The definition of churn can be different depending on how your organisation is tracking it. However, the fundamentals are that someone, in most cases a customer, has stopped interacting with your organisation, whether that be cancelled payments or if they’ve become disengaged with your outreach.
A higher churn rate suggests that there’s dissatisfaction between customer expectations and the value that they are receiving. However, thinking of it from the alternative perspective can be equally as powerful, low churn suggests things such as product market fit, strong brand loyalty and effective customer engagement.
Understanding churn helps companies:
- Gauge the health of customer relationships
- Predict revenue and forecast growth accurately
- Identify friction points in the user journey
Why success can be defined through churn metrics
Although bringing in new business is vital, retaining your current customer base is more cost effective and scalable. This is why retention is the metric used to frame success and churn is the counter-metric.
Here are some churn metrics you may want to keep an eye on:
1. Net churn vs. gross churn
- Gross churn – focuses on the revenue lost from cancellations
- Net churn – Considers the total impact of both revenue lost and revenue gained from existing customers
A business with high gross churn but negative net churn may still be thriving.
2. Customer Lifetime Value (CLTV)
CLTV is directly affected by churn. The higher the CLTV the lower the churn rate, which suggests a greater profitability as well as long-term success.
3. Retention cohorts
Businesses can see how product changes or their onboarding can impact the different customer groups by tracking the user retention. The higher the retention across the groups often correlates with a stronger product-market fit.
4. Leading vs. lagging indicators
While churn is a lagging indicator, success comes from working out the leading indicators that will predict churn e.g. an engagement drop. Making sure that you act on these proactively is the key to success.
The first steps to tackling churn
Before embracing churn, your organisation must put down some foundations in place.
Status of your data?
In order to create a successful churn model, having a dataset that is consistently high-quality as well as being long range is essential. Prior to beginning modelling, your business must review how you’re accessing your data.
There are plenty of risks that might tarnish your data. For example, you might be missing key customer information or events, may not have been tagged it correctly, so are mislaid in your data house. To avoid these specific issues, you may look to implement clean and structured customer records, with mandatory fields so no data is missed. Or you might want to merge all your sources, so events or activity don’t get missed.
A churn model that is built on weak data will produce weak insights.
Assembling the right team
A successful churn programme requires knowledge and technical know-how.
Your organisation may want to build a team to tackle churn. Or you may want to bring roles in, such as data scientists who can develop the churn model, data engineers to prepare the data to move along the pipeline or domain experts who will be able to take any complexities or nuances within your business and change them into useable features.
Churn models need to be usable and actionable
Once you have a churn model in place, it needs to be integrated into daily workflows as well as your decision making.
By embedding the metrics that the churn model has produced into your CRM, your account management or customer success teams can act on the customers that are more likely to churn.
You can also build dashboards that can visualise any risk trends, so teams can react to reduce any chance of customer churn ahead of time.
Remember, a forecasting model that isn’t used just becomes very expensive shelf ware!
An activation plan is essential
Data insights on their own don’t retain customers, you need to act on those insights too.
Firstly, you need to define who is responsible for acting on churn information. For example, would it be your customer success team? You also need to put measurement structures in place – otherwise how will you know the impact you’re having?
Of course, you will also need to test and retest to see if the model and data is correct.
Want to build a churn model that works?
If you want to dive deeper into the questions that will allow you to set up a churn strategy that is ready for success, download our free whitepaper here.
This whitepaper will help your team align on what churn really means, help you assess the readiness of your data, as well as build a churn model that results in actions and not just reports.
