Yandex and Google: what actually differs when you optimise for each

One site, two search engines, and the temptation to assume the work is the same. A large share of it genuinely is: indexation, speed, structure, copy. But the list of things that have to be done separately is longer than it looks, and almost all of it consists of things English-language guides never mention, because Yandex is not in them.

What counts for both at once

Start with the good news, to draw the boundary. No duplication needed for:

• Technical health: status codes, no duplicates, correct canonicals, crawlability.
• Site structure and internal links. Both engines are equally bad at finding a page five levels deep.
• Depth and quality of the copy. The differences are in how it is assessed, not in the principle.
• The link profile. The engines weigh links differently, but a link exists regardless of who is counting.
• The mobile version.

When resources are tight, those five give the most return per unit of effort precisely because they pay twice.

Indexation runs at different speeds, for different reasons

The first thing a new project meets: a site can be fully indexed in one engine and entirely absent from the other, for months.

The exclusion reasons in the two consoles differ in wording and in meaning. Search Console talks about crawling and page quality; Webmaster more often points at insufficient interest in the site as a whole. That distinction matters: in the first case you fix a page, in the second you fix a site.

The practical consequence is to check both consoles and not carry a conclusion from one to the other. A split between engines almost always means something specific to one of them, not a general technical fault.

robots.txt: a directive Google does not have

Yandex supports Clean-param, a directive telling the crawler that the listed URL parameters do not affect page content, so it need not fetch the duplicates. The syntax, per Yandex's documentation, is Clean-param: p0[&p1&p2&..&pn] [path], with parameters separated by ampersands and the path optional (yandex.com/support/webmaster, checked 19 September 2026).

The same page notes that Yandex already strips common analytics parameters on its own: utm_source, utm_medium, ysclid, yrclid, so there is no need to list those.

Google has no equivalent: it works parameters out itself, and the Search Console tool for controlling them was retired. So parameter duplicates are handled for Google with canonicals, and for Yandex with canonicals plus Clean-param. This is exactly the case where one line in robots.txt closes a problem otherwise solved by editing templates.

Regions are set explicitly, not inferred

The most visible difference in day-to-day work. In Yandex a site's region is a setting: you declare it in Webmaster and it directly affects which results the site competes in. Google infers locality from a combination of signals, with no explicit switch.

That produces two different approaches to a regional project. For Yandex: a correctly declared region, an address and phone number on the site, presence in the business directory. For Google: URL structure, organisation markup, local mentions.

It also sets a trap during keyword research: commercial results differ enough between cities that clusters diverge. A regional project has to be clustered in its own region. More on that in our write-up on keyword clustering.

SQI: a metric with no counterpart

Yandex publishes a Site Quality Index for every site, a summary score visible in Webmaster and available over the API. Google has no public equivalent: PageRank stopped being shown more than fifteen years ago, and everything sold today as "domain authority" is computed by third-party services on their own formulas.

The practical value of SQI is limited: it is an indicator, and it cannot be managed directly. As a diagnostic it earns its place: an SQI of zero on a site that already exists and is being filled means the engine barely knows it, and that is a conversation about trust in the site rather than about optimising individual pages.

User behaviour carries different weight

Careful wording is needed here, because nobody publishes exact weights. What is known: Yandex has historically leaned on behavioural signals noticeably harder, and its own analytics product is installed across a large share of the Russian web, so it has a lot of data.

The conclusion is that on a site aimed at Yandex, anything affecting behaviour is worth more: a legible structure, no aggressive overlays, fast access to whatever the visitor came for. What reads as usability in a report functions as a ranking factor in Yandex.

Commercial factors

For commercial queries Yandex traditionally weighs a set of "real business" markers: range of products, prices present, delivery and payment options, contacts, company details. This is an observable property of the results: sites with complete product pages and stated terms sit above brochure sites.

Google assesses similar things through a different frame: quality and reputation signals. The outcome often matches; the work differs. For Yandex you edit the product page and the terms block; for Google, the about page, authorship, and external mentions.

A new site waits longer in Yandex

Set expectations in advance. Google usually takes a new site into its index faster; Yandex needs more time, and early on it is normal to have pages in one engine and none in the other.

What helps: a sitemap submitted to both consoles; an honest lastmod not tied to the build date; internal links to new pages from pages already crawled; external mentions. Yandex warns that processing a submitted sitemap takes up to two weeks, so no result on day three means nothing at all.

What not to do is resubmit the sitemap daily. It does not speed anything up and adds noise to your own diagnostics.

Analytics: two systems, not one

The two analytics platforms count differently and will always disagree. Arguing about which is right is pointless: they define a visit differently and attribute differently.

For a project aimed at a Russian audience the local option is more useful for two reasons: session recordings and click maps out of the box, and a link to the webmaster console. Both set their own cookies, so a consent banner is a working requirement.

What does the work, and what it costs

Both webmaster consoles and both analytics systems are free. The money goes on what they do not show: data about other people's sites.

Prices from the gbseo.ru catalogue as of 19 September 2026; "retail" is the vendor's own price for the same subscription. The full set is 4 999₽ a month against 36 700₽ at retail. The calculator on the homepage prices any subset.

An honest caveat about Western platforms on a Russian project: both are built for a world where the primary engine is Google. They do not cover Yandex positions or local search volume, and should not be relied on for that part. A detailed comparison of the two is here.

Running one site for both engines

1. Both consoles connected and checked separately, exclusion reasons read in each on its own terms.
2. One sitemap, submitted to both, with an honest lastmod.
3. Clean-param added to robots.txt for parameter duplicates; Yandex needs it and it does not bother Google.
4. Region declared in Webmaster; the same intent expressed for Google through markup and contact details.
5. Keyword research done on the project's primary engine, with the second used to check borderline cluster boundaries.
6. Positions tracked separately per engine. A single "average position" across two sets of results means nothing.

Six points cover nearly everything that genuinely has to be done twice. The rest is shared work, which is exactly why it comes first.

FAQ

Can you optimise for Yandex only and ignore Google?

You can, where the traffic split justifies it, and it often does. But the technical layer is shared either way, so "ignoring" in practice means not spending time on Google-specific regional and link work, rather than disregarding it entirely.

Does each engine need its own content?

No, and writing a separate text "for Yandex" usually makes the material worse. The differences sit in site structure, regional settings, and commercial blocks, not in the body of an article.

What do you do about an SQI of zero?

Treat it as a sign the engine barely knows the site, and work on being known: publishing, external mentions, a correctly declared region. There is no direct lever on the number, and services promising to raise it are not selling what they claim.

Is local analytics needed if GA4 is already installed?

For a Russian-audience project, yes, and for the data rather than for rankings: session recordings, click maps, and the link to the webmaster console are not in GA4.

How long does a new site take to index in Yandex?

Weeks rather than days, and processing a submitted sitemap alone takes up to two weeks. A sensible point to draw conclusions is a month to six weeks, assuming the sitemap is submitted, crawling is open, and internal links point at the pages.