Segmentation examples

Selecting users who viewed specific site pages (for example, the product catalog in one of the categories), in order to use Yandex Direct to offer these users a product or service related to these pages.

1. In the for people with condition group, click .
2. Choose the condition: Behavior → View URL. Enter the URL for the page (for example, https:example.com/tariff/). Use @example.com to segment a report for all pages that include example.com.
3. Save the resulting segment.
4. In Yandex Direct, add a retargeting list.

View an example of a selected segment for a Yandex Metrica demo tag.

If we take the data from the Sources, Summary report, we can select the information about users who went to the site and viewed a page with a promotional offer from March 1 to March 15.

1. In the for people with condition group, click .
2. Choose a condition: Behavior → View URL. Enter the URL for the required page (for example, www.example.com/promo.html). Use @example.com to segment a report for all pages that include example.com.
3. Choose the time period for pageviews: click For reporting period, then Choose period.

View an example of a selected segment for a Yandex Metrica demo tag.

In the Sources, Summary report data, we can select users who completed an order for an amount over 10,000 rubles.

1. Find the Sessions in which condition group and click .
2. Set a condition: E-commerce → Purchase amount, then select More and enter 10,000 as the order total.

Sometimes, a high volume of internal traffic on a site can indicate that the Yandex Metrica tag is missing on some pages. Let's look at an example.

A user followed a link from an external source to a page that doesn't have a tag (for instance, contacts.html), then immediately (during the same session) navigated to a page with a tag (in this case, page1.html). This means that two conditions are met: the session starts from page1.html (because Yandex Metrica doesn't have data about the contacts.html pageview due to the absence of the tag on this page), but the page the user navigated from (contacts.html) belongs to the same site that is specified in the tag settings (http://mysite.example/).

To detect the page without the tag:

1. Find the report Content - Landing pages in the Sessions in which condition group and click .
2. Choose the condition: Sources → Last source → Source type → Internal traffic.
3. Only leave the Landing page dimension in the dimensions list. To do this, click Dimensions and remove the others.

This gives us information about the pages that the session started from. To determine how the user got to these pages, add a dimension with the referrer (source) that started the session:

1. Click Dimensions.
2. Select Sources → Referrer.

Now the report shows the pages that the session started from, and the second level of the report hierarchy shows the pages that clicks came from. For convenience, you can move the Referrer dimension to the top of the list of dimensions. Then the first level will show the pages that presumably might be missing the code snippet, and the second level will show where users navigated to from them.

The small number of sessions in this report indicates that everything is probably fine with the tag, and the internal referrals are caused by session timeouts.

View a similar report for a Yandex Metrica demo tag.

A mobile provider has added a calculator for service rates on its website.

To evaluate the effectiveness of the new tool, we need to compare statistics for users who used the calculator with those who didn't (for example, over the first 14 days after introducing the tool). It will also be helpful to evaluate traffic sources. Segmentation can help us do this. Let's assume that the tag already has a goal created for interactions with the rate calculator (“Calculator”).

In the Sources, Summary report, we'll select sessions for new users who completed the “Calculator” goal:

1. Find the Sessions in which condition group and click .
2. Choose the condition: History → New/returning user → New user.
3. Click .
4. Pull out sessions where the “Calculator” goal was completed: Behavior → Goals reached, and choose the goal from the list.

Then refine the data, leaving information about users who viewed the service rates pages with the calculator between March 1 and March 15. The rates pages are standardized on the site, and they all show the calculator. So the condition can specify the URL of the entire “tariff” directory by using the * symbol.

1. In the for people with condition group, click .
2. Choose the condition: Behavior → View URL. Enter the URL of the directory with the service rates.
3. Choose the time period for pageviews: click For reporting period, then Choose period.

This gives you statistics on site users who used the rates calculator during a set time period in the past. Save the segment.

Now let's get average data for site users. Delete the Goals reached condition and save the segment with a different name. Now you can compare the two segments.

You can exclude data with a brand mention from the report. To do this, use segmentation and add a condition in the format!your_brand for the Russian and English versions of the brand, where:

• ! at the beginning — negates the condition.
• at the beginning — matches a string (case-insensitive).

For example, for the company “Alice in Wonderland”, a regular expression may look like !алиса в стране чудес for the Russian version of the brand and !alice in wonderland for the English version. Using this method, you can remove up to 20 phrases from the report. To exclude more than 20 phrases, use the regular expression !~(brand|бренд.рф|brand.ru).