Episode 41 - Identifying Churn Early Warning Signs
Tom Burrell
March 21, 2025
Did someone forward this to you? If so, click to subscribe
Welcome to episode 41 of the Retention Blueprint!
As a Retention consultant, it might seem strange to say that focusing on churn is over-rated.
Often, though, it can be too late if a customer has decided to leave.
This is why I always advocate a focus on retention, not churn.
Namely identifying moments of truth and adapting your processes and actions to help customers achieve their desired outcomes at those critical moments (see episodes 1 and 37 for more on Moments of Truth).
If you do this, the total volume of customers with a propensity to churn will drop significantly, assuming you have a good customer-product fit (see episode 20 on the Retention Hierarchy of Needs and customer-product fit).
Nevertheless, I often see brands approach the churn problem the wrong way.
The idea is to find customers with a propensity to churn and then give them some financial incentive to stay.
The problem with this is that your predictive models are only that, models.
You will invariably flag some customers who look similar to those with a propensity to churn but have no intention of leaving.
When you go out with your discount offer, the customers who take it can often be those without any desire to leave.
Then, when you measure churn reduction against the control cell, you can often see minimal impact, particularly if you factor in the value loss from customers who you unnecessarily gave the discount to.
In this episode, we explore how to approach this differently. By identifying churn early warning signs, you can optimise moments of truth and act on save strategies before churn becomes a reality, sometimes even before the customer even thinks about leaving.
📣 News: New Podcast Episode
This week I had the privilege of talking with Jessica Jantzen, the founder and leader of Inside CRM.
We dived into why CRM teams are undervalued and what to do about it.
You can watch the podcast or subscribe to the Inside CRM Newsletter below.
If you are a CRM marketer and you want to elevate your value, knowledge means power and its never to late to up-skill. Over 1,400 have people have taken my training programs, click below to find out more about my advanced CRM Marketing Strategy 2.0 course that launched in January.
ADVERT
Seeking impartial news? Meet 1440.
Every day, 3.5 million readers turn to 1440 for their factual news. We sift through 100+ sources to bring you a complete summary of politics, global events, business, and culture, all in a brief 5-minute email. Enjoy an impartial news experience.
📰 Top Story: Finding Churn Early Warning Signs
The reality is that churn often stems from multiple root causes. As a result, reducing churn does not rely on a single initiative, which is why the proactive discount-to-stay model can be ineffective.
Most customers with a real propensity to churn show pre-churn signals, often up to 60-90 days before they actually leave and sometimes even years before (Source: Bain).
Companies often blame the customer's exit on their last interaction with the brand, but the propensity to churn usually results from a series of interactions over time and builds to the point where the customer feels the need to break the relationship.
Because these signals build over time, they are often not detected in a churn model, and even if they are, the remedy is usually not a discount.
The key is to identify variables that contribute to churn realisation over time.
Then, use this insight to respond before active customers' propensity to churn rises.
Some examples of early indicators could be:
Behavioural indicators:
Reduced purchase/transaction frequency/volume/amount
Decreased usage frequency
Decreased usage time / shortened session durations
Reduction in key feature usage
Fewer interactions with new features/content
Declining engagement with loyalty programs
Shift to purchasing only discounted items
Decreased cross-category purchases
Lower order variety
Shift to lower-margin products
Abandoning saved items/wishlists
Decreased referral activity
Removing payment methods from the account
Removal of the app from the mobile device
Shift from premium to basic features
Reduced feature exploration behaviour
Using only legacy features (resistance to new offerings)
Decreased customisation of product/service
Switching from annual to monthly billing cycles
Service signals
Unsatisfactory experiential NPS ratings following a service interaction
Repeated service contacts
Change in support ticket tone or frequency
Declining satisfaction scores across touchpoints
Negative sentiment in open-ended survey responses
Reduced participation in feedback requests
Decreased willingness to provide testimonials
Sharing negative experiences on review platforms
Engagement with negative reviews from other customers
Declining satisfaction scores across touchpoints
Communication signals:
Reduced response rates to emails/communications
Decreasing click-through rates in emails (even without unsubscribing)
Reduced interaction with personalised recommendations
Less frequent app notifications opt-ins
Decreased response to in-app messages
Lower engagement with brand community features
Unsubscribes
Account management signals:
Delayed payments
Contract/billing inquiries
Downgrades
My account visits
Competitor-Related Signals
Requests for data exports/migration assistance
Inquiries about compatibility with competitor products
Comparing features with competitor offerings
Mentioning competitor names in support conversations
Of course, not all of these signals can be tracked to a remedial action for a specific customer. For example, an unsubscribe can be an early churn warning signal, but it is not immediately clear how to approach a customer who takes that action.
Therefore, sometimes, you must use the insight in the aggregate to consider how to reduce the signal frequency through reactive actions that affect the entire base.
For example, for an unsubscribe, you might consider improving the personalisation of your communications.
Ultimately, the key is creating a system that identifies these signals and connects them to appropriate interventions.
Start by analysing your historical churn data to identify the most predictive signals for your customer base.
Then create a hierarchy of signals based on their predictive power and actionability.
For each significant signal, develop a targeted intervention strategy.
Some signals require immediate individual outreach (like multiple service complaints), while others inform systemic improvements (like declining feature usage patterns).
Ultimately, the goal is not only identifying at-risk customers - it's understanding the underlying causes of their dissatisfaction and meaningfully addressing them to reduce churn and improve retention holistically.
Final Thoughts
The key to effective churn management is to focus on retention. This means supporting customers in their value realisation process and continuously helping them achieve their desired outcomes.
Most churn is predictable and preventable with the right early warning signals and practical remedial actions to respond to individuals or holistically.
If you reduce churn signal frequency, retention will improve.
Until next week,
Tom
P.S. What did you think of this episode? |
Do you need help with Customer Retention?
When you are ready, contact me to discuss consulting, my fast-track retention accelerator, courses, or training. Or if you are interested in sponsoring this newsletter, get in touch via [email protected]
Reply