We have arrived. 💪🏁
It’s the last week of our customer health challenge, Burn the Churn! But fret not, there will be some exciting cooldown content and a never-before-seen piece of content coming your way soon – subscribe to make sure you don’t miss it! 😎
Now, let’s get back to it. 💪
We’re at the home stretch, working on our endurance. And just like your own health, investing in your customers’ health is a life(cycle)long commitment. That’s why you always need to evaluate and improve your customer health scoring models over time.
So what can you do to stay on top of your health scores?
Let’s take a look! But first, a quick reminder. 👇
Baseline, not benchmarks
At the beginning of the challenge, we talked about common pitfalls when building customer health scores, with “set it and forget it” often being the culprit behind unexpected churn. As your customer base grows and customer journeys change, you’re likely to notice discrepancies between health scores and behavior and you might experience downgrades or even unexpected churn. While a source of frustration, this is also a great opportunity to learn and adjust your health scores.
If you’re creating a customer health score for the first time, it’s pretty unlikely that you’ll nail it. In fact, getting it right – or as close to right as possible – will take time. Remember, you’re going for a baseline, not benchmarks when you’re starting out.
The hypothesis you create when you first build your health score model has to be tested and evaluated continuously to make sure you accurately can predict churn (and/or expansion).
But, when and how often should you evaluate your health scores?
5 factors that impact your customer health scores
The frequency of evaluating your customer health scores is dictated by a number of factors:
- The size of your customer base – Your customers and customer base evolves over time, your health scoring model needs to evolve with them.
- Changes to your product – Just like your customers, your product evolves over time, too. As your product grows more sophisticated and increases in complexity, it’s a good time to think about if there are any specific implications for your health scoring model, e.g. major feature releases or product overhauls.
- Changes in the customer experience (internal) – A revamped onboarding program? New adoption or retention strategy? Or perhaps you’re implementing a digital customer success program? These are all likely to impact health scores. Time to monitor customer health more closely.
- Changes in the customer experience (external) – You might implement a new CS team structure or the business might shift focus or priorities which won’t fit every customer and their needs. These are things that will all impact the customer experience.
- Increase in unexpected or “green” churn – Time to uncover the root cause (it’s often lack of self-evident value, which, to make matters more complex, typically doesn’t have anything to do with adoption.)
Rule of thumb: If you make any significant changes that impact the customer experience, either directly or indirectly, evaluate your health scoring model as a part of the process.
Review your customer health score data
It’s easy to overanalyze when it comes to customer health scoring models. You can make it as complex as you want, but we always advocate for keeping it simple. Especially when you’re starting out.
Your goal is clear: Uncover if your health scores are accurately indicating churn-risk.
Start the exercise by asking these questions:
- Did the customers who you expected to churn, actually churn? If so, why?
- Did the customers who you did NOT expect to churn, churn? If so, why?
Once you have those answers ask yourself this:
- What are the correlating metrics? Are those clear or do they need adjusting?
- Do those metrics weigh into your customer health scores? And if so, is the assigned weight the right weight?
- Did you identify any missing data points that should be included? (Psst! Engagement and community metrics will help you paint a better picture of customer health.)
This should give you a good idea of where and why churn is occurring.
Dark data and false positives
Now, in some cases, the real reason for churn will never be fully uncovered. Whether quantitative or qualitative, there’s always dark data in play.
Let’s say the right metrics weren’t established at the beginning of the customer journey. Over time, those same metrics are continuously measured and assumed to be valid indicators of health, when in fact – they’re not. Hello, dark data.
So what happens then? You get false positives. This is why you need to reevaluate your health scoring model and its related metrics regularly.
|“I recommend that CS leaders increase their investment in their development and maintenance of health scores. A true health score should be a representation of the customer’s sentiment. That sentiment is formed by all touchpoints with the vendor, not just those with the Customer Success organization. Be serious about health scores by assigning responsibility to people who have experience of working with data and customers.”
- Peter Armaly, Senior Director of Customer Success at Oracle
A good customer health score model constantly evolves
A good customer health scoring model should constantly evolve. Create a playbook that includes a cadence for when to review your customer health scores. Eventually, as your models get more sophisticated, you’ll have different evaluation cadences for different customer segments. This will make sure that no data points fall between the cracks and with that – make sure you stay ahead of churn.
🍏 April 5th: Join us for the unmissable grand finale of Burn the Churn
Join inSided's Remco de Vries and Cognite's Alex Farmer as they discuss all things customer health and how you can go from red to green (with a juicy twist) in one epic webinar event.
By Jo Johansson
Head of Content at inSided. Passionate about content ops, words and horses. Connect on Linkedin