Products and Usage Health

Before you can create Product Health rules, you need to send relevant data to Startdeliver.

This data should represent real usage activity — events that show the customer is engaging with your product or reaching key outcomes.
Examples include:

Usage data is usually sent via your product integration, API, or data connector.

In some cases, usage can also be represented by metadata fields instead of individual events.
For example, if a customer should always have at least three active projects, it is better to track that as a customer field such as Active Projects.
Usage data would only show how many projects were created or deleted, but not how many are currently live. A field gives you a live snapshot that can be used directly for health calculations.

When setting up Product Health, think of Products as health buckets.
Each bucket represents an area of value or functionality that you want to measure.

For example, you might have buckets such as:

Each of these reflects a part of your product that contributes to customer success.
Inside each bucket, you will later define specific usage metrics or goals that indicate healthy behavior.

In Startdeliver’s own setup, we track these as separate products:

The goal is not to track every possible action, but to capture the most meaningful indicators of product engagement and value realization.

Once your data is connected and you know what to measure, it is time to create your Products.
Each Product represents a module, feature, or functional area that you want to track.

When creating a Product, you will see four configuration menus.

Define what kind of product it is and how it will be included in health calculations.

Set what is considered Good, Fair, and Poor health.

Example:
If you are measuring Unique Days of Activity and define the goal target as 3–5 days:

Before setting targets, choose if the product is measured on a Customer or User level.

Even if usage data is tied to individual users, you can aggregate it to measure total customer activity.
For example, if all users combined at a customer have usage on 6 unique days, that customer’s health is considered Good.

Measure health per user instead of per customer.
For example, each user must have at least 6 active days to be considered Good.
The overall customer health is then calculated based on the percentage of users who have Fair or Good health.
For instance, a customer could be considered Good if 50 percent of its users meet that threshold.

You can also:

(Only visible if the product is based on usage data)

Here you will see all available usage event types and can connect the relevant ones to the product.
These are the usage types Startdeliver evaluates when calculating product usage health.

This allows you to fine-tune whether health is based on recent or longer-term usage.

Explore additional settings for more nuanced tracking:

Only count Users with this Product
Limit health tracking to specific user types. Ideal for admin-only or role-specific features.

User goal from a field value
Base goals on a custom field value instead of a fixed number, allowing flexible, dynamic targets.

No News is Good News
Use this option when lower numbers indicate good health, such as tracking downtime events or error logs.

No Fair
Apply this when only “Good” or “Poor” ratings are relevant, skipping the “Fair” category entirely.

Adjust for Recent Activity
Account for new customers who are still onboarding by adjusting health targets based on their activation date.

Each customer and user has one Usage Health value — Good, Fair, or Poor.
This represents the average health of all attached products, considering any product weights or exclusions.

Example:
If a customer has three products where two are Good and one is excluded, the average usage health will be calculated from the two active products only.