The term SKU (Stock Keeping Unit) is associated with a product (usually, but not always, a physical one) and identifies the most detailed disaggregate description of the product. Let me give an example – assume you produce shoes. They are your products.
The somewhat aggregated specifications of your product would be: a specific shoe model, in a specific color, made from a specific material.
A given shoe model might come with several available sizes. An SKU can be the code for identifying a specific size. It can even be something more disaggregated, like that size in a specific warehouse or retail store.
An SKU usually corresponds to detailed and specific product description. In other words, you can freely exchange two product items with the same SKU identification without any problem.
An SKU (and its uniquely identified code) is extremely important for inventory management and control. If the inventory is like a building, SKUs are the bricks.
Businesses usually record historical data on the SKU level. Thus, any good inventory forecasting software solution will be based on the analysis of past sales data on a similar level. They usually do forecasting on the SKU level, too.
Let’s take shoes as an example. Sometimes inventory forecasting software solutions aggregate different SKUs and generate a forecast at a higher level of aggregation. As an instance, we might aggregate and forecast the inventory demand for that specific shoe type, disregarding the different sizes.
This way of producing inventory forecasts has some advantages. It is usually more precise for the aggregate forecasted demand, but there are also disadvantages. Sooner or later you will need to split your aggregate to the SKU level.
SKU vs UPC, EAN and barcodes
These terms are often used interchangeably, but they are not the same thing. An SKU (Stock Keeping Unit) is an internal code, defined by each company for its own purposes. The exact same product can carry a different SKU at every retailer that sells it.
UPC (12 digits, used in North America) and EAN (13 digits, used in Europe and the rest of the world) work differently. They are universal codes assigned through GS1, so a given product keeps the same UPC or EAN no matter which retailer stocks it.
A barcode, finally, is not a third kind of code. It is just the graphic format a scanner can read, and it can encode either an SKU or a UPC/EAN.
So why is forecasting done on the SKU? Because that is the level at which a company decides what to reorder, in which variant, and for which warehouse.
Why should you make the most out of your SKU level data?
A couple of reasons are as follows:
- SKU level data will allow you to know more details about your purchases and about the people who made them. You’ll learn about the favorite size or color of a given customer with the help of this data.
- Having all this information at hand, you will be able to improve your marketing strategies. As a result, you’ll send more relevant offers to your customers – ones that are actually in the scope of their interests.
Returns at SKU level
Your SKU-level return rate should be calculated for each SKU separately: returned units divided by units sold of that specific SKU, not of the whole product or category. A size or color that comes back too often inflates your gross sales and hides what customers actually keep, which distorts true net demand. For this reason, it’s best to forecast gross sales and returns separately, or to work directly on net demand. That way, you won’t reorder SKUs that are simply going to return to your warehouse.
Use your SKU level data to encourage your customers to buy more of what they want to buy. It will help you increase engagement, foster customer relationships, and their shopping experience.
Just keep in mind that regardless of your demand forecasting technique and tool, eventually you will be interested in SKUs. Not only for inventory demand forecasting, but also for the inventory and its control. You will need to replenish some SKUs and leave others untouched as they are beyond the level which triggers a new order.
SKUs are of crucial importance as they allow you to track your stock levels for each product and material type. This leads to higher optimization of your inventory levels and more productive inventory management.
Benefits of integrating a solid SKU plan into your business
- Improved Work Efficiency: Using SKUs helps everyone internal understand the products and their movements efficiently.
- Fewer Communication Errors: Having to deal with long product names and variant attributes can cause communication errors and affect the workflow. SKUs help reduces the occurrence of such errors and misunderstandings.
- Simplified Stock Taking: With the help of SKUs, the identification of variants becomes much simpler and easier. This way you can know if your inventory matches the actual stock levels.
- Product Prioritization: SKUs enable you to check out the sales volumes for each variant of your products. Some colors or sizes might be more in demand than others. Knowing your customers’ preferences, you can essentially optimize your production and increase your sales.
- Better Inventory Management: Having reorder points set on each variant on your SKU will highly improve the efficiency of your inventory.
Improve and optimize your SKU plan in order to take advantage of everything that SKUs bring to the table for a business owner or a manufacturer.
To sum up, SKUs are your products, described with all the details you need in order to keep inventory and cost under control.
How to create SKU codes
A good SKU code reads like a short description of the product. Take a running shoe sold in several colors and sizes: RUN-ZX1-BLK-42 tells you at a glance that it belongs to the category RUN (running shoes), that the model is ZX1, that the color is BLK (black) and that the size is 42.
A few simple rules keep codes clear and easy to maintain:
- Keep it alphanumeric and short. Aim for 8 to 12 characters; the example above uses 11, plus the hyphens that separate each element.
- Go from general to specific. Start with the category, then the model, then the color, then the size.
- Avoid leading zeros and confusing characters. Letters and digits that look alike, such as O and 0 or I and 1, lead to picking and data entry errors.
- Never reuse the code of a discontinued item. Old sales history would otherwise get mixed with the new product.
- Create one SKU for every sellable variant. Every combination of color and size needs its own code, so black in size 42 and black in size 43 are two different SKUs.
A consistent structure also pays off in forecasting: it lets you group SKUs by model or by color to produce a top-down forecast, then split it across the individual variants.
SKU Planning
Let us look at how SKUs may be used in planning and SKU demand forecasting. SKU-level forecasting is the prediction of future demand for each individual SKU (each size, color and, if needed, location) using that SKU’s own sales history, instead of forecasting at product family or category level, and any good inventory forecasting software solution can do it. Inputs: historical sales by SKU, seasonality and promotions. Output: forecast demand for each SKU and period (week or month). Use: reorder points and purchase quantities for each SKU.
Let us look at the example of shoes. Occasionally, inventory forecasting software solutions aggregate different SKUs and generate a forecast at a higher level of aggregation. Your shoe company might aggregate and forecast the inventory demand for a specific show type while disregarding the different sizes.
This way of producing inventory forecasts has both advantages and disadvantages. While it is usually more precise for the aggregate forecasted demand, sooner or later you will need to split your aggregate to the SKU level.
Forecasting a new SKU when no history is available
We’ve assumed that enough history of an SKU is available in order to get a good forecast and, obviously, it is not always the case. How can we deal with a very short history or with a brand new product in the inventory?
Things get hard when there is no sales data, especially for a tool that relies on it, as it is for demand forecasting software. The more history, the more accurate the predictions. A thumb rule that is usually true.
However, the time series of sales is not the only source of value. But we’ve to analyze different cases.
A new SKU that replaces an old SKU
This is the case of a new model of a product series. Let’s think about a smartphone: the new model is appearing in the market with huge investments in marketing by the manufacturers. The first thing that comes to mind is that the new SKU will replace the previous and the missing time series history could be inherited from the old SKU. Oh, awesome, close to victory.
However, the new smartphone will cost more than the previous in the first weeks, thus it’s not plausible that the sales of the old model will drop to zero instantly. That is, the movement of sales from the old SKU to the new SKU is gradual and it is commonly called cannibalization.
Thus, a demand forecasting software can exploit such relationship and do a good forecast. The shift of sales from the old model to the new one is gradual and is estimated with a substitution curve over a few weeks: for example, the new SKU goes from 20% to 100% of the combined demand, while the old one declines in parallel. The shape of the curve is derived from previous launches of the same series.
A new SKU comes with a lot of information
Even if no history is available, since it is new, an SKU comes with a lot of useful information. Here we’re going to briefly introduce a few ideas on how to get data to work with.
A new SKU, let’s say a new model of white ripped jeans, has intrinsic data. Firstly, it is a pair of white jeans, then it is a pair of jeans.
Oh, will it be sold in Arizona? Yes, let’s have a look at the sales history of white jeans in Arizona. Uh, a lot of information.
This information can be combined and clustered in order to get insights into our brand new white ripped jeans future demand.
Finally, no matter if no sales history is available. Today, the demand for a new SKU is estimated by leveraging its product attributes (category, color, price range, season) and the sales of similar SKUs already in the assortment, a like-for-like approach. For a new SKU, several solutions can be exploited in order to get a good forecast.
