Contextual advertising is an effective way to promote a resource on the Internet. One of the most useful inventions on the web is search engines. Instead of showing the user useless ads, search engines display only what is directly related to the query. This improves search accuracy and user conversion.
Every well-thought-out resource has a semantic core.
Important! Semantics is a scientific branch that studies the semantic content of each unit of a language. The semantic or semantic core of the site, advertising campaign helps to determine the direction of the resource and its target audience.
Selection of key phrases is the basis of the process of compiling a semantic kernel before launching contextual advertising. The accuracy of hitting the target audience and the results of the campaign depends on its accuracy. Usually, this process is divided into three stages:
Collecting all kinds of phrases related to the subject of advertising (it is important to collect the maximum number);
Cleaning from unused or ineffective ones;
Clustering the remaining phrases.
Definition! Clustering is the process of selecting groups of objects with similar properties, structure, from the original set.
In other words, clustering is the process of grouping keywords for contextual advertising. As a result, it becomes possible to perform an accurate analysis of the results of a particular advertising campaign.
Why do I need to group keywords in contextual advertising?
Ad relevancy increases. Key queries combined into one group help to customize the display of ads for the phrases of a particular cluster. Thus, each cluster receives a separate announcement with its own information call;
A/B Tests. These tests help you choose the most effective from several options. The total site traffic is divided into numerous landing pages with different keywords. As a result, the most effective advertising channel is selected;
Precise definition of search terms. By analyzing the grouping of keywords, you can determine the relevant search queries for an advertising campaign before it is launched, thereby saving a lot of money;
Combining behaviorally similar phrases makes it easier to analyze statistics. Among users of the same cluster, similar behavior is observed for different key queries. This axiom helps to analyze the results faster and more accurately. For example, Google AdWords offers handy group analysis tools;
Convenient management of advertising campaigns is achieved by reducing their number, without losing efficiency. Often, a positive side effect is a reduction in the cost of running advertising campaigns within a single site;
Cannibalization effect elimination. Often, competing pages appear within the same resource that hinder conversions by competing for traffic. Properly selected groups of key phrases will help you create a content plan, develop an effective landing page, and avoid duplication. This event leads to an increase in conversion.
Keyword grouping for Google Ads and other services can be done in a variety of ways. Some can be fully automated, others are only performed manually (as a result, groups are assembled with maximum accuracy). How to carry out clustering, each specialist decides for himself. Usually, the method depends on the goals of the advertising campaign. The result of such work will always be several lists with sets of keywords.
Morphological grouping of keywords
The easiest way to select keywords is to collect by morphological features.
In simple terms, this is the union of words with the same route into one group.
For example, insurance services, insurance, insurance services, etc.
Because of its simplicity, it is easy to automate. There are services on the Internet that will help you quickly group the necessary queries, discard prepositions and prepare them for a general view for insertion into advertising accounts.
Semantic clustering
This is an advanced morphological grouping. Here, keywords and phrases are combined by semantic meaning (due to synonyms).
For example, smartphone repair, smartphone repair, smartphone restoration, etc.
Automatic algorithms make such selection and grouping worse. Automatic services help to do the main work. But, after them, the results will need to be rechecked manually. Often, in the process of manual adjustment, new requests are recorded, or unnecessary ones are deleted.
The best tools are those that offer to manually enter synonyms for a specific query before selecting key phrases.
Concatenation by Intent
A custom intent defines a conversion. The person who came to study the information is not yet a buyer. And vice versa – if you make the buyer look at a product for a long time or bombard him with a lot of unnecessary questions, he will leave.
A common mistake that loses most customers is directing the target customer to a page where they can’t quickly check out. The user leaves after 15 seconds.
Example of request grouping:
How to stick a protective glass on the screen, stick a protective glass on the video, stick glass on the screen with your own hands, etc.
Buy a protective glass for a smartphone, order glass for a screen, sell glasses for a phone, etc.
The first group is information requests, the user is looking for information and is not going to buy. The second group is transactional. When setting up contextual advertising for business (services or goods), transactional requests will be the most interesting, they will receive the bulk of conversions.
Group by sales funnel
The sales funnel is precisely defined for each product stages that the user goes through before making a purchase. Describing a funnel, following it, and measuring results for adjustments is a cardinal rule for every online marketer. Because, at each stage, communication with the user requires separate, clear actions.
For example, a potential buyer is looking for a multicooker. By the queries that he enters, you can understand at what stage of the funnel he is:
Why do we need a multicooker, what to cook in a multicooker, the differences between multicookers and double boilers, etc.;
How to choose a multicooker, top 10 multicookers, characteristics of multicookers, etc.;
Buy a multicooker, order a multicooker with delivery, etc.
In order to lead the user to a purchase, at each stage it is necessary to give the user specific information and separate triggers for moving to the next stage. Today there are no services that allow you to automate this method. However, it is most effective when selling goods (especially expensive ones).
How to split keywords into groups?
The most convenient way to break down key phrases into groups is to descend from general phrases to more specific ones (or from high-frequency queries to low-frequency ones). It is much easier to break the semantic core into large clusters that describe the general essence of the product and service. Then delve into the details of each cluster.
On selling sites, usually the general grouping of keywords in AdWords occurs by product categories (household appliances, computers and accessories, power tools, and so on).
In the service sector, services are often grouped (dental prosthetics, veneers, caries treatment, tooth extraction, etc.).
When several initial groups are ready, you need to go down below. The initial group is divided into child clusters, in which their groups of requests are collected and grouped. For example, for hardware stores, computer equipment is selected (laptops, peripherals and components for computers, displays, motherboards, sound and other components). In services, you can also achieve more specifics (window cleaning in apartments, window cleaning in offices, window cleaning after repair, and so on).
The best option is to combine artificially generated semantics with manual selection. Often, the following links work well for trade and services at the same time:
Brand and object;
Object and characteristic;
Object with toponym;
An object with transactional additions.
In conclusion, we note that the correct selection of key phrases is the key to maximum effectiveness of contextual advertising with minimal costs for an advertising campaign.
