👥Order Retention

This report is derived from transactional sales data and is meant to be used as a reference with the SmartModel. The monthly retention curves in this report are the same actuals used within the model.

This report differentiates from the Customer Retention report by calculating a ratio of orders within an interval since first purchase (e.g. Month 1) divided by the number of orders that were made within the first interval since first purchase (e.g. Month 0). In contrast, Customer Retention does not display the impact of multiple orders, but instead shows the proportion of the originally acquired customers that are making returning purchases over time.

Definitions

Since Day 0, Week 0 and Month 0 will always be 100%, they have been excluded from this report to allow the focus to be on the following periods.

Metric/Dimension	Definition
Order Retention	Cohorted by when they made their first purchase (Cohort Interval), the number of returning orders divided by the number of orders from the cohort (Cohort Size) in the first interval (e.g. Month 0). For example, Month 1 retention reflects the number of orders in Month 1 from customers acquired in the same Cohort Interval divided by the number of orders that the same cohort made in Month 0.
Cohort Interval	This is a control available in the Filters menu with the possible options of Quarter, Month, Week, and Day. Each option groups first-time customers together into cohorts based on the interval selected. For instance, a customer that makes their first purchase on January 15th, 2024 is cohorted into Q1 2024 (quarter), Jan-24 (month), (week of) Jan 14, 2024, and Jan 15, 2024 (day).
Interval Since First Purchase	This is a control available in the Filters menu with the possible options of Month, Week, and Day. Each option calculates the calendar distance since the first purchase.
Segment	Unlike the Customer Retention report where customers are completely segmented by the first segment (e.g. Subscription) they purchase, this defines the segment that was actually purchased on both the numerator and the denominator of the Order Retention formula.

For example, let's assume we're looking at Subscription as a Segment and the cohort of Jan. 2020. There were 10,000 orders of the Subscription in Jan. 2020 by customers acquired in the same month. In Feb. 2020 (Month 1), customers acquired in Jan. 2020 made 3,000 orders of the Subscription Segment. The order retention reflected is 3,000 / 10,000 = 30%.

Order retention is simply a ratio and does not require that the 3,000 orders came from the same customers that made the 10,000 orders. In fact, it's possible some of those customers went on to purchase NonSubscription in Feb. 2020, but that would not be reflected in the order retention because we're purely looking at the Order Retention of the Subscription Segment.