Five customer retention metrics worth measuring
It is easy to call a loyalty program successful if you only look at sign-ups. Signing up costs the customer nothing and guarantees the business nothing. Here are five metrics that show the real picture.
1. 30-day return rate
The share of members who came back within 30 days of their first purchase. It is the most sensitive early indicator — it moves faster than revenue and tells you whether the program works at all.
If it is low, the usual causes are a reward that is too far away, or zero communication after sign-up.
2. Purchase frequency
Average visits per month, measured separately for members and non-members. This comparison is the closest thing to causality you can get without running complex experiments.
One detail matters: compare the same periods and the same locations, or seasonality will distort the result.
3. Redemption rate
What share of earned rewards is actually claimed. A healthy range is 40–70%. Lower means the mechanic is too hard or customers do not know their balance. Close to 100% usually means the reward is too cheap to drive extra visits.
4. Identified share of revenue
What percentage of total sales happen with an identified customer. This one determines the value of all your other data: if only 10% of purchases are identified, your analytics describe a minority. A practical first-year target is 30–50%.
5. Activity decay
How many members who were active 90 days ago no longer buy. This is loyalty churn, and it is the best trigger for automation: send this group a separate offer instead of blasting the whole base.
How not to misread the numbers
Three rules. First, never judge a month without comparing it to the same month last year. Second, separate new and existing customers — growing traffic hides falling loyalty. Third, one metric moving without the other four is usually noise, not a trend.
In the Loyalty.lt partner portal these cuts are available in reports, and segments can drive push and email campaigns. More on the features page.