Real talk alert: averages are easy to understand and calculate, but in reality, most data sets include a few very heavy outliers which significantly manipulate the meaningfulness of the averages. Therefore let's take a look at percentile analysis — you're going to like it!
Percentiles indicate the percentage of scores that fall below a particular value. They tell you where a score stands relative to other scores. For example, a person with an IQ of 120 is at the 10th percentile, which indicates that their IQ belongs to the top 10% and thereby is higher than 90 percent of other scores. Percentiles are a great way to understand the relative standing of a value. Where does a value fall within a distribution of values?
Our Percentile Analysis divides your customer base into ten equal-sized intervals by ranking them e.g. based on their order quantity. For example, the top 10% percentile includes the best 10% of your customers based on their order quantity.
For each percentile, we then calculate their key metrics that help you better understand them:
|Average number of orders per customer
|Percentage of total gross revenue generated by customers in the percentile
|Average Order Value - calculated as Sum of Gross Revenue / Count of Customers - within each percentile
|Sum of Gross Revenue per customer
|Sum of Net Revenue per customer
|Estimated Lifetime Value per customer
|Amount of months your customers remain active
|Product Return Rate
|Percentage of items that are returned; exclusively available for custom integrations, , and .
By default, revenue is excluding taxes, shipping revenues, and discounts. Although we highly recommend using our default definition, but you can modify the revenue calculation under Company Settings > Report Settings. Please note that this will trigger a complete re-import and recalculation of all your data.
Isolating and analyzing your percentiles is powerful if you want to find out:
- Whether your revenue is very dependent on a small percentage of customers
- Whether product returns are a fundamental problem for your business or whether a certain customer group is mainly driving your return rates; you could think about introducing return costs for them
- Where you should set the boundary for the creation of your customer segments
- Who your top customers really are and e.g. appreciate them with a loyalty program
What You Need
For this report to work properly, the following data must be imported:
- Order ID
- Order Date
- Stock Keeping Unit (SKU)
- Items Sold
- Item Price
- Shipping Revenue
- Product Returns