How 7 Companies Calculate their Customer Health Score

Early on in startups the idea of creating a ‘customer health score’ is met with blank stares. Later we realise it's a business survival score that we should look at like we're in Tokyo checking on Godzilla's current location. Below we cover the key criteria that 7 companies use in calculating their customer health scores.

Related: Our List of Tools for Customer Success Managers

Also see: The Honest Truth of How to Sell SaaS to Enterprise

What is a customer health score?

"A customer health score is a value that indicates the long-term prospect for a customer to drop off or, conversely, to become a high-value, repeat customer through renewal or cross-selling or up-selling strategies." Source

Calculating customer health scores is becoming default for SaaS companies along with the rise of customer success as a core philosophy.

Key lessons learned from gathering this list

A customer might cancel but the signs that they were going to cancel come months before.

Some companies have very advanced customer health scores though often a core few factors do 80% of the work.

Get input from your customer success managers because you need to add the empathetic art to the science

You can add 100s of factors if you have the time and resources to do so, including machine learning to advance it further.

Basing it on the customer journey is a nice way to anchor the mind.

You have the normal rating factors like Red, at risk etc but also add one that identifies companies as 'Champions' to provide a new perspective on your health sheet.

1. Upscope's own simple ranking system based on the core factor of usage


Upscope is no-download screen sharing used by teams to effortlessly walk new users through a site's features so they convert into long term buyers.

Upscope's health score is based on usage such as how many agents are using it, how many hours are they using it and how many customers they are screen sharing with.

Even with a simple system, it needs a lot of care as you'll see below.

The main factors for working out the health score

  1. Total number of seconds they used it in the last month.
  2. How much they're spending per second (if they're spending 600 dollars per month and using it 1 hour, that's a problem).
  3. Number of people they screen share with.
  4. Amount of time the top 20% of a companies agents use it per month.
  5. The number of agents that are using it.
  6. The % of agents that used it the previous month that are still using it this month.
  7. How much of the total usage is made by the bottom 50% of agents.
  8. How much of the total usage is made by everybody but the top agent (in case someone has 10 agents and 1 is using it 60 hours per month and everyone else is using it 10 mins).

In the image above the left side is the change in the score between the current month and the previous month based on all factors above.

As each factor is measured in different units (some are in seconds, some in %) they're put on a scale between between 0 and 100 with 100 being assigned to the top 5% (so outliers with massive usage don't put everyone else at zero).

Then they have weighting based on importance so some factors are multiplied by the weighting.

Then they are all summed together then again put on a scale between 0 and 100 with 100 being the top 5%. The top 5% of teams will always be 100.

On the right hand side is a much simpler indicator which is the total number of screen sharing hours for the whole team. This is marked as stable, positive or negative.

There's also an option to filter the list by paying, trialling, cancelling and all.

2. Assignar divide it between product and relationship


Rachel Jennings at Assignar recommends the following to identify those at risk.

Define sub-categories that mean a customer is at risk

  1. Reduction in usage
  2. No/small usage of key features
  3. Too many or no support conversations
  4. Low survey scores
  5. Sudden stop in references/speaking engagements
  6. Overdue invoices
  7. A combination of the above

Define actions to measure for each of the above sub-categories

Then define specific actions to measure like logins and customer interactions for each of these.

Place them into either of 2 main categories labelled Product or Relationship

Product makes up 60% of the total and Relationship 40%.

Score it

Each sub-cateogry is scored out of 5. 1 is danger zone and 5 is healthy zone.

See more on Rachel Jennings calculated health scores at Assignar

3. David Lahey says to map out your customer journey

There's a nice simple flow to how David recommends creating your health score and the step of mapping your customer journey is central to that.

  1. If you are not confident in your data don’t proceed with this exercise until you are.

  2. Segment your data by importance
    Once ready, segment your customer data by Stage, Size, and/or Location e.g:
    Prospect, trial, deployed.
    Small, medium, enterprise.

  3. Get your criteria to measure by mapping your customer journey to return on investment e.g. login, open rates on first email inviting users through to whatever you think actual ROI is for your customers.

  4. Ask your product engineering teams to get the data showing usage, feature adoption etc and match it to the criteria above. Put it all into rows into a spreadsheet.

  5. Add the segments from step 1 to the sheet and then crunch the numbers to determine green, yellow, red.

See how David Lahey recommends applying the green-yellow-read in full

4. Pendo adds 'Champion' to the health scoring categories


Pendo uses Breadth, Depth & Frequency (BDF) to measure a customer’s health but what we like is the addition of 'Champions'.

Identifying champions as much as simply risky or healthy accounts helps marketing, customer success, sales and other teams identify and build new channels and processes that find and convert more champions.


See more on Pendo's BDF customer health breakdown

5. Moe Nada shows you a scoring system that can be used for physical products

This formula concludes the health of the account into four main categories as following:


Each category has set of sub-categories like stability, outage, integrations, enhancements, maturity, adoption, earliness and more.

Sub-categories are all weighted individually and overall health score is evaluated as follows:

Healthy: if the score is more than or equals to 7.5

Infected: if the score is more than 5 and less than 7.5

Sick: if the score is equals to or less than 5

See more on the customer health score for more physical products

6. UserIQ monitors 5 key churn indicators

UserIQ’s customer health dashboard monitors five key variables called “churn indicators” that are used to calculate Health IQ.

  1. Login activity is based on frequency of logins
  2. Feature Adoption is based on the number of unique features that each account is using within the product.
  3. Sentiment is tracked through Net Promoter Score (NPS) surveys
  4. Technical Support is calculated based on the number of support tickets that are fulfilled and closed by an account or customer.
  5. Financial Health is measured by a customer’s monthly rate, whether payments are made on time or delayed, and the validity of the account’s credit card

To truly understand the health score the users need to be put into segments. The post goes into greater detail on that.

See more on UserIQ's customer health score churn indicators

7. Asana's example shows you the work it takes to build a serious heavy duty, machine learning based customer health score


We're not even qualified to comment and breakdown this article but in short, Asana iterated the process and by the end of it included a myriad of factors that most companies don't consider.

Read the article if you want to understand the work a big company puts into iterating a large process in-depth.

Account Health Score (AHS)

Asana first created a 'simpler' health score.

"Asana equipped customer success with what they need to successfully reach out to unhealthy accounts by iterating and expanding their model.

"The Account Health Score (AHS) is that metric, and it does exactly what its name suggests; it’s a measure of how well a team on Asana is doing. This value on a 100-point scale is computed by combining a short list of engagement metrics."

Account Health Score 2 (AHS2)

"When the AHS was created, the data program at Asana was in its infancy. Since then the available data has expanded tremendously.

"Once limited to visits and records of core actions, we now have data about things like how many employees work at the companies that pay for Asana, whether the account’s users are concentrated within a single geographical region or spread across many, and how reliably payments are made."

See Asana's full post on data science and health scores

Also see 12 Questions Every SaaS Company wants Answers for

How 7 Companies Calculate their Customer Health Score
Share this

Subscribe to Upscope Blog!