Myths about setting up contextual advertising

[subornaakter40]

Setting up contextual advertising, contrary to the opinion of the majority, is not such a complicated process. The main thing is to pay attention to some nuances and avoid annoying mistakes. Which ones exactly - we will list below.

First, you need to develop and launch an advertising campaign, then set up analytics
This mistake is one of the most frequently made. There is an opinion that the process of setting up an advertising campaign and analytics should be separated. They say that these works can be delegated to different specialists, and web analytics can be considered a separate service. But they are part of a single set of works performed veterinary email list before the launch of the advertising campaign.

How to deal with RK

Yandex. Metrica and Google Analytics counters have been installed on many resources since their creation. Most likely, they contain previously collected data. Do not neglect such information. Analyze it, take it into account when launching remarketing. This is what you need to resort to at the very beginning, if you already have an accumulated base. It will help in developing segments and the necessary scenarios.

Check how correctly the counters are configured, whether they are up-to-date. Sometimes an old version of Yandex.Metrica is installed, the code is not embedded on every page of the resource, data is not collected from all subdomains, maybe the events on the target do not work correctly. Do not launch a campaign until you have thoroughly checked everything and are sure that the necessary data for analyzing efficiency and optimization has been collected.

Launching YAN requires the same precise semantic core as a search campaign
In YAN, you can often see advertising campaigns that are made according to the search engine model, namely: a semantic core collected using a parser, including low-frequency queries, phrases grouped by similarity and according to the principle of "1 phrase - 1 group". At the same time, the ad itself is written with the addition of a keyword to the title or using a template with auto-substitution. In addition, the advertiser uses the same list of negative words. Such an advertising campaign takes a lot of time, and the effect from it is not that high.

Let us explain what the matter is. In YAN, targeting is based on the behavioral characteristics of the target audience and on the thematic properties of the platform itself. That is, a large list of negative words (including total cross-minus) reduces the reach of the target audience and becomes an obstacle for algorithms.

As practical observations show, when launching large-scale advertising campaigns, low-frequency queries account for 5-10% of the total volume of traffic and conversions. In other words, these are single conversions, which are not enough to understand whether a particular phrase has an effect. Thus, the conclusion is: use high-frequency and mid-frequency phrases, which consist of one to three words. This will allow you to save as much time as possible in the process of developing, analyzing and optimizing the advertising campaign.

Combine phrases into groups based on the product itself and the audience. A group can contain up to 200 different key phrases. The determining factor is their relevance to ads and landing pages.

Let's look at an example: the topic is selection of cars to order .

In one group, you can include such key phrases as auto selection, expert, car selection specialist, and others. In this case, in the ad itself, you need to write approximately the following text: "Are you looking for a reliable (high-quality, proven) used car? We will help you with the search and negotiate the price for our services."

If we were running a search campaign, we would assign keywords to different groups instead of one. This would make the ad more relevant.