Inventories are an essential component of a retail business balance sheet, as they can account for about a fifth of a company's total assets. However, looking at the P&L report, you won't find any inventory. So why do we assume that inventory problems can somehow affect your EBITDA? And why do we even choose to pay attention to this indicator since many financial experts believe that it is not as important as net income?
Let's start with answering the last question first. According to Investopedia, EBITDA (or earnings before interest, taxes, depreciation, and amortization) is an indicator of the overall operational efficiency of a company; in other words, it shows how "healthy" your operational processes genuinely are. If we want to compare the efficiency of our business with that of other market players, then using this indicator will be more useful than, for example, comparing net profit. This is especially true for multinational corporations because taxation conditions, credit rates, etc., may differ from country to country, leading to significant adjustments in net profit, even with the same turnover, operating expenses, etc.
Therefore, to answer the first question, we need to delve deeper into the causes of inventory problems and their consequences and trace this relationship through some calculations. Eventually, we will understand how to avoid inventory problems and improve the operational efficiency of your business.
How does inventory affect your business performance?
The impact of unbalanced inventories on the business is enormous, although it is not obvious, and their size or cost are not reflected in the income statement. However, as they say, “the devil is in the details”.
Being an asset, inventories should be in constant motion and bring profit. The actual sale of goods creates revenue, which is reflected in the income statement and definitely affects EBITDA through gross profit. No goods — sales go down – the gross profit goes down and EBITDA with it. On the other hand, excess inventory might force discounted sales, which also pushes EBITDA down despite increased sales. Plus, if the excess stock goes bad and has to be written off, then we reduced assets and created no revenue, which is the worst case scenario because the company’s costs go up while profits go down.
Additionally, let’s not forget SG&A (i.e., operating expenses), which constitute another component of the EBITDA formula. The impact of inventory on this part of expenses can also be significant and equally under the radar.
Why do inventory problems arise?
Inventory problems bring us to the complexities of forecasting. If we knew exactly what tomorrow’s sales would be, we would have no problem. We would order exactly as much product as we needed without concern for overstock or stock-outs. That's what it would look like in a perfect retail universe. In reality, the problem with demand is not just that it’s fickle; it's also subject to a wide variety of influences. From the more predictable, such as seasonality, days of the week, promotional activities, region specifics, etc., to spontaneous spikes due to changes in weather, national or world events, for example. Simply said, we don't know and never will know exactly how much product we need to meet tomorrow's customer demand. We can calculate it on average, relying on previous periods, and make some adjustments for trends and anticipated events. However, the probability of hitting the exact numbers is still often far from 100%.
When a retail manager says that a product sells 123 units per day, he is most likely referring to ADU = Average Daily Usage, and “Average” is the key word here. In reality, there may not be a single day in the daily sales for the past week or month with a sale of 123. It will be a set of numbers like 99, 108, 133, 140, 107, 128, etc., and that's why we average. But what’s going to happen if we just order that average 123 units? The Gaussian distribution curve helps answer this question. Those who have dealt with probabilities and statistics are undoubtedly familiar with it.
Our average of 123 units — defines the peak of the chart and a 50% probability of guessing. That is, actual demand will either be less at 50% probability or more at the same probability.
Hardly anyone in retail would agree to a 50% probability of a shortage of goods when forming orders since a lack of goods would immediately affect sales and profits. Therefore, the desire to reinsure and purchase more goods is quite logical, but only within certain limits (we will consider the negative consequences of this approach a little later).
The distribution curve clearly shows the problem of inventory management: to achieve a high level of goods availability, it is necessary to order quantities significantly higher than average. Moreover, after passing the 95% probability boundary, each next percentage point requires a noticeable increase in inventory.
The difficulty of retail forecasting arises from the fact that real distribution curves have a more complex shape than the ideal Gaussian curve; they are asymmetric even for fast-moving category A items, and the less turnover an item has, the more these curves differ from the ideal one.
If an item sells an average of 0.5 units per day, that is often 0 or 1 but sometimes 2 or even 3, then to ensure the same 98% availability, we need to order a lot more compared to average sales. As a result, the desire to ensure high availability can lead to a colossal overstock, even resulting in a drop in sales, not to mention other economic consequences. Also note that this curve will not only be different for different SKUs but also in different locations of the same chain because there may be a different price segment, customer profile, etc.
In practice, no one in retail really analyzes distribution curves when forming orders. If the network has 10,000 SKUs and 100 stores, it is incredibly time-consuming to build and analyze 1,000,000 curves. Therefore, this task is most often solved by a combination of a manager & Excel, with each manager coming up with their own formulas and methods. But even if the calculations are performed by an ERP system, the algorithms there are most often relatively simple. And all these methods almost always lean towards increasing orders, because the main thing is not to lose sales. Given all the difficulties we have discussed so far, it is not surprising that not all retailers carefully manage the economics of their inventory, and sometimes, they don't even pay attention to it until it's too late.
“Bipartisan system” in retail: sales at any cost or economics at the head of the table?
Experienced retailers will not deny the existence of the eternal confrontation between salespeople and financial experts when it comes to developing retail strategies. Let's call them the party of "salespeople" and the party of "economists". Ideally, these two centers of business responsibility should be relentlessly engaging in polemics, defending their interests (i.e., performance indicators), and, with competent management and approach, keeping the system in the right balance and harmony.
In practice, however, undistorted balance is rarely achieved. One or the other camp takes power at different times and stages of the company’s development.
Sales, sales, sales. Sales above all
In the initial stage of a company's development, "salespeople" are often the main driving force. It is logical because building up turnover and increasing client numbers is paramount. The demand planning model is approximately the same: average sales plus an adjustment for seasonality, trends, promotional campaigns, and a little extra just to be sure to provide a high level of availability and, eventually, sales.
When the party of salespeople rules and stays in power for a long time, availability can indeed reach 97% and even 98%. So what’s the catch?
Using this kind of planning, overstocks cross all reasonable limits, and this leads to increased logistical and operational costs and starts to affect sales negatively! It can be called overstock chaos. Goods occupy shelves and warehouses; they need to be sold somewhere, and you need to make sell-outs. When customers only buy what's on sale, they don't buy other, more marginal goods. As a result, turnover goes up, but margins go down. On top of that, a product may run out of shelf life and be written off, which also increases costs.
If these are non-food products, they can lie on the shelves or warehouses for quite a long time, but eventually, they become obsolete and lose their marketable appearance. The next course of action is to move them around various retail outlets and warehouses until they are finally sold at a significant markdown or written off. Overall, overstock takes up shelf space that could be occupied by more attractive goods and also increases logistic costs. The company will struggle with storage space and launch projects to expand it without addressing the cause of the problem. Indeed, we don’t need to stress how ineffective that is.
Where's the money?
Next, the party of "economists" comes into play. Especially if a new CFO comes in. The company starts to look at the economics of the business, moves towards moderation, and begins to fight overstocking. The motto "sales at any price" is replaced by "we don't need maximum sales, but profit maximization.”
When quick results are needed, such tasks are hardly completed by adjusting to the demand for individual products, but rather at the level of principles. Inventory turnover in days is most often chosen as a "balancing" indicator, and the criterion is "no worse than competitors". But since it is really difficult to find a balance between lost sales and acceptable turnover, additional rules and restrictions are introduced. For example, goods are divided into several categories (ABC), and for these categories, different combinations of turnover and availability are used, and algorithms are developed or adapted for these combinations. There may be more complex methods, but without a well-thought-out and complex mathematical model for each product in each store of the chain, it will not be possible to achieve an optimal balance between lost sales and acceptable turnover. This will lead to a decrease in sales, which will again raise a "revolt of salespeople" and induce another change in who is in charge.
Getting out of the vicious cycle
Why do "salespeople" win initially? When we lose sales, the formula for calculating the loss is evident to everyone. Simply put, 95% of the availability of goods is 5% of lost sales. We estimate it in money, and if the margin is, for example, 25%, it means that 25% of this amount is what we lost. If it is a large chain of stores, such colossal figures can appear horrifying to the management.
Calculating all the additional costs and losses associated with excess inventory is much more difficult because they are diverse and not always obvious. The stereotype "well, we will sell this product someday anyway" often works, so the incurred costs don’t affect EBITDA and profit. While it may be true for the case of a specific SKU, it is not valid for a large number of overstocks at the company level. Sold overstock on a particular SKU will be replaced by other overstock on other SKUs, and, therefore, it is necessary to systematically evaluate all overstock and its impact on the retailer's numbers. In the short term, an increase in inventory will affect operational logistics costs – salaries to warehouse employees, outsourcing, fuel costs, etc. If overstocking becomes systemic, it will eventually increase fixed expenses, such as depreciation or warehouse rent, which becomes insufficient due to overstocking and has to be expanded. And then there is the cost of money stuck in overstocks, the need for sales and write-offs, and the increased cost of promotional activities to sell stale goods.
When it comes to optimizing inventory without losing sales, we need to be talking about a systemic solution with a long-term impact. It’s not about inventory reduction but optimization, which means that we get the right product in the correct quantity in the right locations and at the right time. It might sound like an impossible dream, especially if you use traditional methods, but it’s no longer unachievable.
You will learn how to find a simple solution to this problem in Part 2 of this article.
