Calculating Customer Health Scores

For many of us, the mention of customer health scores fills us with nervousness, even dread.  This metric is rapidly becoming the panacea solution to all things customer success. But as a highly complex and business specific set of variables contribute to it, each business is grappling it’s way towards building their own customer health score.  Having switched from software to fin-tech in the past few months, I thought it might be worth sharing recent experience.  To start, let’s go back to basics.

What constitutes customer value?

 When a customer achieves value, the product fits their need and a habit is formed, driving customer retention and ultimately advocacy. Customer health score helps us understand objectively and subjectively our relationship with the customer.  The purpose of measurement can be varied: propensity to adopt your product or service OR propensity to renew.  If your product or service is relatively new / innovative / immature, you’ll probably be considering the former (grow your customer base).  If you need to shift the business from acquisition to renewal focus, I’d suggest the latter.

Where should the measure be tracked and by whom?

Ideally, you’ll have a way of tracking your customer health score in your CRM / customer success tool.  Its primary user would be customer success managers (CSMs) to drive their daily tasks, focusing on actions that deliver the required business actions (on-board, adopt, save, renew, nurture, etc.)  Additional use of rolled-up scores (by CSM, cohort, date range, campaign, tenure, etc.) could be developed. This can be linked to appropriate team bonuses, areas to fix or enhancements within the customer experience as well as shareholder / investor reporting.

Can the score be linked to stages of the customer journey / life-cycle?

 This depends on the sophistication of your “customer success tool”, i.e. that manual / automatic triggers link certain events to the current score.  For example, when a customer usage drops, flagging “nurture” stage event could link the score for preceding stage (adoption).  This will become very complex, so I suggest focusing on proving the concept works first, then linking to the headline stages of your customer journey.  Once you’ve got a basic model running (hopefully with the help of some data savvy colleagues), drill down to the sub-stages.  Learning how to respond to a red & green flag events at customer account level will help build a more planned approach to changing the customer experience.  This, in turn improves scores at each stage & sub-stage, as well as building the most effective responses to customer actions (positive & negative).

What primary segments might be in the proof of concept model?

  • High health score (80-100%) – profitable customers: high engagement & renewal probability (nurture activity normally championed by product team)
  • Medium health score (60-80%) – break-even customers: some engagement & reasonable renewal probability (nurture activity normally championed by marketing team)
  • Low health score (0-60%) – loss-making customers: early life customers, low engagement & high risk of churn (nurture activity normally championed by Success team)

What does each score bracket trigger for the CSMs?

Each business will need to build their own model, but here’s an example to get you started. In this example, the scale starts from 0 (customer dramatically scales back volume and usage, serious complaints and threat of competitors), but would typical be below 25% for an unhappy new customer who has low adoption, to 65% points for a happy customer who has high adoption, to 100% for a 4-year-old customer with high recurring usage at advocacy level.  The typical CSM activities per score range could be:

> 80% – advocacy, retention

50% – 79% – relationship management, retention

25% – 49% – on-boarding, adoption, nurture

< 25% – nurture, escalate (relationship management)

This is a rough example, but helps show the segmentation of the treatment flows based on health score – for example, you wouldn’t be trying to promote a new products to a customer who is at risk of churn, etc.  Customers can move into different treatment flows (customer journey process flows), dependent on their health.  The CSM is responsible for reacting to health score changes up & down and applying different treatment.

Notes on health score to consider:

  1. It’s a renewal rather than churn indicator
  2. It’s a balance between your product / service value to customer & value of customer to your business
  3. Mixture of (mainly) objective and subjective measures
  4. In most cases, it indicates customer health at point of sign up (not during sales cycle)
  5. Certain indicators can make score fluctuate wildly (e.g. non-payment, drop in usage)
  6. The importance of the indicators can vary (low NPS = bigger drop in score)
  7. Initial segmentation is simple to allow validation of methodology

Hope this brief overview helps you in getting started or progressing your customer health scoring.  At present, I’m considering the role of CLV (customer lifetime value) and sentiment scoring in building a more complex customer health score, so stay tuned for further insight in the coming weeks!



2 thoughts on “Calculating Customer Health Scores

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s