How can we combine e-commerce key performance indicators (KPIs) such as Page Views  and Conversion Rate with inventory merchandising specific ones such as Sell Through Rate? In this post we will look at how we can evaluate an item’s sales performance using standard e-commerce KPIs for additional context.

The Sell Through Rate is calculated by taking our cumulative sales over a given time period divided by our total inventory. This KPI, when using the product’s entire sales history, gives us the ability to see how far into the lifecycle of a product we are and when using only a given period’s sales we can see how well the product is doing so we can compare it over time.

There are two ways of calculating this KPI and they are similar in function.


The difference between the two formulas depends on whether or not we wish to include Shrink. If we received 1000 units, sold 800 and currently have 150 on hand this implies that 50 units were lost to shrink. Using both versions of the formula will have two very different sell through rates:


Excluding Shrink:


Including Shrink:

In the first one we are basing our calculation primarily on the sum of Current Inventory and Cumulative Sales and explicitly excluding any Shrink since it is not directly related to the item’s potential to sell. In our example, when we reach 950 units of sales we reached 100% sell through implying that we sold all we had available and have nothing more to sell.

In the second formula we are basing our formula solely on Receipts and sell through will appear to be slower and using this method we will never achieve 100% since the 50 units due to shrink can never be sold.  When determining which KPIs to use and how to use them, we often use them not only to measure something in isolation but we also get greater insight when we use them to compare with like items.  Since we don’t know if all our items have the same shrink rate it is best to exclude it from our Sell Through calculation to make a proper comparison. Therefore the Excluding Shrink version of the formula ought to be used.

The formula takes on different meaning when we want to see what this week’s sell through rate is. Adjusting the formula to use only this week’s sales in the numerator and showing it week over week gives us a series, which gives us the ability to chart an item’s progress over time. This, of course, could be adjusted to be any time period that is needed such as month, quarter or season if the optics make better sense. 

Much like the Weeks of Stock KPI covered in a previous post, this metric only gives us a partial view and it needs other metrics to give us the bigger picture. If we’ve reached a cumulative 84% sell-through but it took us 125 weeks to get there, we are either overbought or simply chose the wrong product to sell. If we’ve reached 84% sell-through in one week then maybe we likely don’t have enough inventory and need to consider if we can get some more soon. If the sell-through accelerated in the last few weeks was it due to a promotion or the weather turning in our favor? Consider the following chart showing a sample item’s sales history (Product A) and calculated Sell Through Rate over time. We can see how weeks 7 and 8 show a dip in the weekly sell through. Imagine being at the beginning of week 9 and seeing that dip from the two previous weeks. This would be a cause for concern.

As is often the case, KPIs always offer greater insight when paired with others to give us context. Here is where Page Views and Conversion can give us some further guidance on how our products are doing. Let us look at Weeks 7 and 8 alongside Page Views on a line chart.


We can see that the Sell Through Rate dipped in the last two weeks but so did Page Views. Our conversion rate remained relatively steady hovering around 3%. So it seems our product has not necessarily done poorly, we simply didn’t get the traffic on the product we need. This would warrant an investigation as to why but at least it restores our confidence in Product A.

Compare this with Product B’s history for the same time period.



Here we can see that the Page Views and Conversion Rate did not take the same hit that Product A did. The conversion rate shows an obvious dip on weeks 7 and 8 meaning that we were still maintaining the traffic we ought to expect but the product was just not cutting it as far as our customers are concerned.

So why use Sell Through Rate at all if Conversion offers us similar insight? It’s because Sell Through lets us know how well positioned we are with respect to our immediately available inventory especially when looking at the cumulative performance.


Questions to consider:

  • How can we best make use of Sell Through to compare items with each other?

  • Can we use Sell Through as one of the key elements for Predictive Analysis using Machine Learning algorithms?

  • Can we use Sell Through as a method to classify our collection?

  • What are the key decision making support areas that Sell Through can help us with?

Contact us to discuss these ideas and more!