Post by sumiseo558899 on Nov 7, 2024 2:28:39 GMT -5
Clustering is the process of grouping and distributing search queries across landing pages. Also often called "splitting", it allows you to determine which queries can be promoted on one page, and which will require a separate, optimized landing page that more accurately meets the user's intent (search need).
Clustering is the most important stage in building an effective SEO-promotion website structure, significantly influencing the success of attracting search traffic in all directions corresponding to the website’s subject matter.
In simple terms, developing a content writing service
project structure can be understood as:
collection and cleaning of semantic yard;
distribution of queries across landing pages;
adjusting the site structure: dividing/merging/creating new landing pages to meet unmet demand;
manual optimization of pages using information about queries corresponding to the pages in question.
The process itself can be automated to some extent, but in most cases it requires manual work by the optimizer.
The article deliberately does not examine in detail the services and programs for grouping requests, but only provides examples of using certain services for various types of breakdown.
Types of clustering
By semantic similarity (by word form)
A type of query distribution into groups that uses the similarity of words in groups.
Popular tools: Excel (using filters), SEO Excel add-in, Key Collector in Group Analysis mode.
Pros: fast, high level of automation.
Cons: in automated form – does not take into account synonyms, as well as the degree of commercialization of queries.
Examples
When using grouping by semantic proximity, synonymous queries that can be promoted on one page will fall into different semantic groups.
Group Requests
ushm Angle grinder
Buy an angle grinder
Angle grinder prices
Bulgarian Bulgarian
Buy a Bulgarian
Bulgarian prices
Also, when using this method of division, requests of an informational and commercial nature can be placed in one group and promoted only on separate pages.
Group Requests
Air conditioners Buy Air Conditioner
Air Conditioner Prices
Bedroom Air Conditioner
Best Air Conditioners 2020
The example is exaggerated, but given that in addition to requests with explicit (commercial or informational) intent, there are a lot of requests with implicit intent, correct breakdown by this method is significantly complicated.
By TOP
The most popular way of distributing keywords. It works by downloading lists of URL pages from the top and searching for the number of matches - common URLs for different queries.
Free from the problems described in the examples of the previous method. The same queries with a grouping strength of 2 and the region "Moscow" are used to obtain the following clusters:
news
news
Clustering by this method often implies:
Specifying the region for parsing the TOP.
Specify the strength of the grouping: the minimum number of common URLs for different queries that is required to group the queries together.
Optional: select a grouping method.
Methods
Soft clustering involves grouping queries if each query in the group is associated with at least one other query in the group.
Medium clustering means that to create a group, each query must be associated with one main ("marker") query of the group.
Hard clustering means that there is a connection between all queries in a group.
It is important to understand that the "presence of a connection" is identified taking into account the strength of the grouping. That is, with a grouping strength of 2 and the Hard method, all queries that fall into one group will have at least two common URLs in each TOP.
Schematically, grouping methods can be depicted as follows.
news
Popular tools: Just Magic tool, Rush Analitycs, Coolakov, Key Collector using search engine TOPs download.
Pros: fast, high level of automation.
Cons: By playing with the settings for the type and strength of the grouping, we take risks.
or create an extra page (weaker in terms of assortment/content, as well as in terms of the speed of accumulation of data on the PF), breaking it down in more detail,
or fail to hit the intent of the user’s exact request, deliberately “dumping” on him not the exact corresponding assortment/information, but a wider range of goods/services/information.
By intent (by meaning)
Having delved deeply into the subject and analyzed in detail the needs of users in the search results, we come to what seems to be an ideal way of clustering queries – by their meaning.
Pros: when deeply immersed in the subject matter, it perfectly matches the user’s needs.
Cons: so slow and labor-intensive that full clustering by intent can only be considered for microsites with a small volume of semantic core.
By TOP with intent unification
Initially using the automatic clustering method by TOP (selecting the method and strength of grouping in accordance with the competition in the subject and the project features), the automatically created groups are then manually combined into larger clusters of queries by general meaning.
Based on the user's needs in the search results, it is possible to reduce the strength of the grouping for individual groups of queries, forming more complete clusters and creating common pages for them that are more "powerful" in terms of optimization and complete in terms of content.
Pros: automated primary stage (by TOP), detailed grouping by intent.
Cons: none.
Content completeness should be perceived not only for information resources in the context of article completeness, but also for commercial resources: general, full listings, or service pages that cover the maximum number of related user intents.
By TOP with breakdown by intent. "Superclustering"
Here we act similarly to the previous method. Initially, we use automatic clustering by TOP using any services and basic settings that correspond to the state of the project and competition. Then we additionally manually divide the formed groups into separate clusters that better correspond to the exact demand with the creation of the most optimized, detailed landing pages.
Pros: automated primary stage (by TOP), detailed grouping by intent.
Clustering is the most important stage in building an effective SEO-promotion website structure, significantly influencing the success of attracting search traffic in all directions corresponding to the website’s subject matter.
In simple terms, developing a content writing service
project structure can be understood as:
collection and cleaning of semantic yard;
distribution of queries across landing pages;
adjusting the site structure: dividing/merging/creating new landing pages to meet unmet demand;
manual optimization of pages using information about queries corresponding to the pages in question.
The process itself can be automated to some extent, but in most cases it requires manual work by the optimizer.
The article deliberately does not examine in detail the services and programs for grouping requests, but only provides examples of using certain services for various types of breakdown.
Types of clustering
By semantic similarity (by word form)
A type of query distribution into groups that uses the similarity of words in groups.
Popular tools: Excel (using filters), SEO Excel add-in, Key Collector in Group Analysis mode.
Pros: fast, high level of automation.
Cons: in automated form – does not take into account synonyms, as well as the degree of commercialization of queries.
Examples
When using grouping by semantic proximity, synonymous queries that can be promoted on one page will fall into different semantic groups.
Group Requests
ushm Angle grinder
Buy an angle grinder
Angle grinder prices
Bulgarian Bulgarian
Buy a Bulgarian
Bulgarian prices
Also, when using this method of division, requests of an informational and commercial nature can be placed in one group and promoted only on separate pages.
Group Requests
Air conditioners Buy Air Conditioner
Air Conditioner Prices
Bedroom Air Conditioner
Best Air Conditioners 2020
The example is exaggerated, but given that in addition to requests with explicit (commercial or informational) intent, there are a lot of requests with implicit intent, correct breakdown by this method is significantly complicated.
By TOP
The most popular way of distributing keywords. It works by downloading lists of URL pages from the top and searching for the number of matches - common URLs for different queries.
Free from the problems described in the examples of the previous method. The same queries with a grouping strength of 2 and the region "Moscow" are used to obtain the following clusters:
news
news
Clustering by this method often implies:
Specifying the region for parsing the TOP.
Specify the strength of the grouping: the minimum number of common URLs for different queries that is required to group the queries together.
Optional: select a grouping method.
Methods
Soft clustering involves grouping queries if each query in the group is associated with at least one other query in the group.
Medium clustering means that to create a group, each query must be associated with one main ("marker") query of the group.
Hard clustering means that there is a connection between all queries in a group.
It is important to understand that the "presence of a connection" is identified taking into account the strength of the grouping. That is, with a grouping strength of 2 and the Hard method, all queries that fall into one group will have at least two common URLs in each TOP.
Schematically, grouping methods can be depicted as follows.
news
Popular tools: Just Magic tool, Rush Analitycs, Coolakov, Key Collector using search engine TOPs download.
Pros: fast, high level of automation.
Cons: By playing with the settings for the type and strength of the grouping, we take risks.
or create an extra page (weaker in terms of assortment/content, as well as in terms of the speed of accumulation of data on the PF), breaking it down in more detail,
or fail to hit the intent of the user’s exact request, deliberately “dumping” on him not the exact corresponding assortment/information, but a wider range of goods/services/information.
By intent (by meaning)
Having delved deeply into the subject and analyzed in detail the needs of users in the search results, we come to what seems to be an ideal way of clustering queries – by their meaning.
Pros: when deeply immersed in the subject matter, it perfectly matches the user’s needs.
Cons: so slow and labor-intensive that full clustering by intent can only be considered for microsites with a small volume of semantic core.
By TOP with intent unification
Initially using the automatic clustering method by TOP (selecting the method and strength of grouping in accordance with the competition in the subject and the project features), the automatically created groups are then manually combined into larger clusters of queries by general meaning.
Based on the user's needs in the search results, it is possible to reduce the strength of the grouping for individual groups of queries, forming more complete clusters and creating common pages for them that are more "powerful" in terms of optimization and complete in terms of content.
Pros: automated primary stage (by TOP), detailed grouping by intent.
Cons: none.
Content completeness should be perceived not only for information resources in the context of article completeness, but also for commercial resources: general, full listings, or service pages that cover the maximum number of related user intents.
By TOP with breakdown by intent. "Superclustering"
Here we act similarly to the previous method. Initially, we use automatic clustering by TOP using any services and basic settings that correspond to the state of the project and competition. Then we additionally manually divide the formed groups into separate clusters that better correspond to the exact demand with the creation of the most optimized, detailed landing pages.
Pros: automated primary stage (by TOP), detailed grouping by intent.