Traffic analysis in Yandex. Metrika: a practical guide

If the site receives inappropriate traffic, there are practically no sales and applications. Or each targeted action costs you a lot.

In Yandex.Metrika, you can study visitors by almost any characteristics: from which channels they come to the site, what devices they use, etc. In this article, we will show you which reports will help you in analyzing traffic in Yandex.MetriKa, how to build them and draw conclusions from them.

Let’s start with reports that help you understand how targeted users are visiting your site.

Where does your audience come from?
An analysis of traffic sources is needed if you bring both paid (from advertising) and free (from search results) traffic to your site.
You will need one of the standard Metrika reports:

Add to it additional indicators for achieving goals and conversions by clicking on this button:



If you have configured only actions that directly lead to making a profit (purchase, order) as goals, select the metric “Achieve all goals”.

If some of your goals in the Metrika are related to the intermediate stages of the sales funnel (visited a certain page, went into the basket, looked at N pages of the site), select the achievement of specific goals.

For example:

From the report you will get a general idea of ​​how useful certain advertising channels are for you.

Pay attention to conversions and behavioral metrics. The more interested the audience comes to you, the more time it spends on the site and more content it views (viewing depth).

The percentage of failures, on the contrary, can show inappropriate and “junk” traffic (but not necessarily – sometimes these are problems with the site, as a result of which visitors cannot quickly find the necessary information or solve their problems).

Accordingly, look at sites with high Time Spent On Site and Viewing Depth rates and low bounce rates.
How much does the targeted action cost?
The answer to this question is in the report “Direct – expenses”. It will be displayed in the menu Standard reports → Sources → Direct, if you have previously linked your Metrika and Direct accounts.

We leave the metric “Cost of clicks” and add all the conversion metrics associated with the target visitors to it. We choose metrics based on target visitors rather than goal achievement to avoid biases in results due to repeat applications.

From this report, you get conversions for each type of goal and costs by advertising campaigns:

Next, combine the results by ad groups in Yandex.Direct by adding the appropriate grouping:

This is necessary in order to know in which particular ad groups the conversions are sagging or they are too expensive. Disable groups where bids are expensive for you, even if the conversion is high.

If your conversion rate is low, find out the reason: it may be due to incorrect targeting, or it may be worth improving your ads.

Attention! One report can only contain 10 metrics, so if you have more of them, break it into several reports and copy it into one Excel document for further study of the big picture.
Further we will work with the standard report “Direct Expenses”, because when analyzing traffic it is important to know the cost of the target action. Depending on what exactly needs to be analyzed – requests, sites, gender and age of the audience – we set up different metrics and groupings.


What key phrases does your audience come for
Select the desired ad campaign as the last significant source.

This is necessary to take into account only those cases where user interaction with this campaign led to a targeted action.

Then set up the grouping by campaign and ad display condition, which is the keyword in this case:

All that has turned out, import into Excel – then we work with this file.

Pay attention to the keys that received less than 20 clicks – this number is not enough for further analysis.
For the rest, see:

How many conversions they brought and at what cost
How long does the user spend on the site
How many pages did he visit?
What is the failure rate for this key.

Disable keywords with a lot of clicks and a high bounce rate (30% or more), as well as with visits less than 30 seconds and zero conversions. If there is little targeted traffic, the semantic core needs to be improved – this applies to both search and YAN.

You can analyze Yandex.Direct search ads and ads in YAN only using the Report Wizard in Yandex.Direct. Yandex.Metrika has no functionality for counting impressions and CTR.
What sites is traffic coming from
If you have ads in the Yandex Advertising Network, it is important to know how effective the sites on which they are displayed are.
The report is the same – “Direct – expenses”. Groupings – “Site Type, Site”:

Filter all results with less than 50 clicks to get reliable results. Disable sites that do not bring conversions.

If you get a lot of sites with less than 50 clicks, add the Lead Price and Total Leads metrics to the report. Disable sites with high cost per lead, low conversions, and abnormally high bounce rates.
What gender and age is your audience
In the report “Direct – expenses” select the groupings “Yandex.Direct campaign”, “Gender” and “Age”.

Expand, as shown in the screenshot, the data for each campaign to determine who converts better – men or women. Why separately for each campaign? Even the same product on different sites can be bought by different audiences.
Next, we expand each category by gender and look at the conversions for each gender-age category:

What geographic regions is your audience from
If you are targeting multiple geographic regions, the buying activity can vary greatly. Therefore, this report is indispensable.

In the “Direct Expenses” report, select the Yandex.Direct Campaign and City groups and add metrics:

  • Visits
  • Visitors
  • Clicks
  • Cost per click
  • Refusals
  • Conversion for any goal, if all goals lead to a purchase / order, or Achievement of a specific goal, if there are goals associated with the intermediate stages of the sales funnel.
    Analyze the data in the order in which the groups were exposed – first by campaign, then by city. First of all, conversions are influenced by what conditions you offer in your ad.

Next, the easiest way is to download the report to Excel and calculate the goal price (the ratio of the cost of clicks to the goal achievement) for each city. Here is an example:

What time do your site visit more often?
The answer to this question is in the standard report “Attendance by time of day”

Set up groupings in the following sequence: “Yandex.Direct campaign” – “Day of the week of the visit” – “Hour of the visit”.

Similarly, add metrics for achieving goals and conversions.
To get more accurate data, take a period of at least 3-4 months, or 1-2 months – provided that you have high traffic (from thousands of visitors daily).

You get such a report as a result, in the lines – the names of the campaigns:

If you expand each campaign, you will see a list of days of the week and data on them – how well your audience is converted to each day of the week:

The next level is by the hour. At what hours of each day are conversions high or low:

What devices is your audience coming from?
There is a special report on devices in Metrik:

In it, you will see from which devices visitors come more often and perform targeted actions:

Also, look at the time spent on the site, the viewing depth and bounce rates for all types of devices, and adjust the rates if necessary. If, say, the mobile does not convert at all, put a reduction adjustment.

These were reports that show the quality of traffic – how targeted it is and how expensive or cheap applications it brings. Also in Yandex.Metrika there is a group of reports by which you can study in detail the behavior of your site visitors.
What visitors do on the site
The following reports will help answer this question:

The link map shows which links on the site are most likely to click on (the richer the color, the more clicks):

When you hover the cursor, a summary of the number and proportion of clicks on a specific link appears. For example:

The scrolling map shows what gets more attention when scrolling through the pages. The richer the color, the longer the viewing time:

Form analytics is a visual report on how they interact with application forms:

And also there is a Webvisor, which shows the actions of each visitor on the site. In it, you can see the record of each visit and understand what interests the audience more and what prevents them from performing the targeted action.

Leave a Reply

Your email address will not be published. Required fields are marked *