“Sources, Summary” report

The summary report contains information on all the sources that attracted users to your website.

To view the report: Reports → Sources → Sources, summary.

An example of this report is available for the Yandex Metrica demo tag.

  1. Ways to use this report
  2. Report structure and settings

Ways to use this report

Pinpoint your traffic sources

The report shows the sources that attracted users to your website.



Find out how many users know your website address by heart

The percentage of the following traffic sources may indicate that users are familiar with your site:

Source Description
Direct traffic

Yandex Metrica attributes traffic to this source in the following cases:

  • The user typed your site URL in the browser's address bar.
  • The user opened your site from their browser bookmarks.
  • The user clicked a link added to site tiles in a new browser tab (for example, in Yandex Browser).
  • The user transitioned from a site that uses the HTTPS protocol to a site with the HTTP protocol.
  • The traffic source was not transmitted to the referrer.
Cached page traffic
Yandex Metrica attributes traffic to this source in the following cases:
  • The user opened your website by clicking a locally saved link (the session might have the referrer file:///С:/Documents/User/page.html).
  • The landing page domain contains localhost or 127.
Source Description
Direct traffic

Yandex Metrica attributes traffic to this source in the following cases:

  • The user typed your site URL in the browser's address bar.
  • The user opened your site from their browser bookmarks.
  • The user clicked a link added to site tiles in a new browser tab (for example, in Yandex Browser).
  • The user transitioned from a site that uses the HTTPS protocol to a site with the HTTP protocol.
  • The traffic source was not transmitted to the referrer.
Cached page traffic
Yandex Metrica attributes traffic to this source in the following cases:
  • The user opened your website by clicking a locally saved link (the session might have the referrer file:///С:/Documents/User/page.html).
  • The landing page domain contains localhost or 127.
Evaluate the quality of traffic sources based on the bounce rate

The higher the bounce rate, the less the site is engaging users (bounces are users who leave within 15 seconds of looking at a single page). For instance, a high bounce rate for search engine clicks may indicate that your web pages are not relevant to users' search queries. For more information, see the Search queries report.



Learn which sources produce more converted users
To get this data, first create a goal and use it in your report.
Note. When analyzing converted sessions and users, you have to consider the attribution model.


Find out which sources bring in a certain level of revenue
Note. This requires transmitting e-commerce data to Yandex Metrica.
The report will display traffic sources that generated revenue equal to, more than, or less than the amount you specify.
  1. Find the for people with condition group and click .
  2. Select the condition Ecommerce → Total revenue.
  3. Select More, Less, or Equals and enter the amount.


Find out the average order value each source generates
Note. This requires transmitting e-commerce data to Yandex Metrica.
To find out the average order value, add a special metric to your report:
  1. Click the Metrics button.
  2. Select Average revenue from purchase.


View traffic sources for a specific page on your website

To do this, select the segment with the URL of the specific page:

  1. Find the Sessions that condition group and click .
  2. Select the condition Behavior → Landing page.
  3. Specify the webpage URL.


Report structure and settings

Each session has its own source. Yandex Metrica organizes the sources into groups:

Source Description
Direct traffic

Yandex Metrica attributes traffic to this source in the following cases:

  • The user typed your site URL in the browser's address bar.
  • The user opened your site from their browser bookmarks.
  • The user clicked a link added to site tiles in a new browser tab (for example, in Yandex Browser).
  • The user transitioned from a site that uses the HTTPS protocol to a site with the HTTP protocol.
  • The traffic source was not transmitted to the referrer.
Link traffic Users visited your site by clicking a link on another site. The page URL hosting the link is saved as the referrer.
Search engine traffic Users clicked a link to your site in search results. Yandex Metrica can detect the majority of search engines. Yandex Metrica stores the name of the search engine used, usually along with the search query.
Traffic from social networks Users followed a link posted on a social network. Most social networks identify the user who posted a link or the community it was posted by.
Ad traffic Traffic generated by Yandex services (Yandex Direct, Yandex Market), media platforms, and tagged links (such as UTM, Openstat, or Google Ads). For more information, see the Ad systems report.
Messenger traffic
Clicks on links posted in messengers are identified by the referrer and UTM tags.
Details
Yandex Metrica collects data on the following messengers:
  • Skype
  • Telegram
  • Viber
  • WeChat
  • WhatsApp
Use these UTM tags to identify your source of traffic:
  • utm_medium (required) with the value messenger (recommended) or social.

  • utm_source. The tag is used to identify the messenger. For example, utm_source=viber. If there is no tag in the URL, the tag value is omitted, or the value isn't recognized by Yandex Metrica, then the source will be labeled as Other messenger: identified by tags.

    Full tag list:
    • skype for Skype
    • telegram — for Telegram
    • viber — for Viber
    • wechat — for WeChat
    • whatsapp — for WhatsApp
Recommendation system traffic
Clicks on links in news feeds, personalized recommendation lists, and similar sources. Does not include data from traffic exchange networks and news aggregators (News, Rambler News and Google News).
Details
Yandex Metrica recognizes the following recommendation systems:
  • MirTesen.
  • Rambler.
  • Zen.
  • Flipboard.
  • Google Discover. It does not take into account click-throughs on devices with the iOS operating system.
  • Opera Personal News
  • Pulse (Mail.ru).
  • Sony News Suite
  • Toutiao.
Cached page traffic
Yandex Metrica attributes traffic to this source in the following cases:
  • The user opened your website by clicking a locally saved link (the session might have the referrer file:///С:/Documents/User/page.html).
  • The landing page domain contains localhost or 127.
Internal traffic If the user stopped doing anything on the site and the session timeout expired but the browser was still open, when the user later resumes activity Yandex Metrica registers a new session with “internal traffic” as the source.
Note. For accurate detection of internal traffic, make sure that the correct site URL is shown in the tag settings (in the Tag section).
Mailing traffic
The user clicked a link in an email. This includes:
  • Clicks on links. Detected using UTM, Openstat, and from tags.
  • SMS messages and push notifications. Detected using UTM tags. At the same time, utm_source must have the value sms or push, and utm_medium must not be empty (can have any value).

For more information, see How to correctly create a tag.
Click-throughs via QR codes

Traffic to the site via QR codes is detected using UTM tags.

For more information, see How to correctly create a tag.
Undefined

Undefined means that some of the session characteristics couldn't be identified, although the session itself was registered in the system. For instance, Yandex Direct traffic may have Yandex Direct: Undefined listed as the source. That means that Yandex Metrica recognized the ad platform, but couldn't identify details such as the campaign, keyword, and so on.

Why sessions aren't recognized

There are many reasons why sessions might be “undefined” in Yandex Metrica. For instance, if the source of Yandex Direct traffic appears as Yandex Direct: Undefined, this could mean that there was a lag between generating the yclid tag in Yandex Direct and registering the session in Yandex Metrica. This can happen in the following situations:

  • The user accessed the page by clicking an ad, left the tab open, and later refreshed the page.
  • The user clicked a link from an ad, but then forwarded it to someone else. The repeat session using this link won't be associated with the click from Yandex Direct, either.
  • Redirects on the site may cause yclid tags to disappear from the link and lose the referrer.
Source Description
Direct traffic

Yandex Metrica attributes traffic to this source in the following cases:

  • The user typed your site URL in the browser's address bar.
  • The user opened your site from their browser bookmarks.
  • The user clicked a link added to site tiles in a new browser tab (for example, in Yandex Browser).
  • The user transitioned from a site that uses the HTTPS protocol to a site with the HTTP protocol.
  • The traffic source was not transmitted to the referrer.
Link traffic Users visited your site by clicking a link on another site. The page URL hosting the link is saved as the referrer.
Search engine traffic Users clicked a link to your site in search results. Yandex Metrica can detect the majority of search engines. Yandex Metrica stores the name of the search engine used, usually along with the search query.
Traffic from social networks Users followed a link posted on a social network. Most social networks identify the user who posted a link or the community it was posted by.
Ad traffic Traffic generated by Yandex services (Yandex Direct, Yandex Market), media platforms, and tagged links (such as UTM, Openstat, or Google Ads). For more information, see the Ad systems report.
Messenger traffic
Clicks on links posted in messengers are identified by the referrer and UTM tags.
Details
Yandex Metrica collects data on the following messengers:
  • Skype
  • Telegram
  • Viber
  • WeChat
  • WhatsApp
Use these UTM tags to identify your source of traffic:
  • utm_medium (required) with the value messenger (recommended) or social.

  • utm_source. The tag is used to identify the messenger. For example, utm_source=viber. If there is no tag in the URL, the tag value is omitted, or the value isn't recognized by Yandex Metrica, then the source will be labeled as Other messenger: identified by tags.

    Full tag list:
    • skype for Skype
    • telegram — for Telegram
    • viber — for Viber
    • wechat — for WeChat
    • whatsapp — for WhatsApp
Recommendation system traffic
Clicks on links in news feeds, personalized recommendation lists, and similar sources. Does not include data from traffic exchange networks and news aggregators (News, Rambler News and Google News).
Details
Yandex Metrica recognizes the following recommendation systems:
  • MirTesen.
  • Rambler.
  • Zen.
  • Flipboard.
  • Google Discover. It does not take into account click-throughs on devices with the iOS operating system.
  • Opera Personal News
  • Pulse (Mail.ru).
  • Sony News Suite
  • Toutiao.
Cached page traffic
Yandex Metrica attributes traffic to this source in the following cases:
  • The user opened your website by clicking a locally saved link (the session might have the referrer file:///С:/Documents/User/page.html).
  • The landing page domain contains localhost or 127.
Internal traffic If the user stopped doing anything on the site and the session timeout expired but the browser was still open, when the user later resumes activity Yandex Metrica registers a new session with “internal traffic” as the source.
Note. For accurate detection of internal traffic, make sure that the correct site URL is shown in the tag settings (in the Tag section).
Mailing traffic
The user clicked a link in an email. This includes:
  • Clicks on links. Detected using UTM, Openstat, and from tags.
  • SMS messages and push notifications. Detected using UTM tags. At the same time, utm_source must have the value sms or push, and utm_medium must not be empty (can have any value).

For more information, see How to correctly create a tag.
Click-throughs via QR codes

Traffic to the site via QR codes is detected using UTM tags.

For more information, see How to correctly create a tag.
Undefined

Undefined means that some of the session characteristics couldn't be identified, although the session itself was registered in the system. For instance, Yandex Direct traffic may have Yandex Direct: Undefined listed as the source. That means that Yandex Metrica recognized the ad platform, but couldn't identify details such as the campaign, keyword, and so on.

Why sessions aren't recognized

There are many reasons why sessions might be “undefined” in Yandex Metrica. For instance, if the source of Yandex Direct traffic appears as Yandex Direct: Undefined, this could mean that there was a lag between generating the yclid tag in Yandex Direct and registering the session in Yandex Metrica. This can happen in the following situations:

  • The user accessed the page by clicking an ad, left the tab open, and later refreshed the page.
  • The user clicked a link from an ad, but then forwarded it to someone else. The repeat session using this link won't be associated with the click from Yandex Direct, either.
  • Redirects on the site may cause yclid tags to disappear from the link and lose the referrer.

The report supports all settings available in Yandex Metrica.

Select a question to find a solution.

No data or not enough ad data
Data differs from statistics in another system
Internal traffic increased
No traffic from a particular site
Problem getting data on UTM labels

Yandex Metrica receives information about different advertising systems from labels, which are parameters in the ad's link. If an ad system doesn't add a special label to your link, you can manually form the link using the UTM or Openstat label. For the complete list of values, see the "Advertising systems" report section.

There could be several reasons for this.

Imagine the path from the user clicking the link to the Yandex Metrica report:
  1. Creating the label for the link.
  2. Clicking the link and through to the site.
  3. Loading the code snippet for the tag on the site.
  4. Collecting data for the label and analyzing reports.
As a result, there are many steps between the first and last stage where data could be lost.
Imagine the path from the user clicking the link to the Yandex Metrica report:
  1. Creating the label for the link.
  2. Clicking the link and through to the site.
  3. Loading the code snippet for the tag on the site.
  4. Collecting data for the label and analyzing reports.
As a result, there are many steps between the first and last stage where data could be lost.
Reasons at the label creation stage
  • Labels aren't formed correctly: required values are missing, parameter names are incorrect, there are superfluous question mark symbols or spaces, and so on.
  • When copying ads from one ad system to another, the labels weren't changed, so the traffic is linked to another ad system.
Reasons at the click stage
  • The user stopped the page from loading and closed the page before it opened. This could be an accidental click on a mobile phone or tablet.
  • The page didn't load due to problems with site accessibility or it loaded incorrectly in a certain browser.
  • The ad specifies a URL that redirects to a page that doesn't have the Yandex Metrica tag installed.
  • The ad specifies a URL that redirects to another page, and the UTM labels are lost.
  • When the user goes to the site, other scripts that are loading block the tag from working.
  • The ad specifies an invalid URL that doesn't open a page.
Reasons at the code snippet loading stage
  • The Yandex Metrica tag isn't installed on landing pages or it's installed incorrectly.
  • The tag didn't load on the site because the user's browser (or the corporate proxy server) has a plugin that prohibits downloading tracking tags.
  • The user's antivirus system blocks the tag from loading.
Reasons at the report analysis stage

If you are comparing reports for Yandex Metrica and another system or ad management account, data might differ. This is due to the fact that services have different ways of registering a user's interaction with a site.

For example, in the ad manager, a click might be registered for going to the site as well as for opening details about an ad or going to a social media page. Yandex Metrica only counts the users who go to the site.

You might also have different attribution models chosen in ad managers.

See the additional information and tutorial video in the section About the Labels report group.



Yandex Metrica and other web analytics systems and ad managers have different ways of registering a user's interaction with a site and calculating statistics. For example, in the ad manager, a click might be registered for going to the site as well as for opening details about an ad or going to a social media page. Yandex Metrica only counts the users who go to the site.

Also check that:

  • You are comparing correct data for the same time period in Yandex Metrica and the other system.
  • Last click attribution is used by default in Yandex Metrica reports. If necessary, change the attribution settings in Yandex Metrica to make them match the attribution settings in your system.
This might happen for the following reasons:
  • Excessively narrow filters are set in Yandex Metrica in Settings → Filters. As a result, not all of the data is collected.
  • Yandex Metrica filters robots out of statistics. It's possible that the other system included robot traffic in their statistics. You can check for robot traffic on the site in Reports → Monitoring → Robots.
  • The tag didn't load on the site because the user's browser (or a corporate proxy server) has a plugin that prohibits downloading tracking tags. Antivirus systems with high security might also prevent tags from loading.

Also make sure that the Yandex Metrica tag is installed on all the domains and subdomains that the other system is collecting statistics for. To see data for domains that have the Yandex Metrica tag installed, go to Reports → Content → Popular.


This might happen for the following reasons:

  • Robot filtering is turned off in Yandex Metrica in Settings → Filters. As a result, the reports show data for all the robot traffic on the site.
  • The other system has overly strict filters defined.
  • The other system's tag isn't installed on all the domains, including mobile versions of the site.


This might happen for the following reasons:

  • One of the site's landing pages doesn't have the code snippet embedded on it.
  • The site might have JavaScript code that runs before the snippet and performs an internal redirect on the site. For example, this could be code for A/B testing. You can detect this code by a redirect on the site (this is usually a redirect to the same page on the site with extra parameters added to the URL). To solve this issue, contact your webmaster.
  • Your hosting provider might have enabled protection from DDoS attacks. This could add redirects before going to the site. Contact your hosting provider and ask them to disable this protection.

Choose the protocol your site uses:

In this case, you won't see the traffic source if this source uses a secured HTTPS connection.

  • You can switch your site to a secure HTTPS connection. You can get an encryption certificate for free through the Let's Encrypt project, which is a non-profit certification center supported by many prominent representatives of the internet industry.
  • If you can edit links on the source website, add the utm_referrer label to them and specify the source site as the value. For more information, see About the Labels report group.

The source site might be encrypting the referrer. In this case, you won't be able to solve the problem within your site or Yandex Metrica.

Add the utm_referrer tag to the website's links and specify the source site as the tag's value.


Imagine the path from the user clicking the link to the Yandex Metrica report:
  1. Creating the label for the link.
  2. Clicking the link and through to the site.
  3. Loading the code snippet for the tag on the site.
  4. Collecting data for the label and analyzing reports.
As a result, there are many steps between the first and last stage where data could be lost.
Reasons at the label creation stage
  • Labels aren't formed correctly: required values are missing, parameter names are incorrect, there are superfluous question mark symbols or spaces, and so on.
  • When copying ads from one ad system to another, the labels weren't changed, so the traffic is linked to another ad system.
Reasons at the click stage
  • The user stopped the page from loading and closed the page before it opened. This could be an accidental click on a mobile phone or tablet.
  • The page didn't load due to problems with site accessibility or it loaded incorrectly in a certain browser.
  • The ad specifies a URL that redirects to a page that doesn't have the Yandex Metrica tag installed.
  • The ad specifies a URL that redirects to another page, and the UTM labels are lost.
  • When the user goes to the site, other scripts that are loading block the tag from working.
  • The ad specifies an invalid URL that doesn't open a page.
Reasons at the code snippet loading stage
  • The Yandex Metrica tag isn't installed on landing pages or it's installed incorrectly.
  • The tag didn't load on the site because the user's browser (or the corporate proxy server) has a plugin that prohibits downloading tracking tags.
  • The user's antivirus system blocks the tag from loading.
Reasons at the report analysis stage

If you are comparing reports for Yandex Metrica and another system or ad management account, data might differ. This is due to the fact that services have different ways of registering a user's interaction with a site.

For example, in the ad manager, a click might be registered for going to the site as well as for opening details about an ad or going to a social media page. Yandex Metrica only counts the users who go to the site.

You might also have different attribution models chosen in ad managers.

See the additional information and tutorial video in the section About the Labels report group.