Inventory is one of the retailers’ most significant assets, so it should not be a surprise the number of resources dedicated and technology developed for its optimization, such as software, artificial intelligence, and machine learning tools for demand forecasting and supply chain optimization.

However, despite all these efforts and the growing sophistication of inventory management systems, there is still an Achilles heel that no retailer has been able to overcome: phantom inventory.

Phantom inventory refers to goods that appear in the inventory system but are not available for sale, either because they are lost in the back room, out of display in the store, or simply no longer physically exist.

Phantom inventory is exceptionally costly and significantly affects retailers’ bottom line, as it essentially prevents products from being available to customers. But that’s not all. Phantom inventory also hurts inventory management, demand forecasting, online sales, analysis, and decision-making capabilities, to name a few. This blog will present the consequences of having phantom inventory problems for retailers.

Loss of sale

Loss of sales is the most obvious effect of phantom inventory since you cannot sell a product that is not physically available to the customer.
Phantom inventory generates a decrease in shelf availability, making it impossible for customers to find the products they are looking for. Studies show that when a customer faces a stock-out, they may decide to substitute a similar item, switch brands, choose to shop at another store or abandon the purchase. Regardless of the decision, the retailer will be affected, receiving 2-3% direct sales losses depending on the category.

Inventory management

As we have seen, one of the origins of phantom inventory is the discrepancy between physical and system inventory levels. Let’s consider that according to the GS1 US Survey, the average retailer in the United States has an inventory pressure of only 63%. The chances of having phantom inventory are pretty high.
Beyond accuracy, phantom inventory generates a chain reaction, disrupting processes that originate from recorded data, such as automatic purchases based on reorder points. An imbalance between the two inventories (systemic and physical) can generate stock-outs when the lack of stock in the store is identified too late; or overstocking, in case the system has registered a lower quantity than the actual one.

Non-compliance in online sales:

The adoption of e-commerce is a reality that has accelerated in recent years, putting pressure on retailers facing the highest levels of competition.
Discrepancies between the systems and physical stock create havoc in meeting online demand.
According to Peoplevox’s E-Commerce Fulfilment Report, 34% of retailers have delayed the shipment of a purchase order because they mistakenly sold a product that was not in stock. Phantom inventory leads to accepting purchase orders that cannot be fulfilled because the inventory is not actually available in the store. If you don’t pay attention, this problem will only be identified when an operator notices that there is insufficient produce to fill the order. Thus, phantom inventory tricks retailers who find themselves unable to meet their customers’ expectations.

Poor information, analysis, and conclusions

The existence of phantom inventory is undoubtedly a problem of data, a discrepancy between physical stock and that recorded in the system. And as the saying goes: garbage in, garbage out. This means that when a retailer has a phantom inventory problem – and we have already established that it is very likely to have one – the analysis and conclusions drawn from this data will be compromised.
Phantom inventory can lead to misreporting sales under projections, leading the retailer to think that a promotion is not delivering the expected result, or a product was not as successful as thought. The sales team could be analyzing and concluding without knowing that the product was not in stock or that the promotion was never implemented because there was not enough merchandise, even leading to take actions in terms of product mix or promotional strategy.

On the other hand, incorrect inventory recording will affect the supply chain. Phantom inventory will distort sales, and consequent demand projections, since the system will show available stock that has not been sold when in fact, it is not displayed correctly or available for sale. Thus, the inaccuracy will be perpetuated in the future, affecting the following projections, sales plans, store performance measurements, and the replenishment system without even knowing it.

At Frogmi®, we know that phantom inventory is highly costly for retailers. Solving the complicated dilemma of keeping stock under control while delivering the best assortment and service level to customers is no easy task.
In our experience, an efficient solution to address phantom inventory is to support the operation with advanced analytics, artificial intelligence, and machine learning to identify deviations in product sales. Studying demand patterns at a granular level will make it possible to investigate the possibility of facing a case of phantom inventory. The information may be used to trigger alarms -or even better, targeted tasks- to perform a product’s field auditory and analysis.

Implementing these technologies together with a task manager at SKU level will improve inventory assertiveness, reduce stock-outs, ensure shelf availability, and consequently increase sales. Thus, retailers can begin to make decisions based on reliable information and improve their results by paying attention to the data and the in-store operation.

Let’s put ourselves in the following situation: an in-store walk-through.

The objective is to experience the store from the customer’s perspective to assess whether the standards defined by the company are met.

When walking down the aisles, it is impossible to avoid noticing empty shelves, locations that do not have enough stock, out-of-stock products generating dead areas, and others that cover more than the optimum, taking advantage of open spaces.

When checking the system, the inventory indicates that there is sufficient stock. Then, the replenishment of the product is coordinated since -theoretically- the backroom has enough inventory. It only needs to display it and make it available to customers.

However, the problem does not end there. When looking for the products, it is discovered that they are not there, there is no stock inside the store or in warehouses. This situation is called phantom inventory.

What is Phantom Inventory

Phantom inventory is inventory that does not exist in reality but appears to be in stock, like a ghost. Thus, the inventory system shows existing on-hand units, even though it is not actually in the warehouse or available on the shelf for the customer to see and buy.

These discrepancies between system information and reality can originate for several reasons, such as typing errors in the sale, carelessness in inventory receipt, and product shrinkage. But regardless of its origin, the detection and management of phantom inventory become complex since the system considers that the product is in stock, even though the store has no physical units.

The pain for retailers is evident when they believe that a product is available but – in reality – it is not, risking OOS and the associated loss of sales, as we saw in our previous blog. Another drawback of phantom inventory is its perpetuity. As the system indicates that the retailer has enough stock, the reorder point is never reached to restock, maintaining the situation over time. Thus, a pivotal act to break this vicious cycle is detecting signs of phantom inventory.

The symptoms of phantom inventory

To manage phantom inventory, the first step is to identify it and locate these gaps. However, as phantom inventory can originate for several reasons, it is difficult to detect the root cause for its management. That’s why retailers should focus on recognizing the visible symptoms, those that make the existence of phantom inventory evident.

One such symptom is stockouts. We have already seen that one way to recognize this symptom is to have eyes and hands in the store, spend time physically verifying where there are stockouts, and then check them against the system to identify an inconsistency in the inventory.

Considering that the average retailer handles around 16,000 SKUs simultaneously, this manual task can be challenging to tackle. In addition, it is clear that the larger the assortment, the greater the complexity of the job, making the management of phantom inventory uphill.

The second option for pinpointing symptoms of phantom inventory will be to rely on data analytics and artificial intelligence (A.I.) tools that can predict potential stock outs through analytical observation, identifying patterns and inconsistency in the data. There are three alerts that may suggest some deviation and phantom inventory risk.

1. Decrease in turnover: a decrease in the turnover of any SKU in respect to the normal or the projection.
2. Sales $0: the data shows that there are no product sales, despite having positive inventory. It can also be considered an alert when sales show anomalies concerning the demand projection.
3. Changes in sales trend: The sales pattern of a product changes abruptly, for no apparent reason.

One of the most significant advantages of using a data analysis system with A.I. and Machine Learning is that the algorithms refine the predictions as they are used. In this way, the alarms will become more and more accurate, improving the ability to diagnose phantom inventory.

What to do to treat your stores

Once the symptoms and products at risk of phantom inventory have been recognized, the hypothesis must be validated. Regardless of the method used (manual or through A.I.), a cross-check should be made between the system and total physical inventory (units displayed in the store plus backroom units). This evaluation will shed light on inventory disparities and where adjustments need to be made so that the system information reflects the store’s reality again.

At Frogmi, we know that managing phantom inventory is a complex process and one that must be performed constantly. Regardless of the methodology used to detect the problem, the solution involves manual in-store work to validate the possible inventory discrepancy. Our SKU-level task management solution is ideal for addressing this challenge from both methodologies, as it allows you to view information at the store-product level.

For example, a manual phantom inventory survey can be performed. One of the functions of Frogmi is the ability to generate tasks from a product (or the absence of it), which means that when stockouts are identified during a store walk-through, an automatic replenishment task can be created by simply scanning the product. If the area in charge of replenishment does not have enough units to replenish, a case of phantom inventory will have been found, enabling its immediate management.

On the other hand, if an A.I. system is used to alert phantom inventory, Frogmi can be easily integrated to generate automatic validation tasks specific to each store at SKU level. Thus, each store will have the opportunity to manage those products that behave atypically in sales or inventory turnover.

In our experience, the ideal plan of action is to use both tools simultaneously, facilitating the diagnosis of phantom inventory and decreasing the associated losses. This reaps the benefits of A.I. while empowering store personnel with self-management tools, improving inventory management, shelf availability, and, consequently, sales.

Worldwide, food waste is one of the main challenges facing modern society. FAO studies indicate that 1/3 of the food produced worldwide is wasted. According to The Guardian, levels reach 45% of total production in the US; 35% of seafood, 30% of cereals, and 20% of meat and dairy products each year.  

In parallel, the impact of Covid-19 has exacerbated food insecurity across the world. FeedAmerica studies show that 1 in 6 Americans lack reliable food access, while paradoxically, tons of food that are fit for consumption are wasted every day. 

Food waste also has an environmental impact. We are talking about the resources used for production and transportation, such as soil, water, and energy. According to ReFed’s analysis, the cultivation of uneaten food is responsible for 14% of the total potable water used and 4% of greenhouse gases in the US. 

The effect of food spoilage does not end there. It also has economic consequences. According to The Guardian, in the United States alone, 218 billion dollars a year are lost in producing, processing, and transporting food that is not consumed, equivalent to 1.3% of the country’s GDP.  

 

For food retailers, such as supermarkets and C-stores, food waste represents one of the biggest challenges in inventory management and store operations going forward. They must keep in mind a crucial factor for their industry: perishability. Unlike other retailers, the food area handles a wide assortment of products with low shelf life, such as fruits, vegetables, and bakery products — and the more varied the mix of fresh produce, the greater the retailer’s risk of food waste.

Today we see how the triad of economic, social, and ecological factors push retailers to take responsibility for waste reduction as a global improvement opportunity. Fortunately, technological solutions come to support stores to prevent, control and reduce food waste. 

All sales projections made it clear that online commerce would boom. According to Statista, “E-commerce sales worldwide will expand frenetically in the coming years. In 2018 they reached US$2.8 trillion and are expected to increase to US$4.9 trillion by 2021.”  However, no one expected to face a pandemic and its effects, which have accelerated the growth of e-commerce, surpassing all expectations. 

Although some retailers saw sales declines during the pandemic of up to 60%, supermarkets with online sales saw considerable progress, reporting growth of up to 300% in March 2021. 

Supermarkets also benefited from the development of grocery delivery, which has a projected annual growth of 30%, according to studies by Civic Science. This service has been on the rise as consumers find it attractive to select items online and receive them at home, scheduling the day and time, allowing one of the week’s chores to be done from the comfort of home. 

Along with the growth of grocery delivery, Dark Stores are emerging, allowing supermarkets and other retailers to have exclusive space to store and prepare products for e-commerce. This format, which has already been adopted by major retailers such as Bed, Bath & Beyond, Walmart, and Carrefour, has several competitive advantages. It minimizes operational and housing costs for companies and improves inventory accuracy. But the most significant benefit is felt by customers who get faster delivery, with promises of getting their products in less than an hour. 

Retailers must be prepared to face some challenges that arise within this new dynamic. Among them, studies such as “Leveraging the physical network of stores in e-grocery order fulfillment for sustainable competitive advantage” highlight the major factors to be considered: integration between online and physical sales channels, logistics coordination, and inventory management.