Keeping your product catalog fresh can be an important part of bringing customers back for repeat purchases. Few things are more exciting than bringing new inventory into your catalog.
How will customers react to new products? How can you tell if they are selling well enough?
Short of consulting a crystal ball, it can be a mystery to figure out how much to order without prior sales history. Here are some key points of data to use as a guide for placing your initial order.
Look at other new product launches
Consider what you have done in the past since this is the most readily available data. Have other new products increased sales quickly upon release? Or does it take your customers weeks or months to warm up to a new product? Examine the month-over-month increase for previously launched products, assuming similar launch resources and plans are being invested in this new product.
As an example, previous product launches with sundresses, maxi skirts, and swimsuit coverups saw strong sales in the first six weeks, then a drop-off during the following months.
Applying that to our new styles of beach-ready summer dresses, we can estimate there will be a drop in sales after six weeks.
Look at trends within the same category or brand
Look at products with similar attributes to your new line. This could include category, collection, brand or vendor.
For example, Cora’s Candles show solid sales across all available scents. When considering Gingerbread as a new scent release, we can look at the sales velocity of Christmas Tree and Spiced Cider flavors to approximate the future sales for the Gingerbread candle.
Looking at other scents, we see a pattern of a strong increase in the second and third months after launch. Seeing this pattern in several other launch scents can inform you how the new Gingerbread candle might perform.
History of the option sets
Option sets are different colors, sizes, or flavors of the same item. For example, when you look at t-shirts, you can look at the history of the different sizes (small, medium, large, etc.). The proportions each size sells at will guide your initial order of the new t-shirt. If you sell 25% size small, 50% mediums, and 25% larges, use this distribution when determining sizes of the new style to order.
Merging sales history of similar products
Merging sales history is ideal when you have an old product being phased out and a new product coming in that is very similar. Link the sales history to the new product so that even though you don’t have any sales history, you have a good idea how it is going to sell based on the old one.
An example here is a food item selling in a 2 oz. size and being replaced with a 2.1 oz. size. Merge the sales history so you can see how the 2.1 oz. will perform based on how the 2 oz. size sold. The change in size is not significant enough that we expect a change in the rate of sales.
How to handle seasonal product with a short sales history
Top-down forecasting uses sales trends for categories and forecasts the future sales based on the product contribution to the category.
Seasonal forecasting references what happens 12 months ago. If you look ahead to how a product sells in August 2019, you must look at how it sold in August 2018 and even August 2017. When you have seasonal items with less than 12 months of sales history, consider using top-down forecasting.
If an item has only been for sale this summer, you must use top-down forecasting because we know this category is seasonal. Take sunglasses: They tend to sell better in the summer, not the winter. Use that seasonal trend.
Once you have a little sales history, you know a particular sunglasses style contributes to 2% of the unit sales. Tack that 2% to the seasonal wave. As the category dips in the winter, you are looking at 2% of that lower number will be forecasted to your new seasonal product with only a little sales history.
While at first glance it may appear you have no data for your new product, there are metrics you can use to guide your purchasing decisions. Consider recent product launches, trends in the same category of brand, option set performance, and merging the history of similar products. All these are excellent starting points for forecasting your new product meet customer demand.