How does a new website build its own keyword library?

Having worked on websites for many years, I attach great importance to the accumulation of keywords. To this day, I have worked in countless industries, and in each industry I have accumulated at least 20,000 useful keywords. Because these key trainings can help me save a lot of time when I re-enter this industry.

When you enter an unknown new industry, the first thing many SEOs do is to dig out keywords and build their own vocabulary. In fact, this kind of thinking is wrong, because you don’t understand this industry, so the words you dig out are meaningless. When searching for keywords, many SEO companies do not consider user behavior and search purposes, which results in no traffic even if the website is ranked for certain keywords.

Actual combat cases:
Xiao Ming took on a website optimization project for the second-hand mobile phone industry, so he excitedly began to dig for words using "second-hand mobile phones" as the root word, and obtained the results "Beijing second-hand mobile phones, Shanghai second-hand mobile phones, Guangzhou second-hand mobile phones, second-hand mobile phone recycling, second-hand mobile phone repair". At this step, Xiao Ming thought it was great, I already had a word brand, I just needed to put Beijing second-hand mobile phones and Shanghai second-hand mobile phones on the website, and there would be traffic. So he posted a large number of second-hand mobile phone articles on the website and made a large external link for second-hand mobile phones. Half a year later, there was no traffic and no ranking. Next, we open a second-hand mobile phone page and see what the ranking website looks like:

On the list page of second-hand mobile phones, users tend to reflect the brand, color, memory and how new it is in the title. Users will not write "Second-hand iPhone 7 Plus Bright Black for Sale" in the title because they are not SEDs and do not understand why they should write "second-hand mobile phone". Therefore, the words that Xiao Ming dug out cannot be applied to such a page, but keywords such as mobile phones only need to appear in the title of the page. What users see on the website is "Apple iphone7plus bright black 1286 Chinese version 90% new iphone" while what users see on the search engine is "Apple iphone7plus bright black 128G Chinese version 90% new iphone - Beijing second-hand mobile phone".

Simply put, website optimization does not mean that you have to optimize a keyword and the word you want to rank must appear in the title of all pages. Instead, 80% of SEOs use a large number of keywords on each article page and list page to optimize a long-tail keyword.

Let’s look at an example of incorrect website optimization:

There are so many keywords in the title, but the content is not related to this industry. In fact, the competition for website optimization is not high, and such operation can also help the company obtain a high position.

Actual combat case 2:

Xiao Ming once again received an order for optimizing a Beijing housekeeping website, so he began to think about optimization strategies and keyword layout. He could not directly dig out words like last time, but had to understand the industry first. Xiao Ming's thinking is like this: a user wants to find a housekeeping company in Beijing, and he is concerned about which housekeeping company in Beijing is good and which one is cheap. So he started to optimize which housekeeping company was better, which was cheaper, and which had more efficient service. However, after half a year, there was still no traffic.

Here is actually the SEO not jumping out of its own thinking, just considering the problem from its own perspective, instead of from the user's perspective. I found that most of the websites are doing XX housekeeping which is good, XX decoration which is good, Xx training institution which is good, which English training is good. The problem lies in such keywords. Although SEO feels that there is traffic, it is actually very small.

When SEO is working on such long-tail words, they will use article pages to do it, so a page like this will appear: "Which housekeeping company in Beijing is the best? If you are looking for housekeeping, go to a certain housekeeping company in Beijing." It seems that there is nothing wrong with this title, and it also contains keywords and long-tail words, but in fact the user's thinking is "Which housekeeping company is there near Chaoyangmen in Beijing, with better service?" When users need a service, the first thing they do is to look for it based on their geographical location. Why do classified information websites and rental websites have to be divided by region? Because users want to find services near their location, rather than living in the second ring road and looking for a company in the sixth ring road for services.

In addition to the user's purpose, there are also the user's search habits and behaviors. 99 percent of SEOs will think that words like Qingdao wedding photography are more valuable, but the reality is that the words that users search and click on more frequently are "Qingdao Photo Studio, Qingdao Wedding Photography Personality, Qingdao Wedding Photography, and Qingdao Seaside Wedding Photography" and other keywords. Users in different regions have different search habits and names for different industries and products, so when building a vocabulary, you should consider it more from the user's perspective.

If you work in a company, you can use one sentence to summarize how to create a vocabulary library: "products for the whole country" and "services for the region". I often find that many companies don't understand SEO at all, and they don't know what Shenzhen elevators or Beijing mold temperature controllers are. If you must use a regional word, you must at least ensure that your product is as famous as "Chongqing Noodles", otherwise how will your users know that there are elevators for sale in Shenzhen? Of course, it’s not that this type of keyword cannot be ranked, but there’s no need to use the main site to do it. You can just divide it into two levels and you’ll be able to rank. The reason for providing services in the region is very simple. A decoration company in Shanghai cannot satisfy users in Beijing. So now when searching for decoration companies in Beijing, even if your website of a decoration company in Shanghai ranks first, users will not click on it. Even if there are some clicks, there will be no conversion, so such words are wasted.

So how do you build an effective keyword library?

Traditional methods of mining words include "Baidu Promotion Assistant (e.baidu.com), Baidu drop-down, related search, 5118, Aizhan, chinaz, CiKu.com, Zhuici.com, Locomotive Collection, Python Crawl, Jinhua Webmaster Tools" and so on. Without evaluating the effectiveness of the keywords extracted by these tools, as an SEO you should understand that your competitors also know these tools. In most cases, the keywords you dig out are the same as those dug out by your competitors. Invisibly, you have gained many more competitors. Moreover, in most tools, the indexed keywords we query are often caused by a large number of SEO searches. Therefore, you will find that even if your industry core words rank first, there will only be displays but no clicks.

Among the above tools, in addition to some simple downloadable tools, Pthon crawler is mentioned, which is difficult for many SEOs. As a personal webmaster, I have been troubled by the difficulty of digging out words, and I have studied it for three months. To use Pthon, you need to understand regular expressions and databases. I also asked a technician to help write a script, but I found that the keywords dug out were not as good as I imagined, both in quantity and quality. At least 60 percent of the words need to be thrown away. If the website you are optimizing does not reach a vocabulary of more than one million, you don't need to understand it specifically.

Like all SEOs, I used to try to use this method to dig keywords, until I found a better channel, so I gave up all of them. If I am in an industry, I will not dig out words first, but add the OO of the webmaster of this industry. Pay to directly purchase the opponent's statistics. Of course, the success rate will not be very high. After testing in many industries, websites operated by outsourcing companies will be easier to buy. Generally, as long as you can contact SEO, there will be no problem with the websites made by the company itself. Although the cost will be relatively high, when I was building a website for a mechanical enterprise, I bought the statistics of the competitor’s number one ranking. I found that the main keyword had an annual display of 150,000, but the click-through rate was only 3,600. So when I was optimizing the website in the early stage, I decisively gave up the idea of ​​the main keyword, and instead covered all the long keywords with more than 10 IPs in the statistics in my content pages. Within a month, the website achieved the result that the main keyword was not ranked but the traffic exceeded 500. As industry traffic continued to focus on my site, coupled with a little promotion and drainage methods, the website ranking continued to rise, and I built it from scratch in 60 days.

Because of the paid purchase format, it may not be recognized by many SEOs, but it is often the most real in statistics, and the others are false. Even the display volume in the bidding, if the index is brushed, the display volume will also deceive you. In addition to the paid method, I suggest that you use the webmaster tool to query your opponent's Baidu weight, then download all of his keywords, and then filter and retain those industry-related words within the 100 index. This way you can avoid head-on confrontation with your opponent to the greatest extent possible. Of course, after you filter the vocabulary, you still need to filter the keywords again, retaining those within 5 million results and deleting those above 5 million. When you have put all these words up, you will naturally have a certain amount of traffic (including the results of fuzzy matching). Then, extract all the vocabulary in the statistics after three months of website optimization, expand it, and finally layout it on the website. You will find that it is a very simple thing to double the traffic and increase the ranking in a short period of time.

Once you have a vocabulary, you also need to classify the keywords. If it is a product word, you need to divide it into product type, product color, product brand, product material, product purpose, product price, etc. After you have completed the classification, you will find that you will match almost every product attribute/type/color/brand/material/price/purpose. Even if the search volume for these words is not high, the cumulative amount is immeasurable. Keywords in different industries should have specific classifications, so we won’t list them one by one here. When you have a lot of keywords in your vocabulary, you can use Excel macros to manage them. Using Excel to expand the long tail is also a good choice. If you have a large vocabulary and are worried about duplication, you can use the [HiStar De-duplicate Text Tool].

As an excellent SE0, what you need to do is not to compete with others for the seemingly beautiful cake, but for the residues in the corner that no one pays attention to. Only in this way can you have the opportunity to overtake others. There are actually many words with indexes in search engines that no one is doing, and there are also many words without indexes but with traffic that no one is doing. If you can find them, you will find that doing SEO will be very easy. When you have a long-tail word that can be ranked, don't forget about it. Instead, when you find that there is a page ranking, immediately create N more pages to occupy the homepage of a certain long-tail word. Generally speaking, if you can rank more than three pages on the homepage for a long-tail word, you are considered qualified.

In summary, when you enter an industry, you need a lot of keywords to support your work. Whenever I need to optimize an industry site (medium to high competition), I will organize at least 5,000-20,000 keywords. As long as these keywords rank above 20%, my website will have traffic, instead of just focusing on a core word that has display but no clicks. In addition to purchasing paid statistics, you can also obtain accurate traffic by acquiring websites. Although the cost is higher, when you have acquired more than 10 websites in your industry, you have at least mastered the basic needs and search purposes of users in this industry, then your efforts are worth it.