> For the complete documentation index, see [llms.txt](https://developer.harness.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://developer.harness.io/feature-management-experimentation/use-fme/feature-management/setup/default-treatment.md).

# Set the default treatment

Learn when Harness FME serves the default treatment and how it differs from the default rule.

### Overview <a href="#overview" id="overview"></a>

The default treatment is returned by the SDK when targeting rule evaluation is skipped. This happens in the following scenarios:

* **The flag is killed:** If a feature flag is killed, the default treatment overrides the existing targeting rules and is returned for **all** customers.
* **The end user is excluded by limit exposure:** If you [limit exposure](https://developer.harness.io/feature-management-experimentation/use-fme/feature-management/targeting/limiting-exposure.md) for a flag, users outside the exposure percentage are not evaluated against the targeting rules and receive the default treatment.

If targeting rules are evaluated and the end user does not match any rule, the user receives the treatment defined in the default rule, not the default treatment.

{% hint style="info" %}
You can set any of the treatments in your targeting rules as the default. Harness recommends choosing the safest treatment for your customers as the default treatment (that is, *off*, *old version*) when beginning the rollout. However, when you complete the rollout, you may want to make the new experience the default in the case that feature is accidentally killed before being removed.
{% endhint %}

### How default treatments differ from default rule <a href="#how-default-treatments-differ-from-default-rule" id="how-default-treatments-differ-from-default-rule"></a>

It’s important to understand the difference between the default treatment and the default rule, especially when using limit exposure:

* When you call the SDK’s `getTreatment`, it is called for 100% of users who encounter the feature flag. Every user will receive some treatment, whether on, off, or something else.
* Limit exposure restricts the percentage of users who are subject to targeting rules and the default rule. For example, if you set limit exposure to 10%, then only 10% of users are evaluated against your rules. The remaining 90% will receive the default treatment.
* For scenarios with multiple treatments, such as `version1`, `version2`, and `unallocated` (which means no feature exposure), you might:
  * Set limit exposure to 10%, exposing 10% of users to targeting rules.
  * Configure the default rule to split that 10% between `version1` and `version2`.
  * Set the default treatment to `unallocated`, which 90% of users will get.
* Limit exposure helps control user exposure during rollouts and experiments, allowing gradual inclusion without impacting users outside the exposed percentage.
* In simple two-treatment cases, limit exposure and default treatment can sometimes produce equivalent outcomes.

## Troubleshooting

### Users receive the default treatment instead of matching a rule

If you [limit exposure](https://developer.harness.io/feature-management-experimentation/use-fme/feature-management/targeting/limiting-exposure.md) for the flag, users outside the exposure percentage are not evaluated against the targeting rules and receive the default treatment. Review the limit exposure percentage for the flag.

### All users receive the default treatment

If the flag is killed, the default treatment overrides the existing targeting rules for all customers. Restore the flag to turn the targeting rules back on. Go to [Use the kill switch](https://developer.harness.io/feature-management-experimentation/use-fme/feature-management/manage-feature-flags/use-the-kill-switch.md) for details.

### A user who matches no rule does not receive the default treatment

If targeting rules are evaluated and the user matches no rule, the user receives the treatment defined in the default rule, not the default treatment.

## Next steps

* Go to [Control treatment](https://developer.harness.io/feature-management-experimentation/use-fme/feature-management/setup/control-treatment.md) to understand when the control treatment is returned.
* Go to [Fallback treatment](https://developer.harness.io/feature-management-experimentation/use-fme/feature-management/setup/fallback-treatment.md) to configure the treatment an SDK returns on error.

{% @harness-feedback/feedback %}
