Quantitative operation on website SEO

Intermediary transaction http://www.aliyun.com/zixun/aggregation/6858.html ">seo diagnose Taobao guest cloud host technology Hall

In the management position of the SEO, operators, usually encounter such or such a problem, so sometimes when we talk about SEO people should be This "God" must also be a "God", every day to deal with the boss, colleagues and their own various "why", in these problems, which will have a similar "when Can the rise?" When can you do it? How long will it take? "and so on," and so on. Some of the "unreliable" or "unanswerable" questions, but when we think about it, are these questions really only "God" know? Today I'm going to talk a little bit, Some experience in the operation of the website, as well as the development and expectation of the natural search flow index.

Mention analysis, you need to have data, then today we focus on this data analysis, the search engine traffic operation is based on the main indicators: the number of Web pages included, the number of visits (IP), keyword quantity, For the above search performance indicators of the station support data is: The number of people who generate web data, the amount of data in the site, the amount of data generated by the person daily.

So, when we list these key data indicators, we all know that these indicators are interrelated and useful.

Personnel ———— Station data ———— included data ———— search term data ———— traffic (IP)

Have) Have) Have) Have)

From the above figure can be seen, people and search traffic indicators are linked together, then the ultimate natural search traffic indicators to implement the SEO or operations team members of the body is natural, but the key is to implement how much? How long is it to be done? These important nodes in the index, we need to do data analysis, To resolve these problems, you need to first resolve the site in the search engine existing performance, just good estimates, how to resolve that:

Included data, search Word data, traffic data, is a search engine performance in several important indicators, these values directly present a Web site search engine performance, we can use a common word-"weight", then these indicators "upstream" The basis is: The station data and the person that produces the data as well as the daily work data.

1. How much data is included in the station is included in the search engine? Here is a ratio, I will call it "included than" with the amount of data collected than on the site (Total page).

2. Key words than included how many keywords that? Here is a ratio, I will call it "keyword than" with the current number of keywords in the station (CNZZ, Baidu Statistics can be provided) than the current amount of collection.

3. Flow rate than these keywords generated how much traffic that? Here also has a ratio, for the time being called it "flow ratio", with the station flow ratio on the keyword amount (the proposed data model of the length of the latitude extended to more than half a year, so that data out, will be more accurate);

Of course, the above similar comparative analysis, there are many, eventually we will find that these seemingly independent data, in fact, there is a deep correlation, these values, can reflect the current site, a search engine performance, the specific analysis method, I use a case brief introduction:

There is a site, the current site Baidu daily average 8w IP, Site page data total 40w+, Baidu is currently included in the amount of 200,000 pages, communication, the daily average of about 4w, there are currently 20 editors, the demand is: to do 400,000 daily average IP, how long, how much money to spend?

Analysis of the site in the search performance:

1. Analysis included ratio: 40/20 = 2 (2 data produced one included, 50% included)

2. Analysis of keywords: 20/4 = 5 (5 pages generate a keyword, 5:1 of the keyword production)

3. Analysis of Flow ratio: 8/4= 2 (a keyword produces 2 IP, macro-look at the flow of single keywords)

4. Analysis Workload: Edit 40/day * 20 people = 800 articles/day

So, according to the present site in the search engine performance data, do 40wip/every day, need 20w keywords/daily, need 100w included page, need 200w station page, and editorial staff work task rating, then need 20 people work, need 6 years 8 months; some people say it is not too long, should say, Only according to the data analysis is this, but we can see subjectively, some are not reliable, because we ignore the objective problem changes, in the process of the site in the search engine performance (can be considered to be the weight) of the changes, the impact of brand words, the number of team personnel changes, content resources to collect channels of change, according to experience, This end result is generally better than the formula settlement results (because the site is mostly doing better) but if your site these values are not moving, I think the results will not be too far.

Some people would say that if one of my big flow words were ranked up, the whole is going to change, but what I'm trying to say is that if you zoom in on the data model and lengthen your time, you'll find that in macro data, single keyword rankings fluctuate, with little impact on overall flow, especially in large web sites, because The fulcrum of the flow is the long tail, the long tail of the fulcrum lies in the content, so the macro data, the flow of the fulcrum of the content of the data, rather than a few large flow of keywords ranking.

Some friends like to seriously, I do not want to say anything more, this is just me. In the operation of some level hundreds of thousands of, millions IP site operation, for macro data analysis after the experience, there must be deviation or in the scope of application has problems, (such as small sites, non-content sites, may not apply), and suggested that everyone, Split column (or industry) for analysis, because if you are a similar portal to a number of industry sites, this value must be very large, such as I encountered, education, games, macro down, the single keyword of the audience difference is very large, education information may be within 1:2 (a word to bring no more than 2 IP), And the game may be at 1:40 (a word to bring 40 IP) or even higher, of course, this is each station is different, their own.

This article by Zhang Original, www.2huacha.com reprint finishing