold food from going to landfill. Conversely, there are also administrative fines (for example, in France) for those not complying with the EU hierarchy or additional taxes on sending unsold food to landfill (for example, the UK ‘landfill tax’)5. Because of the variations across countries and the complexity of the rules, taxation has not been included in our calculations or discussion but could form the basis for a future study.

From the above quote, it appears that retail has the lesser problem with food waste. However, even 0.3 million tonnes from one country such as the UK (UK House of Lords, 2021) is a significant amount of waste. Following the European Commission’s Waste Directive (amended 2023, https://environment. ec.europa.eu/topics/waste-and-recycling/waste-framework-directive_en), retailers across Europe and the UK have engaged in initiatives to redistribute, recycle and repurpose food waste, rather than just send it, packaging and all, to landfill.

Among members of the ECR Food Waste working group, questions were asked about how much it costs retailers to deal with unsold food, and whether this cost had a significant impact on profits. To address these concerns, our first objective is to look at how unsold food is exited from the business. Our second objective is to explore the costs and losses associated with each exit route and our final objective is to develop a ‘what if’ analysis that can be used by retailers to assess the value of each route.

The project lasted from September 2020 to March 2025. The length of time taken reflects the lack of visibility of food waste data in retail businesses and the difficulty in compiling the evidence needed for the project. During this period we observed, through our case study companies and ECR Food Waste meetings, that the management of food waste in retail was changing rapidly. The main changes in this period are:

• Increased calls by civil society groups, governments and the food industry itself (including retailers) for mandatory reporting requirements8 and for consumer food waste prevention initiatives, such as the EU’s European Consumer Food Waste Forum (Project 2021- 2024)

• Moves from store manager discretion to centralised algorithms for markdowns and disposals, developed in conjunction with third party software providers10;

• Move from “zero to landfill” to “circular economy” principles. However, spectrum of practices globally covers from ‘everything to landfill, no markdowns’ to ‘full programme of varied exit routes’;

We also observed the complexity of managing food waste. Intuitively, giving or throwing food away may seem a quick and costless act. However, even in the home, time is expended in emptying bins, sorting food waste for composting or reuse of leftovers, transporting food items to food banks, and there is a cost to having additional bags and bins for those activities. Care has to be taken with best before and use by dates, and with food hygiene. Scale that up to the volumes of food and beverage items in a store, over 365 days a year, 24 hours a day and even tiny, marginal costs per item could end up having a significant impact on the bottom line profits.

We asked the following questions:

1. What does it cost retailers to exit food waste from their stores?
2. Could the cost of unsold food to retailers turn operating profits on lines into operating losses?
3. And when we know these costs, how could that data inform retailers strategies to manage the cost of unsold food?

Our objectives for the project were to:

• Map the exit routes for a sample of fresh, short expiry categories across a sample of retailers in Europe, identifying the principle causes of food exit.

• Identify the costs, both visible and hidden, of each of the different exit routes, from donation to people, through to land fill and everything in between.

• Develop an interactive model that can calculate the true cost of food waste, at an item level, considering the costs to exit the product, and the potential costs of prevention, such as lost sales and reduced gross margin.

We carried out interviews and observations with four food retailers, two in the UK and two in Europe. In each case, loss prevention and environment teams were implementing new strategies and processes to combat food waste. We were able to visit a city centre store of one of the case study companies and walk through the food waste management process. In addition, we had between 3-4 online discussions with each company team, including data analysts working with the teams. In order to collect numeric data we constructed a spreadsheet stating the data needed to develop the cost of food waste model. All the case study retailers provided internal and publicly available progress reports.

In two cases, we were given data relating to food waste. One of these sets of data related to an internal project on three specific items, looking at aggregate volumes and costs of waste across all stores for those SKUs. The other set of data related to daily figures for loss and waste collected by store managers. We were also given the outcome of a time and motion study carried out by one of the stores, to ascertain the length of time for one person to check one SKU on shelf in store. We also received information about work being done to reduce losses in supply chains and distribution centres but this report narrows the scope to in store unsold food.

The disparate information that we obtained enabled us to map the various exit routes of unsold food in each organisation and to develop insights into good practices and challenges facing the industry. However, it also became clear that each organisation was very different, and that in many cases, there were not yet clear mechanisms for collecting and analysing the data needed to develop the model in most retailers. One data analyst confirmed that data was held in different systems within the organisation, and others confirmed that ongoing IT developments (for example, to create centralised mark down algorithms) were in the process of tackling this issue. Therefore, in discussion with ECR Retail Loss, it was decided not to proceed with the planned questionnaires to members who were not part of the case studies.

Instead, we attended seminars and discussion group meetings with ECR Retail Loss to strengthen the mapping process and the development of the true cost of unsold food model. The model was presented to members of ECR Retail Loss on two occasions and feedback was used to refine the initial model, which is presented in this report. There is still scope to work with retailers who can further develop the model and report the findings back to the team, therefore not compromising sensitive commercial data.

We also carried out a detailed review of previous academic and industry projects, that supported our findings from the case studies on most wasted foods and exit routes. We were also able to find additional proxy measures that contributed to the development of the model.

Alongside this project, researchers at the University of Portsmouth (including Professor Jack) and Newcastle University have been investigating the way in which food waste is reported publicly through the Annual Reports, Sustainability Reports and Websites of major retailers in the UK and Europe. They have found that while all retailers aim to meet commitments to reduce food waste, there is significant variation in the definition of food waste, KPIs and mode of reporting. This has implications for the internal management, data collection and reporting for retailers, as ultimately the systems in place will need to feed into external reporting that may become mandatory within the next 5-10 years.

Since 2015, more retailers are measuring their efforts to reduce the food waste in store that they have some control over, to encourage consumers to waste less food and work with their supply chains to reduce food waste. They are under intense pressure from government, civil society groups and the public to do so but there is also a high degree of motivation from individuals inside retailing who feel driven to reduce food waste. There is an imperative in the industry to be seen to meet the UN SDG 12.3 goal of halving food waste by 2030. Whilst there could be accusations of greenwashing, virtue signalling or profiteering from waste, our interview evidence shows that the commitment by ECR Retail Loss members goes beyond loss prevention or profit enhancement, being seen as simply ‘the right thing to do’. At the same time, we find that being the ‘right thing to do’ requires significant investment in information systems technology and management time, and a willingness to take a hit on profits.

Although a few retailers have gone beyond what we have found in this project, others are still at the early stages of tracking food waste, and could use this model to help make the case for investment. The ideal is an integrated system in which all food waste is identified, weighed and costed, costs are imputed to individual SKUs, sales to customers are reduced and less food is stocked, incurring fewer storage and handling costs, and allowing for more reductions in the price of food.

Complications and caveats

Although some retailers were willing to share information, our study was still restricted by a lack of data. Even those further down the road were not able to collate all the data needed as it sat in different accounting system silos in the organisation. However, we were able to obtain sufficient data to construct a model that shows the cost of unsold food to an organisation, and to assess the scope for future developments.

Our three major findings were that:

1. The hidden costs of using alternative exit routes to landfill impose a significant fixed cost on retailers when they ‘do the right thing’.

1. Despite efforts to identify and report food surplus and food waste, a significant percentage of unsold food is unexplained and unreported. This unknown loss need to be included in calculations but it was unclear how retailers were accounting for unknown loss and there are perverse incentives for both including and excluding this figure from food surplus and food waste figures.
2. The true cost of food waste should take into account cost to serve and profits foregone but could still have unintended consequences for retail managers’ decision making.

What does it mean for organisations?

The irony, of course, is that retailers incur costs and consumers spend money on food that will be thrown away, at a cost, in the home. Because purchasing in retail is driven by sales, our interviewees were acutely aware that getting more accurate numbers on food waste would improve supply chain forecasting. In addition, setting the tone for consumers would also improve inventory and purchasing, reduce handling and storage costs to the benefit of everyone. However, the move towards centralised algorithms to drive markdowns and manage unsold food has also demonstrated that losses related to unsold food have a significant impact on profitability. Our model demonstrates this impact and highlights the realisation that food surplus and waste could be a ‘profit killer’ if not managed carefully.

What does it mean for consumers etc understanding the issue?

Communications around food surplus and waste need to be handled with care but there are some myths around unsold food that could be addressed. For example:

Myth 1: it costs nothing to give or throw something away. Counterintuitively, there is a cost to the retailer (or to any other party, including consumers) if they give or throw away food items for which they are paying.

Myth 2: that it does not matter if retailers overstock because they can make money from selling on food waste. We found that on the contrary, although there were a small number of opportunities to sell edible food onto recycling, in the majority of cases the retailer has to pay a third party to take the unsold food away. On the subject of overstocking, retailers forecast demand based on past sales to consumers and surprisingly, over 97% of all food stocked is sold.

The more difficult message to consumers is summed up in a quote from nearly 100 years ago.

In other words, you have to pay all the costs for the chickens you didn’t buy to be brought to the shelf, refrigerated there, checked, disposed of if unsold, and replenished.

The main message to consumers is that reducing food waste takes everyone in the food supply chain. Retailers do incur costs dealing with unsold food. Consumers may be sceptical of the Social and Environmental Reports that retailers produce but improved internal and external reporting is likely to make the retail message clearer and enhance the partnerships of retail, consumers, government and civil society groups to meet food waste reduction goals.

Food loss and waste reporting– a growing problem worldwide

Alongside this project, researchers at the University of Portsmouth and Newcastle University have been investigating the way in which food waste is reported publicly through

the Annual Reports, Sustainability Reports and Websites of major retailers in the UK and Europe. They have found that while all retailers aim to meet commitments to reduce food waste, there is significant variation in the definition of food waste, KPIs and mode of reporting.

This reflects the findings we found in our interviews, that measuring food loss and waste (FLW) is a complex issue. Agreed protocols may be in place but not the systems and metrics to enable full disclosure. In the absence of reporting regulations, it is difficult to assess practices adopted, metrics used, and improvements made. This means that there is a jointness problem between ESG objectives and accounting. In this study, we compare the current state of play for the ten largest food retail companies in UK and ten in Europe through an investigation of FLW disclosures over a five-year period. Findings suggest that there has been an acceleration in reporting in line with company policies and industry protocols. However, transparency is lacking. We observe varying disclosure levels and measures with some companies changing their targets, metrics and baselines. We also find evidence of inconsistencies and inaccuracies within statements.

In fact, we only found two UK companies offered like for like definitions and targets at any time, which makes comparing the progress of different retailers in reducing food very difficult.

To address the issue of food loss and waste in a systematic and comprehensive way, there is a need for guidelines to increase transparency and consistency, and an assessment of drivers of change to establish whether firms are motivated by ESG or net margin objectives. In 2024, UK retailers and other organisations called on the British government to introduce mandatory reporting requirements for food waste reduction in the industry.

Mapping the exit routes and associated costs

From our interviews, the following were identified as the most wasted items in store. Previous academic studies and industry reports have found very similar lists.

• Fresh baked on site bread (highest by weight)

• Fresh produce

• Fruit and veg, and particularly bananas, soft fruits and salads

• Meat and fish (highest by value)

• Concessions and in store deli counters (for example, for fish, cooked meats or sushi) create waste that is dealt with by the store rather than the concession holder.

A detailed companion study has been conducted for ECR Retail Loss entitled Food Surplus Management in Retailing: A Global Perspective which provides detailed background for the qualitative aspects of the management of unsold food.

Note: this table and subsequent analyses do not include repurposing, upcycling or recycling of surplus food into new products for human consumption. There were too many different examples and permutations of these schemes to allow for a proxy mapping or figure to be produced. However, such schemes appear to represent a very small percentage of the food surplus and waste currently handled by retailers, although this is likely to grow given the success of some of the new products developed in partnership with manufacturing suppliers.

From our interviews, observational store visits, review of data provided and attendance at ECR Food Loss meetings, the following emerged as the costs associated with managing unsold food. However, it is also very clear that there are many permutations and variations on how unsold food is handled, the nature of the item (fresh produce, meat, bakery for example), which exit routes are used, the costs incurred and the third parties involved. This means that it would be very difficult to create one model that could reflect all possible outcomes related to the true cost of food waste. Therefore, we have set out an indicative model that highlights the elements that need to be considered and from that, suggest some heuristics that should be taken into account during a retailer’s buying, forecasting and planning processes.

Our findings fall into three categories, reflecting the factors that can damage an item’s contribution to margin and that could turn the profit envisaged in the buying process into a loss for the retailer overall. The three categories are:

• The Hidden Costs of Unsold Food

• Unexplained total loss related to unrecorded items of unsold food

• The potential impact of using cost to serve rather than purchase cost to assess the true cost of unsold food.

The Hidden Costs of Unsold Food

People costs

All our interviewees were clear that the most important cost was the time spent by staff on checking items on shelf, marking down or removing for different exit routes.

One case retailer has carried out a timed study of selected chilled and fresh produce items, utilising a new app based on an algorithm. They estimate that each check of a block of items on a shelf or in a cabinet takes on average 1.9 minutes.

Unsurprisingly, many retailers are experimenting with algorithms that optimise the number of times each SKU needs to be checked, (for example, algorithms designed to alert staff to near out of date items only), and the routes through store. For example, one European retailer has a system of markdowns based on reductions through the day automatically applied at the till and handled through the EPOS/electronic shelf tags.

A surprise contentious element of the time spent checking unsold food was the time to record the exit route of near or out of date items. Our case study participants, supported by feedback from ECR Food Waste members, were aware that to complete tasks set in an allotted time or shift, staff omitted the more complex recording of, for example, charity donations or missing items that needed to be tracked down in other parts of the store. This is particularly the case where information needed to be transferred from a manual record used in the aisles to a computer system held in the back of the store. Our model includes the time to record but the implications of the task of recording being omitted is discussed further in the section ‘unreported and unexplained losses’ below.

More detailed calculations based on standardised industry data was provided for us by West Monroe Partners and our case study participants. The costs are based on a rate of 25 cents per minute, as above.

All the Labour Times and Costs per item include time to record of 0.02 minutes (€0.005).

Figure 2 illustrates the considerable time and distance expended in checking, collecting, collating,

Figure 2: A schematic diagram of the checking process for an SKU.

Consumables

Where items are being separated out for recycling, redistribution and so on, colour coded bins and bags are often used. Tags, bags and other consumables are used for markdowns and damaged items.

There is a cost to buying these items in bulk but the cost per item is negligible. For the purpose of the model, it is assumed that the cost of consumables is within the rounded up figures for the cost of time, above. On a store-wide database the cost of consumables could be added to the final cost of food waste

Costs in addition to time spent and consumables

Surplus bags/boxes and similar schemes

These schemes involve packaging sundry items of surplus food and offering them for sale in store via an App or customer messaging. They can be quite costly for a retailer to manage. The items need to be bagged and recorded, staff time is needed to oversee the collection and if not taken, the items need to be unbagged and sent down another exit route. Third party companies providing Apps charge a small annual fee per year per store (typically around €50), and a commission for every bag reserved through their App. The attraction of the scheme for retailers is that it is said to bring in customers who may not otherwise have visited the store, and the store receives some income from the sale although this may not actually result in a profitable sale. The main value of the externally managed App lies in a possible intangible reputation and marketing boost for the store. Internally managed similar schemes still incur the costs of time to bag and time to re-sort if not sold but do not incur the external fees or commissions

There is a high variability in the additional cost offset against the additional income from the scheme, and for the purposes of the indicative model here, we use only the people and consumable costs as above, and assume that management time is included in the management cost figure below.

Charitable redistribution services

There are several possibilities here, ranging from the collection of individual bags or crates of items from a store by a local charity on a regular basis, through to a fully managed service from a provider where items are collected from a more centralised location and then redistributed. The transport costs from the distribution centre are paid by the charity. The main cost in store is the time and consumables, as above. However, if the items are transferred to a central location, and collected, then the fee paid to the service provider and the costs of transport to the distribution centre are assumed here to be the same as for animal feed, anaerobic digestion, incineration and landfill below.

Fees paid for animal feed, anaerobic digestion, incineration and landfill

When retailers dispose of unsold food items by these routes, costs tend to be based on tonnage. In store, the items have to be bagged or sorted into bins. This sorting could be done on the trolley and then transported to the back of store area, or transported and then sorted at the back of store. Separate areas need to be maintained to store the bins/bags until they can be collected by one third party operator who delivers the bins to the different processing plants or by a different operator for each exit route. However, it is more likely to be the case that the bins are consolidated at the retailers own distribution centres or other location, and collected from there by the third party.

An estimated cost per tonne (gate fee) would be €60 per tonne or €0.06 per kilogramme.

In addition, there could be a transport cost for removing the bins to the retailer’s regional central location. Hygiene considerations mean that the bins cannot simply be put on lorries returning to the distribution centre from stores. If this is done, then a further cost is incurred for cleaning.

The RHA in the UK use €1.80 per mile for a 7.5 tonne lorry (a load of 2900 kilogrammes) as a rule of thumb in calculating costs. €0.03 per kilogramme for a journey of 50 kilometres.

Another additional cost may be incurred in countries where anaerobic digestion, incineration and landfill incur a fine or tax cost. This is discussed below under the heading of ‘Unexplained total loss’.

Furthermore, there is the risk that food sent for animal feed may be contaminated (for example, bakery may inadvertently contain meat from sandwiches included in the wrong waste container) and then additional costs may be incurred in either the retailer or contractor resorting and resending the food down an alternative route. However, this complexity is not included in our calculations.

Recycling/upcycling

This is a complex area with many different schemes ranging from reusing food instore to make soups for staff or customers, through to working with a third party manufacturer to produce own brand products or the third party’s branded products that may then be sold in store. There are additional costs associated with Hazard Analysis and Critical Control Points (HACCP) and food inspection protocols and with sorting foods into hygienic storage areas.

For the purposes of our indicative model, we are assuming that the items are collated in store, transported to a central facility and collected by the third party manufacturer.

The additional costs here would be the €0.03 per kilogramme for removing the items to a more central location and a notional €0.005 for additional hygiene and food safety costs incurred. It might also be the case that the retailer receives a nominal income for the waste product, although we were not given any data that enabled us to estimate the value of any such income.

Management costs

Retailers assign loss prevention and data analysts to the control of food waste, along with accounting staff, operations managers and IT support. This time is a fixed cost to the organisation but represents a significant investment by the retailer in dealing with unsold food. There are also the costs associated with the time the store manager spends dealing with food waste task supervision, reporting and training. In addition, there are also the costs associated with certification schemes, inspections and QA management systems.

In addition, training costs need to be taken into account, particularly in stores with high staff turnover. Complexity and integration of systems and processes, including a move away from store manager’s discretion in handling markdowns, surplus and waste, to centralised algorithms. The new systems work better when comprehensive staff training and communications are put in place.

It is very difficult to assign a standard cost to management time per SKU in a store. We assume a cost €0.001 per item, intended to be indicative of the fact that every item stocked in store carries a management cost overhead.

Opportunity costs

We considered whether opportunity cost for staff and management time should be included in the calculations. The argument is that if staff did not spend time on processing and managing waste, then they could engage in other activities. However, it is more likely that rather than alternative tasks being assigned, fewer staff would be employed. Hence, a straight saving would be made. Insufficient data was available for loss prevention alternative activities, and as this was a small fraction of the overall cost, opportunity costs have not been included in this analysis

Unexplained total loss

Not all unsold food could be accounted for by the redistribution of food surplus or food waste. Some formed part of the unexplained total loss in food and drink. One explanation is that staff do not have the time or incentive to investigate and accurately record all unsold items, if they have gone missing from the shelf. This figure tends to be written off as an expense by the store, rather than included in the food surplus/food waste figures. One perverse outcome of the need to record all food waste could be that because of the time it takes to classify and record items that are damaged, spoiled or beyond their expiration date, items are left unrecorded. Also, the tasks of classification and recording require trained staff, who may not be available.

However, where tax incentives are given for food donations, for example in France and Germany, unexplained losses may be lower as still edible items are more likely to be recorded but this still leaves a significant number of items that cannot be donated and which may be unrecorded. Tax incentives can also be perverse, in that items may be recorded as donated when damaged or spoiled. Tax incentives for donations have not been included in this model, as they vary across countries. However, it is relatively simple to add them back into the calculation where needed.

There is a risk that with different charity schemes being handled in store, alongside surplus bags and staff schemes, that the records of which items go to which exit route may be conflated. This is particularly a risk where there are incentives in the form of tax rebates or staff merit point, say, which are a temptation to record more as a charitable donation. The risk also exists simply because colleagues are working at speed to meet time targets.

The main problem is the lack of data relating to the ratio of unreported/explained loss to unreported/ unexplained loss relating to fresh produce. We found or received a very mixed set of numbers from case study companies, reports and verbal feedback.

In our data, the best case scenario was for every 3 items recorded with an explanation as being food surplus or food waste, 1 was unrecorded or unexplained. In our worst case scenario, 1 item was recorded with an explanation of the exit route, but 2 were unrecorded or unexplained. The problem is that the loss may only be reported at purchase cost with any additional costs being difficult to trace. The items may have been exited as food surplus or loss, or may have been stolen or be part of an inventory error. If staff colleagues are being monitored on speed, there may be a perverse incentive not to record involved. Better record keeping would enable more precise data to measure costs related to unsold food and give a better estimate of the hidden costs of unsold food.

In the Scenarios section, we have taken the FMI average total losses per store of 3%16 and a worse case reported/unreported ratio of 1:2. This simply reflects that the permutations are almost endless and that a conservative accounting approach has been taken based on the very limited evidence made available to us.

Although all retailers tracked the losses of food items through their inventory systems, there was a lack of data concerning the additional costs. Overall information for reporting purposes is usually provided by weight. We found less evidence of the existence of data analysis on trends or regular reconciliation to reduce the unexplained total loss across the retailers.

Calculating Cost of Sales and the need to avoid silos

Purchase costs

Where food waste and donations are stated as a monetary value, rather than in tonnes, the loss is usually measured as the purchase cost (unless on a sale or return contract, when the loss is often shown as nil). When items are discounted to below their purchase cost, then there is an additional loss of gross margin. This raises an important issue in respect of forecasting, inventory and buying patterns. Buyers may be liable to assume, and the systems that they use may be programmed to assume, that all items except a nominal total loss for wastage, are going to be sold at full price. This means that the gross margin calculations and contribution to overheads used to calculate the profitability of an SKU will be inaccurate. A further issue is that the stores will be penalised for the markdown and unsold items, rather than the central buyers and inventory managers who made the decision to buy.

If purchase cost is used, and items have been sold at a mark down or not sold at all, then arguably, a true cost of food waste should also take into consideration the profit foregone. However, this is likely to be evident in the difference between the Buyers Ideal Scenario and the actual profit or contribution achieved from sales.

Cost to Serve

Ideally, all the costs of bringing the item to the shelf, storage and handling should be included in an ideal calculation, as that effort is also lost when food is wasted. However, tracking these overheads and imputing them to SKUs or food waste is very difficult. A separate exercise involving floor space saved and other cost-benefit exercises would be needed.

We can estimate the cost to serve from the annual financial statements of retailers. A simplistic explanation is that a food retailer will state the cost of sales to be all the direct costs of selling food including all store overheads. The Gross Margin % then acts as an indicator of Cost to Serve. Operating margin the gross margin less the administrative costs of the head office and net margin is operating margin less tax and financing costs.

However, there are variations in how retailers in different countries classify their direct costs. example, Sainsbury plc in the UK has one figure:

“Cost of sales consists of all costs that are directly attributable to the point of sale including warehouse, transportation costs and all the costs of operating retail outlets.” (Annual Report 2024 p.140 Note 3.2)

However, Walmart plc splits operating costs into Cost of Sales and Operating, Selling, General and Administrative Costs, making it more difficult to find a comparative figure.

In 2024, across six companies within ECR Retail Loss, Operating Profits were in the range 0.4%-4.2% and Net Profits were in the range 0.28%-2.4%.

For those companies disclosing all non-head office costs as one figure, it seems that the cost of sales including all directly attributable costs is around 95%, given a Gross Profit % of 3%-7%. Therefore, as a worst-case scenario, we have taken Cost to Serve as 95% of the sales price of an item.

Cost to replenish

As shown in Figure 2 above, there is an additional cost to replenish shelves once items are exited from sale or have gone missing and are unreported.

We have used a nominal figure of €0.025 in line with the cost to inspect in our model (Appendix B)

Investment in IT and hardware

All the retailers that we spoke with have invested heavily in systems to manage unsold food. However, the cost and depreciation of the systems and hardware have not been considered here, as the data was not available. This raises the question whether the decisions to invest are based on a cross- functional understanding of the impacts of the cost of unsold food.

Summary

A detailed breakdown of the costs per exit route can be found in Appendix C. In order to appreciate the level of costs associated with each exit route, Table 3 is a reconstruction of Table 1 above, with the identified costs and losses for 1,000 units held, assuming all 1000 items are exited through that one route in a column. Ticks imply that cost could not be identified or are costs associated with each individual SKU.

The unsold food model (See Appendices for detail) works on the basis that food might be checked more than once before being sold, marked down, redistributed perhaps via a staff shop or charity, or disposed of by another route. This is mapped out in Figure 2 which can be found in the previous section.

The ‘What If’ model recognises that the cost of unsold food effects contribution to margin and this might lead to losses that cancel out the profits made from food sold.

We created a hypothetical case study, because none of the case organisations could supply all the data that we needed. This was largely due to the fragmentary nature of their systems, and the lack of coherent data that we have discussed above. We used figures provided from case studies, and we supplemented these figures with benchmarks taken from reliable sources.

From the indicative model, we suggest that the true cost of unsold food is the cost associated with the possibility of any item purchased being marked down or exited from the business as food surplus (redistribution) or food waste. The whole system of checks, marking down, bagging up and so on, exists for the possibility that any unit on the inventory might not be sold at full price.

The basic model follows the flow chart in Figure 3 and the equation in Figure 4.

A glossary and examples can be found in the Appendices.

Based on our case study, companies and industry data, we can estimate that the purchase cost of food donated, re-distributed or thrown away equates to 1% of the retailers’ sales revenue. This is typically the number that would appear on the retailers accounting records as the cost of food surplus and waste.

However, the retailer also needs to consider the possibility that around another 1.8% of the sales revenue will be needed to cover the hidden costs of managing ALL FOOD NOT SOLD AT FULL PRICE. That’s food surplus and waste but also markdowns and unexplained losses.

Breaking down this 1.8%, about three-quarters of these hidden costs relate to checks and markdowns (1.35%). About a fifth relates to the possible cost of identifying and managing mislaid and stolen items (0.37%). The remaining hidden costs are the additional costs of giving food away or sending it (for example) to anaerobic digestion (0.06%) and management costs (0.01%). This is shown in the pie chart in Figure 5.

Rearranging the figures shows how much of the hidden costs are related to just food surplus and waste, and how much to maintaining the system that identifies near or out of date, and missing, items in store. This breakdown is shown in Table 4 and additional calculations can be found in Appendix D.

Managers who need to get the message across that dealing with unsold food costs the business more than expected, could draw on the idea of the iceberg in Figure 5.

While these scenarios and the model need to be tested and refined using a much larger dataset (not yet available) over many different items, they are indicative of the high cost of unsold food to a retailer.

Figure 5: hidden costs of unsold food

Scenario 1: A Buyer’s (perfect) Calculation

1000 units of a chilled product are purchased at a cost of €3.25 per unit and all 1000 units are sold at the full price of €5.00 per unit. The gross profit margin is 35%

Note: This is termed the ‘perfect’ calculation as in reality, buyers usually do build in a margin for waste when making procurement decisions. It is also acknowledged that buyers need to make decisions between waste avoidance and on shelf availability. However, the aim here is to show how that margin should be assessed and calculated from first principles.

Scenario 2: A buyers (perfect) Calculation but now including the cost of quality / expiry date checks

Again, all items are sold at full price but each item is checked at least twice for damage, out of date or other issue. This gives an additional cost to be considered of €51.00. The gross profit margin reduces to 34%

Scenario 3: Now including the costs of selling some items are sold at a markdown price

Out of the 1000 units purchased, 800 (80%) are sold at the full price of €5.00 per unit and 200 (20%) are sold at a 35% mark down price discount of €3.25 per unit. That is, a mark down price which is equal to the purchase price (65% of the sales price). Each item is checked at least twice. The gross profit margin reduces to 29%.

If the items are sold at below purchase price, for example at 50% of the sales price, the impact of the loss on the purchase price would decrease the gross margin further, to 24%

Scenario 4: As well as markdowns, some items remain unsold (food surplus/waste)

Out of the 1000 units purchased, 800 (80%) are sold at the full prices of €5.00 and 190 (19%) are sold at a 35% mark down price of €3.25 per unit. 10 units (1%) are unsold. Of these 10 units, 6 (0.6%) are redistributed as food surplus and 4 (0.4%) are sent to incineration as food waste. The gross profit margin reduces further to 28%

Scenario 5: Markdowns, food surplus and unexplained losses

Out of the 1000 units purchased, 800 (80%) are sold at the full prices of €5.00 and 170 (17%) are sold at a mark down price of €3.25 per unit. 30 units (3%) are unsold. Of these 10 units, 6 units (0.6%) are redistributed as food surplus and 4 units (0.4%) are sent to incineration as food waste. 20 units (2%) are unexplained and unreported losses. The gross profit margin reduces further to 27%.

The Worst Case Scenario

Scenarios 1-5 use the purchase cost of items to calculate the cost of unsold food. In this worst case scenario, we make the hypothetical assumption that the cost to serve is used in place of the purchase cost. This will give an end-to-end profit and loss view of food waste and indicates that overheads incurred in assuring on shelf availability should be managed alongside the prevention of food waste.

For this worst case scenario, we use the following:

Out of the 1000 units purchased, 800 (80%) are sold at the full prices of €5.00 and 170 (17%) are sold at a mark down price of €3.25 per unit. 30 units (3%) are unsold. Of these 10 units, 6 units (0.6%) are redistributed as food surplus and 4 units (0.4%) are sent to incineration as food waste. 20 units (2%) are unexplained and unreported losses.

However, instead of using purchase cost to calculate the gross margin, the cost to serve is used. In this example, the Cost to Serve is estimated as €4.75 (95% of the sales price – see Pages 23-24).

The gross loss margin is -4%

A less drastic view of the situation that still gives a fuller picture of the true cost of unsold food is to consider the profit foregone on unsold food items.

The following table 5 summarises the cost of unsold food and gross margins, if profits foregone are included as well as markdowns, food surplus, food waste and unexplained losses. It shows the changes in basis points (bps) from a baseline of a 35% profit margin (the buyers ‘perfect’ scenario).

The NET margin of an item sold at €5.00 and purchased for €3.25 is likely to be 1-2% of the sales price, after overheads are taken into consideration. Therefore, if the gross margin falls below, say, 33.5%, then it is unlikely that the item will make a contribution to net profits. If the gross margin falls below zero, then the item purchased has a high probability of contributing towards a loss. In an assortment of items, it is likely that the net effect of these positive and negative contributions will be smoothed out, but the overall profits of the retailer will be reduced as a consequence of dealing ethically with unsold food.

We found that the probable cost of unsold food is €0.027 for every €1.00 purchased or 2.7%. This equates to €0.018 for every €1.00 sale price or 1.8%18. Therefore, for food items with a NET margin of less than 1.8%, it is likely that the cost of handling unsold food could eliminate any profit on an SKU. Therefore, when making buying decisions, we recommend that at least 2.7% of the purchase price is built into the anticipated profit margin.

SO WHAT IS THE TRUE COST OF UNSOLD FOOD?

A full calculation of the true cost of food waste would include the cost to serve of items wasted in addition to the cost of handling food not sold at full price, the additional costs attached to different exit routes and consequential losses.

Detailed data was not available from retailers to enable this full calculation but we do demonstrate the following key points:

• There are hidden costs in the handling of unsold food that decrease the contribution to margin of a food item stocked.

• Unrecorded/unexplained losses matter but there are costs/trade-offs associated with the time to record an item and to provide an explanation.

• Decisions to buy and stock inventory based on gross margin and contribution to overheads need to allow for the possibility that any item purchased might remain unsold.

What have we learned about different exit routes?

What do our scenarios tell us about which costs can be managed/mitigated?

The cost of markdowns should be managed, and the majority of retailers are now using centralised algorithms linked to EPOS to find the most cost-effective means of achieving markdown efficiency. This is achieved through reducing the number of checks required per SKU, which would have a significant impact on the cost of unsold food.

It appears to be marginally more cost effective to choose food redistribution over food waste exits for unsold food, particularly where this can be done instore. However, in order to comply with food hygiene regulations, this is not always possible.

Implications for the management of unsold food

As with our previous report for ECR Retail Loss on ‘Buy online, Return in Store’, the demarcation between Loss prevention, Buying, Inventory and Storage, Finance and other divisions within a retail organisation can prevent a co-ordinated response through the threat to the bottom line of unsold food. However, it should also be acknowledged that retail managers across functions are also consumers, and we find that they are often genuinely exercised about the ethical dimensions of food waste and the cost of doing the right thing. Cross functional teams would align both the ‘moral imperative’ not to waste food, and the data analysis related to the true cost of unsold food and its impact on the bottom line. Ideally, better forecasting, buying and inventory management could alleviate prices for consumers, and reduce losses and overhead costs for the retailer.

There needs to be alignment throughout the organisation to prevent both socially unacceptable food waste and mitigate risks to the bottom line. One solution would be to have cross-functional teams that include the CSR and loss prevention colleagues, but also data analysts, buyers, store managers, customer service and others.

Buying decisions need to recognise the potential level of mark downs and the cost to serve. This would, in turn, feed into the forecasting models which could eventually allow for reduced prices to customers and higher staff wages, for example. Put simply, there is a significant cost to getting the forecasting wrong. The SKU could very quickly become loss-making and have a significant impact on the contribution to cost to serve and on the bottom line.

The main message to consumers is that reducing food waste takes everyone in the food supply chain. Retailers do incur costs dealing with unsold food. Consumers and activists may be sceptical of the Social and Environmental Reports that retailers produce but mandatory reporting is likely to improve the value of measures and initiatives reported. There is a risk though that retailers would not do more than meet mandated requirements and so lose their incentive to innovate in this field.

This comes back to the message that all stakeholders need to be aligned in reducing food waste. The Cost to Serve is vital in that message and a further development of this research project would be to attempt to collect data relating to these additional costs that are incurred on unsold food.

Our report can be summarised in these six key messages for senior managers in retail:

1. All areas of the business need to share in the responsibility to prevent both socially unacceptable food waste and mitigate risks to the bottom line. One solution would be to have cross-functional teams that include the CSR and loss prevention colleagues, but also data analysts, buyers, store managers, customer service and others.
2. Using this report, food retailers could develop external communications with consumers, suppliers and other stakeholders that set out the cost of handling surplus and unsold foods, and that they are willing to share the cost of meeting societies’ goals to reduce global food waste. This would help to correct widespread misconceptions or myths that it costs nothing to dispose of unsold food and that retailers sell less than the actual figure of around 97% of food stocked.
3. Food retail businesses can use the information in this report to make clearer decisions concerning the improvement of forecasting and inventory management, and about exit route choices. For example, our work indicates that donating food to staff carries less cost than sending unsold food to be converted into animal feed, subject to any scheme being used fairly across the business.
4. Our work also highlights the need to choose key performance indicators (KPIs) that do not create internal conflicts. Food waste KPIs or Total Loss KPIs in the Loss prevention Function should be aligned with KPIs to optimise gross profits in Product Procurement Functions, for example.
5. Importantly, our research supports the case to invest in sound record keeping procedures for unsold food in order to support better decision making. This is how retailers can separate known shrink (food to be `given to employees or donated) from unknown shrink (damaged, theft or unknown outcome).There are two key risks. The first is that to reduce the staff cost of handling unsold food, items may remain unreported or unexplained. This means that the cost of handling unsold food is underestimated, and the communication concerning social impact of the business’ action to reduce food waste is diluted. The second risk is that where tax incentives are available for redistribution of food for human consumption, items may be reported incorrectly, leading to overestimation of cost and social impact.
6. The driver for tackling food waste in recent years has been the environmental targets put in place on an international level. The UN Sustainable Development Goal 12, which seeks to ensure sustainable consumption and production patterns across the globe, contains the imperative in 12.3.1 to ‘reduce capita global food waste at the retail and consumer levels and reduce food losses along production and supply chains, including post-harvest losses’. Although most retailers are working towards the 50% reduction target by 2030 for in-store waste based on SDG 12.3, there is still work to do throughout the food supply chain.

Next Steps:

We hope that you have found this report useful and informative, and can put into action one or more of the six key messages. Due to restrictions on available data, our findings are limited in scope and are designed to provide an indicative guide to the cost of managing unsold food in a retail environment. We would very much welcome feedback on this report. Also, we would like to work with retailers and service providers to refine the model and the data used. Please contact ECR Retail Loss to discuss how we can share the model with you, in return for feedback and the summary results from your own data analysis to colin@ecrloss.com.

Table 8 shows the costs of staff time and additional costs for each exit route discussed above.

Calculations from previous research

From an academic perspective, there is only one previous project that appears to have addressed a similar problem to the one that we examine in this project.19

Their calculation is that the

Economic cost of food waste. =. (purchase) cost of wasted product + personnel cost for waste

management + cost of waste disposal

Mattson, Williams and Berghel found the proportions of these costs to be 85%:9%:6% or 85%:15% (or roughly 6:1) for purchase cost to cost of food waste.

Using their logic, this means that for every €1.00 of purchase cost spent on an item that is then becomes food waste, the additional cost of that item that is thrown away is €0.17.

Our model goes beyond this to bring in mark downs, unexplained and unreported losses and cost to serve. However, we do not have the large data set that the Williams and Buerghel (2018) study was able to develop with one retailer.

We created a spreadsheet based on the Cost of Returns model developed for ECR Retail Loss in 2018, which in turn was based on equations by Asdecker (2015). The original model for returns built in the notion that goods might be returned and resold, then returned again before disposal.

Using the limited scenarios in the next section, we find that for every €1.00 spent on purchasing, around €0.027 needs to be allowed for the possibility that the item might not be sold, based on purchase cost. (See Appendix A p.36.) For our scenario, this means that the probable cost of food waste is €87.75 for 1000 items at a purchase cost of €3.25 and a sales price of €5.00. This can be reduced to €27 for every €1000 of items purchased or €18 for every €1000 of sales. This gives a ratio of roughly 6:2 (3:1) rather than 6:1 in the Mattson et al project above.

We compared this figure to estimates produced by one of our case companies, as shown below:

They added together a purchase cost of food wasted (not for human consumption) and the cost of staff time to handle food waste:

€millions

Purchase cost of food waste  825

Staff costs 160

Total cost 985

Using the Mateson et al. calculations above, this gives us a cost per euro of €0.023 or €23 for every €1000 of items purchased (2.3%), or €15 for every €1000 sales (1.5%) based on a blanket mark up of 35%. This gives a ration of around 5:1 or for every €1.00 of purchase cost spent on an item that is then becomes food waste, the additional cost of that item that is thrown away is €0.20.

The research for this report was supported by ECR Retail Loss. This report is intended for general information only; it is based upon a review of the available literature together with primary research undertaken with retail organisations in Europe. Individuals or companies are advised to take professional advice regarding their specific needs and requirements prior to taking any actions resulting from anything contained in this report. Any such actions taken by individuals or companies are entirely at their own risk. Companies are also responsible for assuring themselves that they comply with all relevant laws and regulations, including those relating to intellectual property rights, data protection and competition laws or regulations. The images used in this document do not necessarily reflect the companies taking part in this research.

© May 2025, all rights reserved.

About the Authors

Professor Lisa Jack, FCCA, began her career as an auditor, before moving into teaching professional accountancy and management studies. Lisa moved to Portsmouth in 2009 where she has specialised in research in management control and accounting for the food industry and forensic accounting.

Professor Chris Simms is an expert Innovation Management and Entrepreneurship with a focus on NPD and packaging, particularly in the food industry. He has undertaken collaborative work with a number of organisations within the public and private sectors, with extensive experience working on Knowledge Transfer Partnership projects.

Professor PK Senyo specialises in FinTech and Information Systems in the Department of Decision Analytics and Risk at Southampton Business School. His research focuses on how the use and adaptation of new digital technologies impact individuals, organisations, and society. His current research interests include financial technologies (FinTech), Financial Inclusion, Platform Ecosystems, Artificial Intelligence, Blockchain, ICT for Development (ICT4D), and Digital Innovation.

The authors are very grateful to the four organisations who participated as case studies for this research project, as well as all the other retailers who took part in interviews and who offered feedback and additional insights via events within ECR Retail Loss.

The authors can be contacted through Professor Lisa Jack at Lisa.Jack@port.ac.uk

Additional work acknowledged

Professor Josie McClaren (University of Newcastle), Dr Raina Wu, Dr Roza Sagitova, and Mr Steven Iorfa and Ms Dina El Tabey (University of Portsmouth).

About ECR Retail Loss

ECR Retail Loss is part of ECR Community, a voluntary and collaborative retailer-manufacturer platform with a mission to ‘fulfil consumer wishes better, faster and at less cost’. Over the last 25 years, ECR Retail Loss has acted as an independent think tank focused on creating imaginative new ways to better manage the problems of loss and on-shelf availability across the retail industry. Championing the idea of Sell More and Lose Less, it is open to any retailer and manufacturer to join. Its work is supported by research funding provided by Checkpoint Systems, NCR, Genetec, Retail Insight, RGIS and Upshop.

For further information: www.ecrloss.com.