Measure, predict, and prevent churn

Uncover engagement trends and risks to intervene and reduce customer loss.

Analytics Challenge

Business Challenge

KPI Impact

Analytics Challenge

The challenge of calculating churn is that it’s based on human behavior and the driving factors impacting changes in the metric are often different each time you analyze it.

To uncover the key drivers behind churn, you need to be able to analyze all user activity across operational, behavioral, price points, product usage, and more. While it’s easy to assess that churn rate is changing, it’s difficult to quickly look across all these segments to analyze exactly which factors or combinations of factors are driving those changes in time to inform decision-makers.

Churn Rate

Churn is the percentage of users who cancel or stop using your service in a specific period of time.

Churn = (Users at start of period – User at end) / Total remaining users

Analytics Challenge

The challenge of calculating churn is that it’s based on human behavior and the driving factors impacting changes in the metric are often different each time you analyze it.

To uncover the key drivers behind churn, you need to be able to analyze all user activity across operational, behavioral, price points, product usage, and more. While it’s easy to assess that churn rate is changing, it’s difficult to quickly look across all these segments to analyze exactly which factors or combinations of factors are driving those changes in time to inform decision-makers.

Business Challenge

Measuring and predicting churn is important for optimizing revenue and engagement. Even a small increase in churn rate can have adverse effects on the business’s ability to grow. Worse, the effects of higher churn rates tend to compound.

While you could attempt to overcome churn by growing acquisition and engagement metrics, acquiring new customers tends to be more expensive than retaining them. Meaning, if you’re not calculating churn and the factors behind it, you’ll see higher CAC and reduced revenue.

By diagnosing your churn rate and why customers are churning, your business will have the opportunity to intervene and prevent churn in the first place.

KPI Impact

Most businesses know that it’s easier and more cost-effective to retain customers than to obtain new ones. Uncovering and addressing a churn trend that reduces churn by even a small amount can have a compounding effect as customers join (and stay).

Let’s imagine a subscription SaaS company with 10,000 customers has an average churn rate of 10% MoM. At an average of $70 per customer per month:

     10,000 customers = $700,000 MRR / 8.4 M ARR
     10% churn = -1,000 customers
     -1,000 customer = -$70,000 MRR / -840k ARR

But an analyst finds that customers who sign-up in Germany receive onboarding emails 8 hours later and have a higher rate of churn. When the marketing team intervenes, the churn rate drops to 8% MoM. This single change equates to:

     8% churn = -800 customers
     -800 customers = -56K MRR / 672K ARR

Factors to include when analyzing churn

To check every factor driving churn, track churn at the account or customer level. Each row should represent a unique customer, with one record per time period.

As you build your datasets, make your data as wide as possible. Start with the recommended fields below and add as many descriptive variables as you can to augment your analysis.

Schema Blueprint:

  • One row per customer
  • Metric column =
    Status (Active/Churn)

Fundamental

Account ID

Monthly Subscription Value

Txn_ Amount

Product SKU(s)

Product Category

Product Type

First Order Date

Last Order Date

Last Activity

Days Since Last Log In

Calendar Month

Customer Acquisition Date

Customer Age

Customer City, State, Zip

Customer Tier

Augmented (Optional)

Customer Engagement Score

Email Engagement

# Users on Account

Customer City, State, Zip

Transaction Type

Promotion

Loyalty Program

Marketing Channel

Account Source

Email Subscribed (Y/N)

Account Deleted (Y/N)

Account Deleted Reason

CSAT

# Customer Success Tickets

Schema Blueprint:
  • One row per customer
  • Metric column =
    Status (Active/Churn)
Fundamental

Account ID

Monthly Subscription Value

Txn_ Amount

Product SKU(s)

Product Category

Product Type

First Order Date

Last Order Date

Last Activity

Days Since Last Log In

Calendar Month

Customer Acquisition Date

Customer Age

Customer City, State, Zip

Customer Tier

Augmented (Optional)

Customer Engagement Score

Email Engagement

# Users on Account

Customer City, State, Zip

Transaction Type

Promotion

Loyalty Program

Marketing Channel

Account Source

Email Subscribed (Y/N)

Account Deleted (Y/N)

Account Deleted Reason

CSAT

# Customer Success Tickets

Upwork Logo

See how Upwork finds new paths to customer growth with Sisu.

Read the case study

More resources for augmenting your analysis

Fast Segmentation Analysis to Diagnose Changing Customer Behavior

In this walkthrough, we show how a data analyst at a large retailer can use segmentation analysis in Sisu to comprehensively analyze experiments and diagnose changing customer behavior.

Read more

Two new ways to answer why, faster: Text and segment analysis

Two new ways to answer why, faster: Text and segment analysis

Read more

Analyze churn with Sisu, free.

Must be valid work email.
Get a demo