One of the reasons why loyalty programs exist is to gather better data on customer behavior: if a customer is incentivized to identify themselves when they interact with our brand (presenting their loyalty card/info to collect rewards) we have a much clearer view of that customer’s interactions and purchases.

Which loyalty metrics are you tracking? Here are three you might be missing. http://ow.ly/ACj01

We have been a part of over 3,500 loyalty programs at Sweet Tooth. In that time we’ve used some cool metrics and gathered some pretty strong data to analyze; it’s a data nerd’s paradise! So when I knew I was writing a guest blog post for RJMetrics, I had to do it on data.

Here are three loyalty program metrics that most retailers won’t know, or use, but should. Where appropriate, I’ll include what we have found in our data from over 3,500 merchants.

Realized Customer Lifetime Value

Customer Lifetime Value (CLV) is the total profit (some use revenue) that a customer will generate over their entire lifetime. There are two ways to look at CLV:

  • Predicted CLV: How much profit you expect to come from a customer
  • Realized CLV: How much of the lifetime profit you have already made

Realized CLV is an extremely effective way to segment and target customers.

How to calculate it

To calculate realized CLV, take a given customer’s predicted CLV and divide by their actual lifetime profit. This will show how much of a customer’s lifetime value they have achieved, as a percent.

realized-02

How to use it to improve your business

  • Use it to find your brand promoters. If a customer has realized close to 100% of a high predicted CLV, then it is very likely that they are happy with you. This is the perfect time to introduce a referral campaign. Remember, the total value that a customer brings isn’t just in profit that they directly generate; brand promoters can help find more high-value customers just like them.
  • Use it to measure the accuracy of predicted CLV. If a significant number of customers are regularly not realizing their predicted CLV, it means that the method being used to predict CLV is not accurate. Keep in mind that it takes time for a customer to realize their full CLV, so be sure to only analyze customers who have had sufficient time to do so.
  • Combine it with cohort analysis to improve your customer experience. When combined with cohort analysis, realized CLV can point you to cohorts of customers that likely won’t realize their full CLV potential. You can determine what is causing this low CLV, and improve the overall customer experience. The same can be said for a cohort of customers who have high CLVs.

How does your realized CLV compare to your predicted CLV? http://ow.ly/ACj01

How do you compare to other ecommerce retailers?

An average or range for this figure isn’t very useful, as this metric usually has a lot of variance. In general, retailers should aim for their customers to reach 90% of their (individually) predicted CLV. Note that if you’re using a general CLV metric for all of your customers (instead of a CLV for each customer or segment), then you should be aiming for approximately 70% accuracy.

A more practical way to measure realized CLV performance is to set intervals where customers should be at a certain realized CLV. For example, after a year we might set a target realized CLV of 30%.

Breakage Rate

A loyalty program’s breakage rate is the ratio of points that do not get spent to the total number of points earned. Put simply: what percent of points aren’t spent?

Any retailer with a loyalty program can make use of this metric. But it can also be applied to other promotions as well. Any time you offer a customer some sort of discount or gift you can use a metric similar to breakage. The percent of unused abandoned cart coupons, the ratio of customers who didn’t accept a free gift during checkout, etc, are all useful situations for calculating breakage.

How to calculate it

To calculate breakage, find the total number of points that have not been spent, then divide this by the total number of points issued. This can be calculated for retailer’s lifetime, or during a particular time period. For example: the calendar year.

loyalty-02

How to use it to improve your business

  • Determine the financial liability of loyalty points. Accountants will use breakage to determine the outstanding financial liability of points for a loyalty program, usually towards their fiscal year end. Accountants love breakage (free money!), but it often indicates that something is wrong with your loyalty program.

BreakageIsBad

  • Use it to determine how engaged customers are with a loyalty program. If customers aren’t spending points to get rewards, then they are not likely to be active, loyal customers. It might also signal that your customers don’t know how to participate in the loyalty program.

How do you compare to other ecommerce retailers?

From our data, the average breakage rate is about 30%, but there is a ton of variance. Breakage will vary according to how frequently a customer visits or purchases, how engaging your program is, if you have point expiration, and if you have frequent communication with the customer, to name a few.

Average Time to First Spend

The average time to first spend is a measure of long it takes for customers to go from creating a new loyalty account to the first time they spend their loyalty points. This could be days, weeks, months, or (hopefully not) years.

The average time to spend metric is used to determine how long a customer takes to go through one “full cycle” of a rewards program. Basically, how long does it take for a customer to spend points and feel rewarded?

How long until your customers first start spending loyalty program points?http://ow.ly/ACj01

Knowing where a customer is in this rewards cycle is extremely useful for creating promotions and segmentation.

The average time to spend is directly related to a customer’s purchase frequency as well as what actions are rewarded. If a customer is making regular purchases, they’ll accrue points faster and will have a lower average time to spend. Similarly, if a loyalty program rewards customers for several actions, such as reviews or social sharing, then the customer will take less time to reach a spending threshold.

How to calculate it

First, calculate the time in between a loyalty account being created and its first point spending event. Calculate this on a per account basis. If you have a large customer base, use a sample size (make sure it’s a statistically significant sample size). Take the average of these and you’re done.

time-to-spend-02

How to use it to improve your business

  • Optimize customer communications. If a customer is approximately 90% through their average time to spend, it is a great time to ask them to perform an action that will earn a reward. They’ll be more motivated than usual to earn points because they’re close to having enough points to earn a reward. You could send them an email letting them know that they receive points for referring their friends, or send them a coupon giving “double points” on their next purchase, or any action that you determine valuable. These campaigns are often our most successful, in terms of engagement. All of this is possible because we understand the average time to spend.

How do you compare to other ecommerce retailers?

From our data we see an average time of 92 days to spend points. Note that we also see a lot of variance in this number because there are a lot of variables that affect the average time to spend.

Loyalty & Data: Best Friends Forever

Like I said, one of the main reasons why loyalty programs were created was to give a customer an incentive to identify themselves at each interaction. This incentive comes in the form of collecting reward points.

Loyalty programs are great for developing happy, loyal customers that spend more and refer their friends, but don’t forget that loyalty programs are also great at giving data-driven retailers better data.

ad-13

  • http://www.gozkart.com Sanaya Shaikh

    good and intresting article thankz for sharing