> ## Documentation Index
> Fetch the complete documentation index at: https://docs.openlayer.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Evaluation and delay windows

> Learn about evaluation and delay windows in Openlayer

Evaluation and delay windows are part of every monitoring test. In this guide, we explain their use and highlight their differences.

You configure both under **Evaluation windows**, in the **Data** section of the test
configuration. See [Tests configuration](/tests/test-configuration) for the other sections of
a test.

## Evaluation windows

To evaluate Openlayer tests, a dataset is always required. While in development, the training and validation sets are the natural choices, for monitoring, **the dataset used by tests is defined by an evaluation window.**

<Tip>
  **What is the evaluation window?**

  The evaluation window defines the period used to accumulate data, and form the dataset used to evaluate a test.
</Tip>

The **Cadence** determines how often the window closes and the test runs:

* **Scheduled** runs the test on a fixed schedule, such as Daily. Each run uses the data
  published since the previous run.
* **Manual** runs the test only when you trigger it, with **Run test** from the test's
  overflow menu.

Let’s look at a concrete example to clarify the process:

1. Imagine we want to monitor Nulls on the feature `Age` for our production data. We would navigate to the test creation page and click the *Missing values* test.
2. In the **Data** section, we set the cadence to Daily.
3. After the test is successfully created, the platform starts accumulating production data being published.
4. Then, once the first day passes, it uses the data accumulated in that day, evaluates the number of missing values, and updates the test status.

Each test can have its own cadence. This is important because each value monitored has its peculiarities. For instance, longer evaluation windows can smooth out seasonal data. On the other hand, shorter windows can be appropriate if quick reactions to sudden changes are needed.

Two toggles change what each run looks at:

* **Include historical data in each run** widens every run to also cover data published
  before the current window, instead of only the data accumulated since the previous run.
* **Anchor evaluation windows** pins the windows to a fixed point in time. Turn it on and you
  set a **Start date** and a **Timezone** — for example, `Sep 1, 2026, 12:00 PM` in
  `America/Los_Angeles`. Every window boundary is then measured from that instant, so a daily
  test always closes at the same hour of the day you chose.

## Delay windows

We have seen that evaluation windows are defined to accumulate the data used by tests. Now, we will explore delay windows.

<Tip>
  **What is the delay window?**

  The delay window defines the gap between the test evaluation time and the end of the evaluation window.
</Tip>

The **Evaluation delay** toggle is off by default, because most tests can be evaluated at a regular cadence defined solely by the evaluation window. However, in some cases, a delay is also needed. For example, for performance tests, which usually require ground truths to be computed, a delay window is needed because the labels are not available at the same time as the data is published to the platform.