SEO key phrase data evaluation
The majority of the SEO work focuses on the 3 points of " key phrases, web site web content as well as link framework", each of which requires to rely on data analysis abilities. As a initial step, the objective is to understand the search needs of users and also the search practices of different customers in various circumstances, and also to represent the content of the site to locate the search traffic that is presently ignored. It plays a straight function in directing the instructions of internet site content as well as web link building.
Throughout the numerous SEO online forums, the following inquiries usually develop when describing keyword evaluation.
1, just how to classify keyword words
2, comparable words exactly how to take care of
3, just how to utilize these words
In the face of a lot of entries and intricate types of initial word lists, it is evident that pure handbook handling is not realistic and also pure program handling is certainly not exact. So you require to make use of a great deal of suggestions, utilizing the program to process the initial word list, result the prospect collection and also specified reference data, in the handbook to select the method, effective and rapid as well as ensure the high quality of the above problems, the basic procedure is as follows.
1, to acquire the original word list, I am below with the Baidu promo API, you can additionally use Baidu phoenix metro nest, search engine drop-down box and also relevant search, competitive data, etc., techniques and prefabricated devices, not to say extra, the data is as complies with.
2, the initial word list for handling.
A lot of individuals will certainly utilize the typical search combination, bookkeeping for around 60% to 70% of the percentage, each combination contains the "word" and also "variable".
For instance: "where to rent out economical apartments in Beijing", where "Beijing" is a geographical variable, "Beijing" itself can change any type of city, district, area, street or other geographical associated term Words "Beijing" itself can change any city, area, area, road or various other geographically related term, while "where to lease affordable" is a suffix, which is created dead on the template rather than dynamically readjusted like a variable.
On top of that, words with a high search volume normally correspond to multiple words with the very same meaning, such as "where to rent cheaply in Beijing", "where to lease inexpensively in Beijing", and so on
. The next thing to do is to make a initial collection of the original word list to provide dependable information for the next hand-operated selection. The copying is based upon my individual routines.
The following example is based upon my personal behaviors: for each and every word in the initial word list, the prefixes and suffixes of the word are extracted and the number of occurrences of each suffix in all words as well as the overall number of searches.
This is the original word list after handling.
This is the computed suffix data.
3, hand-operated testing, extraction of variables.
Suffixes have actually been discovered, however still except the matching variables. As a result of the complexity and variety felixncxg446.simplesite.com/450310684 of the variables, it still calls for a fast glance by hand.
As the previous step has actually been calculated on the suffix information, the suffixes in descending order by total search quantity, to locate the prefix, suffix search quantity of the first 200 suffixes, and afterwards a quick glance over the key phrases including this suffix are what variables, are taped.
The style of the record can be free to play, combined with the real scenario, document the information required for their own evaluation.
The way I have done this is as adheres to, heaven boxes on the left are for comparable terms, i.e. "similar terms with the exact same search demands yet different search behaviour". Heaven box on the right is the corresponding high-frequency search mix, having different mixes of patterns or various variables.
4, the use of data.
This is not a dealt with method, depending upon the real scenario of the site. But the more unified way is that comparable affixes are put into a page, high search volume fastens put title, low search volume fastens reasonably installed in the web page, such as the blue box part of the above graph.
A overall of seven words, first find the complete search quantity representing each word suffix, the search volume in front of the word suffix concern on the title, the staying search volume of reduced reasonable ingrained in the page, for instance, at the bottom of the page plus the text summary of the web page, such as ' individuals who have actually seen the page additionally browsed:" city task search network, city task search need to be employed, city task search information network"'
Due to the low search quantity of the word suffix if a lot of variables, the possible long tail traffic plus a great deal, as a result of the small level of competition, the Web has these long-tailed words in very few web pages, if the website has a web page or even more pages can be full in these long-tailed words, after that the possibilities of these words will certainly show a lot.
Below is a feeling of the internet search engine to confiscate the quality of the index collection. Baidu is divided into 3 degrees of indexing, premium collection, common library, low-quality library, high-quality collection of the variety of pages made up 20% of the complete index collection, yet can fulfill more than 60% of the search need. In the quality of the library of website will have the chance to reveal, so the very same course of search terms, it is best to reasonably ingrained in A, B, C and other types of pages, in case A put up part of the web page did not get in the high quality library, it does not matter, perhaps B, C can fill out it.
There are many individuals will experience this circumstance, " work search network, job search to find work" is to " work search" as the root of words, the search quantity is high, need to be put on the title, that has been "recruitment" as the origin of the word The search volume of "recruitment network, employment and also work search" is also extremely high, a title can not place a lot of high search volume of words, just how to do?
It is recommended to make use of various web pages to bring the equivalent traffic for words with different origins however the exact same demand as well as challenging competitors. For example, " work search" and "recruitment", the equivalent page web content is employment info, so if "job search" already has a web page, we recommend making use of the very same information, as well as work search page different If there is already a page for "job search", it is recommended that a brand-new collection of web pages be developed with the very same information, yet with a different phone call method and also a various theme to capture the recruitment website traffic.
There is likewise a word co-occurrence problem, with tried several times, to discover "Ma Yun" "Ali cloud" "Wan network" this semantic connection is rather good, for search phrase evaluation, such as news consulting sort of The search term relevance is not strong terminal, you can manually establish a few seed words, export the matching article, by evaluating the seed word context to extract the TAG words with strong significance to the seed word. There is a prefabricated toolkit on the net: word2vec, which can conveniently accomplish the above needs. But for regular site key words evaluation, I directly really feel that it is not really valuable.
Additionally don't consider any type of program that can do it in one action, the program is only in charge of handling, the final decision still depends upon individuals.