We are all aware that when implemented properly, loyalty programmes are wonderful for both retail firms and their customers. However, the adjective “excellent” alone can never satisfy a true expert in online retail. To determine how “amazing” anything actually is, she wants to examine the statistics. And cohort analysis and customer loyalty analysis can also be useful in this regard.

This raises a lot of concerns, including how to evaluate our loyalty programme. What metrics ought to we consider? What are the criteria? What figures are considered “good” or “bad”?

Retention

Let’s begin by stating the obvious. When you manage a loyalty programme, you want your customers to shop more frequently and return to your store. You wish to improve retention.

The math is straightforward: when you keep consumers, you spend less on marketing (and related costs). In order to keep clients from switching to your rival, you can return some of that money to them in the form of discounts, gifts, coupons, and other incentives. In theory, customer retention is a win-win situation.

In actuality, only if the repeat purchase takes place, is it a win-win situation. I want to be very clear about this: when assessing a loyalty programme, you shouldn’t concentrate on:

how many repeat visits you have on your website and how many individuals have opened the emails you send out for your loyalty programme
the number of viewers of your loyalty commercials, etc.
Of course, these figures are crucial determinants of other choices, such as whether you selected the appropriate email subject line or whether the appropriate information is present on the appropriate pages.

However, you need consider the sales data if you want to determine whether your loyalty programme is effective or not:

Sales Volume Revenue Profit
Sales figures, repeat customers’ numbers, etc.

What period of time is ideal?

Retention each month? Every week retention? Everyday retention?

A good query. Some retail companies have relatively consistent retention rates. Some don’t. Consider a shop that sells coffee capsules. People go out and buy coffee capsules if they run out on their own, without any more advertising.

How frequently are they repurchasing is still an open subject. Do they regularly purchase? most likely not Do they shop weekly? Maybe. once a month? More probable

You can avoid speculating on it. The easiest thing to do is to compute the average retention window by looking at your past data. As these various timeframes could be extremely far apart, this will require some statistical acumen, but once you have a reasonable average, you can easily build up your cohorts using that number. For instance, if you notice that your retention changes daily, your cohorts should also be segmented daily. Having cohorts for the first, second, third, and fourth quarters is OK if your consumers only return once every three months.

Measuring Retention Using Cohort Analysis is an essay I wrote on my blog that goes into greater detail and does it from a more analytical angle.

How can I go about doing that?

To utilise cohort analysis, follow these steps:

utilising several third-party tools (eg. Mixpanel, Kissmetrics, Google Analytics, etc.)
using several visualisation tools, such as Tableau, to extract your raw data into an Excel or Google Spreadsheet, or by setting up a script to perform the computation on your data server automatically (if you’re interested, here’s an article of mine about SQL for Data analysis).
Conclusion: Your choices and business
Three key ideas from this essay are as follows:

Check retention. And while you’re doing it, pay attention to your sales figures.
Utilize cohort analysis as a method of analysis.
Always exercise good judgement! Instead of following market benchmarks, consider your own numbers and make decisions!

If you’re measuring properly, you’ll have a far better understanding of your loyalty programme and will find it much easier to measure and optimise it. The information can also be used to enhance customer satisfaction polls.

Interested in concrete outcomes? Download the case study about one of our customers, LUISAVIAROMA, where we describe how the fashion retailer increased revenue by €16 million in just two years.