---
metadata:
  - name: generator
    content: Diplodoc Platform v5.52.0
alternate:
  - https://yandex.com/support/varioqub/en/create.md
  - https://yandex.com/support/varioqub/ru/create.md
  - https://yandex.com/support/varioqub/tr/create.md
  - href: en/create.md
    type: text/markdown
    title: Markdown version
  - href: llms.txt
    type: text/markdown
    title: llms.txt
---
> **Documentation Index:** Fetch the complete configuration index at https://yandex.com/support/varioqub/en/llms.txt

# Creating an experiment

## Step 1. Go to Varioqub

Open your [Yandex Metrica](https://metrika.yandex.ru/) account, select a tag, and click **Varioqub** → **Experiments** in the sidebar.

{% cut "What it looks like" %}

![](_assets/exp.png){.scr-exp}

{% endcut %} 

## Step 2. Create a new experiment

On the experiments page, click **Create experiment**.

{% cut "What it looks like" %}

![](_assets/create-ex.png){.scr-exp}

{% endcut %} 

A page for setting up a new experiment opens.

## Step 3. Customize the experiment 

Give your experiment a name to distinguish it from others and add a description (optional).

Some settings are only available with certain plans. Experiment parameters that aren't available with your current plan are marked with the ![](_assets/lock-32.svg) icon. To learn more about different plans, see [Additional functionality](https://yandex.com/support/varioqub/en/paid-functionality.md).

Configure the settings in the sections available with your plan.

{% cut "Sections with settings" %}

<div class="first-col-04"> 

#|
|| **Section** | **Field** | **Note** ||
|| **Traffic distribution** | 

* Equal (default): Traffic is equally distributed among all experiment variants throughout the entire duration.

* <!--[Dynamic](create/traffic-distribution#dynamic)-->Dynamic: Traffic is automatically distributed among variants based on the selected conversion metric (available with the **Standard** and **Extended** plans). | 

{% cut "Screenshot" %}

![](_assets/create-ex-traf.png){.scr-exp}

{% endcut %} 
||
|| [**Conditions**](https://yandex.com/support/varioqub/en/create/conditions.md) | 

**Date range**: Duration of the experiment, from 1 to 90 days. | 

{% cut "Screenshot" %}

![](_assets/create-ex-conditions-1m.png){.scr-exp}

{% endcut %} 
||
|| ^ | 

**Audience share**: The percentage of users to be involved in the experiment, from 1% to 100%. | 

^ ||
|| ^ | 

**Pages included in experiment**: Website domain and URL path without a protocol (`http`/`https`). | 

You can set conditions using regular expressions and combine them with logical `AND` or `OR`.

{% cut "Screenshot" %}

![](_assets/create-ex-conditions-2m.png){.scr-exp}

{% endcut %} 
||
|| ^ | 

**GET parameters**: GET parameters that are separated from the URL path by a `?`. | 

^ ||
|| ^ | 

**Regions**: Regions included in the experiment. | 

They are determined by the IP addresses of site visitors.

If you want to exclude certain regions from the experiment, select them in the list and toggle the **Exclude selected regions** option.

{% cut "Screenshot" %}

![](_assets/create-ex-regions.png){.scr-exp}

{% endcut %}
||
|| ^ | 

**Platforms**: Devices on which you want to run your experiment. | 

They are determined by the User-Agent of site visitors.

{% cut "Screenshot" %}

![](_assets/create-ex-plat.png){.scr-exp}

{% endcut %}
||
|| ^ | 

**Custom parameters**: Custom parameters that can characterize site visitors. For example, by indicating that they're an important customer, have registered on your site, and so on. | 

This setting is only available with the **Standard** or **Extended** plan.

To take custom parameters into account in the experiment, make sure that your site sends them.

To apply a condition in the experiment, click **Add condition**.

{% cut "Screenshot" %}

![](_assets/create-ex-param.png){.scr-exp}

{% endcut %}
||
|| [**Targeting**](https://yandex.com/support/varioqub/en/create/conditions.md#targeting) | 

You can divide site visitors into groups by using custom user IDs. To do this, enable the **Use custom uid for usersplit** option and fill in the **Select a list of custom uids for the experiment** field below. | 

This setting is only available with the **Extended** plan.

{% cut "Screenshot" %}

![](_assets/create-ex-targeting.png){.scr-exp}

{% endcut %} 
||
|| [**Intersections with experiments**](https://yandex.com/support/varioqub/en/create/intersections.md) | 

**Select the experiment to exclude the intersection**: Specify which of the ongoing experiments shouldn't overlap with this experiment. | 

This setting is only available with the **Extended** plan.

{% cut "Screenshot" %}

![](_assets/create-ex-intersection-1m.png){.scr-exp}

{% endcut %} 
||
|| [**Metrics**](https://yandex.com/support/varioqub/en/create/metrics.md) | 

* **Main**: Main metric that should determine the experiment's success.

* **Advanced**: Depending on your plan, you can add between 1 and 10 additional metrics for a more in-depth analysis. | 

If you have either the **Standard** or **Extended** plan and select a conversion metric as your primary metric, you can also enable [Dynamic traffic distribution](create/traffic-distribution#dynamic).

{% cut "Screenshot" %}

![](_assets/create-ex-metrics-1m.png){.scr-exp}

{% endcut %} 
||
|| **Experiment type** | 

* [Visual editor](https://yandex.com/support/varioqub/en/create/types/visual.md)
* [Redirect](https://yandex.com/support/varioqub/en/create/types/redirect.md)
* [Flags](https://yandex.com/support/varioqub/en/create/types/flags.md)
* [Ad units](https://yandex.com/support/varioqub/en/create/types/ads.md)
* [Widgets](https://yandex.com/support/varioqub/en/create/types/integrations.md) | 

The method you'll use to modify the experimental version of the page.

{% cut "Screenshot" %}

![](_assets/create-ex-types-1m.png){.scr-exp}

{% endcut %} 
||
|# {wide-content}

</div>  

{% endcut %}

To check that you've set up everything correctly before launching your experiment, use the **Sample size calculator** and **Experiment testing** tools.

{% cut "About these tools" %}

{% list tabs %}

- Sample size calculator

  Use the sample size calculator to [evaluate the statistical significance](https://yandex.com/support/varioqub/en/create/tools.md#calculator) of the experiment with the given settings.

  {% cut "Sample size calculator" %}

  ![](_assets/create-ex-calc.png){.scr-exp}

  {% endcut %}

- Experiment testing

  Use experiment testing to [check](https://yandex.com/support/varioqub/en/create/tools.md#check) that the changes have been applied and display correctly.

  {% cut "Experiment testing" %}

  ![](_assets/create-ex-test.png){.scr-exp}

  {% endcut %}

{% endlist %}

{% endcut %}

## Step 4. Save the experiment

If you have configured the experiment and are ready to start it, click **Save and run**. The experiment will start on your specified date.

To save the settings without actually running the experiment, click **Save as draft**. You'll be able to return to this draft and run the experiment later.

Click **Cancel** to discard all unsaved changes.

To view a report on the results of a completed experiment go to **Varioqub** → **Experiments**, then click **View result** under that experiment's name. To learn more about reports, see [Experiment results report](https://yandex.com/support/varioqub/en/stat.md).


<!-- source: en/_includes/feedback/feedback.md -->
<details>
    <summary class="button">Contact support</summary>
    <div style="padding: 15px;
     margin: 10px 0;
     background: #FFFFFF;
     border-radius: 10px;
     border: 1px solid var(--yc-color-line-generic);">
     <iframe style="background: #FFFFFF;"
        height="700"
        width="100%"
        frameborder="0"
        src="https://forms.yandex.ru/surveys/1705/?iframe=1&lang=en">
    </iframe>
</details>






<!-- endsource: en/_includes/feedback/feedback.md -->