In his 1987 magnum opus about customer retention and churn, Rick Astley sang, “Never gonna give you up, never gonna let you down, never gonna run around and desert you. Never gonna make you cry, never gonna say goodbye, never gonna tell a lie and hurt you.”
To this day, it remains a solid philosophy with which to underpin a customer retention strategy. In short, do your level best to keep your customers, and keep them for the most part you will.
Of course, understanding Astley’s profound ideology is one thing, but achieving it is quite another. You only have to look at the work Starbucks puts in to maintain the lauded customer retention strategy that it does. The company’s strategy is based around a mix of personalisation, customer service and a loyalty rewards program. It has been developed over years and is honed continually. This isn’t to say that you should do what Starbucks does, but simply to illustrate the point that identifying what approaches will work for you and maintaining a top-drawer retention strategy are tasks in themselves.
Fortunately, they’re tasks that have been made much easier by data and, indeed, nigh on every company is sitting on exactly the data they need to maximise retention and minimise churn.
Of course, whether they’re using that data is a different matter. (Spoiler: they’re not).
Here’s a simple, two-step breakdown of the error many companies make when overlooking customer retention as a means of driving growth:
1. When companies want to grow, they typically focus on attracting more new customers
2. Attracting new customers costs more than retaining existing customers
Now we’ve got that cleared up, let’s explain how you can use your existing data to reduce customer churn, keep customers for longer and increase the lifetime value of these customers.
The first thing to know is that all sorts of data types can be used to improve retention and minimise churn. Some types of data are more valuable for these purposes than others, and the more relevant types of data you have the better, but almost every company has enough of the right types of data. That includes things like customer purchases, feedback, website usage and social media activity.
By pulling together a combination of relevant data types, it's then possible to spot patterns and trends that may suggest a customer is likely to leave. This means companies can intervene to stop customers leaving when it seems likely that they might (e.g. with an offer of some sort) and can tailor their service to reduce the aspects that may precipitate customers leaving.
At Peak, we apply a variety of modelling techniques to the combined data, including counting process-based survival, logistic regression, random forest survival analysis and recurrent neural networking. By combining these approaches, we can predict very accurately when any given customer is likely to leave and get an idea of why that may be the case.
What all this means in the long-run is happier, more satisfied customers, who are less likely to leave your business, will stay for longer and are more valuable. Why? Because you wouldn’t give them up, let them down, run around or desert them. You wouldn’t make them cry, say goodbye, tell a lie or hurt them.