Tips that do not work for search robots.

Official robots from search engines will see the original content without any problems, and the copymaster will receive misinformation and, in fact, crooked content.

Doing this in an online store is a bit of a hassle (we just always do it with a margin, to the maximum), but for service sites and content projects – masthave.

We prohibit the preservation of archived copies of the site in search engines.

Sites are often parsed directly from saved copies in serp’e. To make copies of the site pages unavailable in the search results, you need to place a simple code in the head section.

<meta name=”robots” content=”noarchive” />

We actively use these and some other methods when promoting websites and have never received sanctions from search engines.

Not so long ago, one of our sites was massively complained to Yandex technical support. They say that their content is hidden there, it’s not fair, we should be in their place, and so on according to the classics. It was not crowned with success. Yandex sent standard unsubscriptions, and we are still standing in the top.

Semantics, technicality, linking – the foundation of the result

No cheating of behavioral factors works without a competent structure of the online store, thoughtful clustering, linking and other standard but important work. If after the implementation of the work there are not at least 1000 live visits per day, then something is wrong.

For 9 months we have:

moved to a new engine and changed the design;

implemented 35,000 commercial keywords;

we have developed and implemented 12 TOR for the modification of the product catalog;

implemented internal linking with synchronization of product names in the 1C database;

we wrote in python automatic generation of arrays of text content with the possibility of updating in a couple of clicks directly on the site.

To cheat behavioral factors, parsing competitors, and other needs of the online store, we have assembled a separate server based on X99-T8D with two Xeon E5 2078v3 and 128 GB DDR3.

The pf cheat was carried out stepwise, no more than 20% of the natural traffic from the search with automatic adjustment of daily volumes depending on positions and phrasal frequency.

All clicks are successfully registered in Yandex.Webmaster and the robotness in the metric rarely rises more than 3%. This means that the fraud systems of Yandex.Metrica and Yandex.Webmaster cannot identify our bots.

Conversions are a strong signal for Yandex. Bots left orders with a floating conversion rate, filled out callback forms and purchases in 1 click. All this is tracked in Yandex.Metrica and perfectly boosts search traffic after pf updates.

To protect the site from possible sanctions, we permanently disable the Session Replay and connect the Clarity service instead. It is free and does not transmit any data to search engines.

After 11 months of work, we managed to reach ~79,000 visits from search engines per month from an abandoned site with killed behavioral ones.

This is a small traffic with a very smooth growth. When running on novoreg, the results would appear faster.

95% of the traffic comes from Yandex, and Google requires a separate budget and time to build up the link mass. Cheating pf in Google is unprofitable and ineffective – there is no point in doing it, links make the top.

Over the year, the financial dynamics of the online store gained momentum and by the end of January 2022, the total statistics for 12 months looks like this.

The conversion rate from search traffic corresponds to the market average in the subject of auto parts. Several million rubles can be safely deducted from the income – these are lost and abandoned orders.

The most inefficient channel with negative profitability is contextual advertising. Lots of bots, low average check, 0.3% conversion rate.

In the dry balance

The start of the work begins with a reserch.

It is impossible to enter most niches without protection from bots and pf adjustments.

With poor or absent optimization, the pf cheat does not work.

To catch the filter – you have to try hard.

Posting live positive reviews about an online store gives an advantage in ranking.

There is no need to promote query clusters if all positions in the top 10 are occupied by marketplaces and aggregators.

The more search traffic – the more difficult it is to twist the pf. You need to try to reach large volumes in the shortest possible time.

Someone does SEO, and someone makes money.

I attach a survey. I am interested in your opinion.

Abnormal traffic growth in the regional online auto parts store

Even my reviewers (sites with product reviews) beat all records in terms of traffic and commissions for sales. These were the most abnormal figures for the entire existence of the projects. Perhaps this will never happen again.

After March-April, there was a very smooth decline. Yandex’s semantics frequencies, real traffic, and order volumes began to decline. This lasted until September.

Traffic has become sawtooth. This has become especially pronounced in Yandex. According to statistics from the Webmaster, this video is very clear:

Website with reviews of household appliances and electronics

Online store of spare parts for LADA cars

The situation is similar for dozens of other ecom projects. Search traffic has become very dependent on news. Since the beginning of September, tens of millions of users have been increasingly reading news telegram channels and watching relevant videos on YouTube.

After September 21, traffic in the entertainment and commercial segment began to fall sharply. When studying the patterns of behavior , the current situation is as follows:

  • Monday – Wednesday – normal traffic and sales activity;
  • Thursday – traffic decrease by 5-10% compared to Wednesday;
  • Friday – traffic decrease by 5-15% compared to Thursday;
  • Saturday – traffic at the level of Friday, sometimes small movements;
  • Sunday – traffic growth by 3-5% in the evening compared to Saturday

Leave a Reply

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