Customer Retention is a Data Science: RFM
Customer Retention looks very different depending on the marketplace in which you find yourself, but the same data elements persist across industries and can be harnessed to transform your sales history into a superpower of data and actionable insights.
If you’re trying to attract customers for a commodity service such as dental check-ups, offering a calming three-wick candle may be the perfect shiny object. If you’re looking to become the go-to lunch pickup for a business team, your mobile app should be super simple, support group orders, and offer some kind of reward based on frequency of usage. If you’re selling thousands of widgets B2B for the purpose of downstream distribution, then (womp) neither of those are going to be especially compelling. In a B2B space, your customers do business with you based on their business needs.
In the absence of being able to offer free tranquility in a glass cube of wax, you need to know who to talk to and when. That’s where the data science *jazz hands* comes in…
Understanding the sequencing of customer behavior and being able to consistently and systematically read the tells of that behavior offers a powerful forecast of loyalty. In order to quantifiably create this forecast, a risk score can be calculated based on three behaviors.
- Recency: How recently has this customer purchased?
- Frequency: How frequently does this customer purchase?
- Monetary: How much money does this customer spend?
For all of the incredibly powerful data available in the marketplace (for a price, of course), there is immense power in the data that is readily available internally. By evaluating each customer’s RFM (Recency, Frequency, Monetary Spend) activity along standard timing intervals and across the portfolio, a risk score can be assigned to each customer and each behavior. The aggregate view of this risk score highlights the steadfast loyalists as well as those customers who are at risk or likely already lost.
Stratifying customers based on relationship health and retention risk is not enough, though. In order for the information to truly mean something, it must compel someone to do something. This may take a variety of forms depending on the particular technical and operational ecosystem:
- If sales reps are managing their book of business within a CRM, then these scores can be fed to the CRM so that they appear within the customer’s metadata.
- If inside sales teams are working call lists within a queuing system, customers within a particular (high) risk range can be added to those queues as a recovery exercise.
- If marketing teams are looking to encourage loyalty, customers in a certain (low to medium) risk range can be targets for loyalty rewards.
None of these are incredibly complex concepts, but its critical that these calls to action be incorporated into standardized processes so that each of them can be naturally integrated into daily operations, repeated at scale, and measured for efficacy.
ProfitOptics has ideated, solutioned, designed, built, implemented, and maintained these types of targeted solutions for our own customers for 15 years, and we’d love to talk to you about what is possible.
***************************************************************************
Sidenote: Its important to acknowledge that mobile apps and candles are also candidates for this type of data-driven operationalization. At ProfitOptics, we can help you build and monitor user behavior on your mobile app. And while we can’t manufacture your candles for you, we can help you monitor your candle-loving customer base. Reach out to us to talk more about what we can accomplish together.
