Set off her hijab--an analysis of the website of people-centred

Intermediary transaction SEO diagnosis Taobao guest Cloud host technology Hall

Preface: This is not the problem of website analysis, not even "online" "offline" problem, but a commercial problem.

From the beginning of contact with the site analysis to the present, more and more feel the "people" as the core of the site analysis of the importance. Without a written definition, I explained to myself, "collect information about your site visitors (Visitor) through various channels (web analytics and non-web analytics technology), analyze and study their demographic characteristics (demographics) and behavioral characteristics, Then analyze and optimize the marketing channel, the way of communication and the content of the website on the basis of the understanding of the visitors. So, here is the human-centered site analysis and the so-called user-centric is not exactly the same, even more than the scope of the site analysis, to the CRM and Business analysis extension.

We know that there are two main methods of web analytics, User-centric and Site-centric, and we usually say that the site analysis mainly refers to Site-centric, that is, to be able to track the content (such as Web sites, applications, advertising media) as the core, Analyze data such as flow sources, click Streams (click Stream), content viscosity, etc. In this narrow web site analysis, visitors are treated as a "dimension" (dimension), but if you count the reports of all the Web analytics tools you know, you will find that the dimension is seriously overlooked, except for a few technical indicators, such as chicken ribs. It gradually degenerated into an unreliable indicator (Metric), such as "number of unique visitors".

It's like we're sitting in front of a terminal in the Internet, dealing with data from web Analytics in simple or advanced ways, making all kinds of ugly or stunning forms and graphs, and getting all kinds of people wringing or dancing. I don't care about the other end of the terminal is looking at your site is actually Garfield or Lightning dog. Indeed, this is a problem for the entire industry and the technical and legal environment; But if you can't resist arranged marriages, it's your fault to kiss through the veil on the night of the bridal chamber.

Perhaps our website analysts are dismissive, but in the brothers industry, another kind of web analytics technology has long been widely used. This is the User-centric method used in the audience analysis of media websites in the field of network media purchase. Compared with site-centric, user-centric mainly has two benefits, one is to provide competitive intelligence (competitive FDI), this is not within the scope of this article, interested friends can contact me; One is to provide information on the demographics of site visitors such as gender, age, education, income, etc., such as interest, the frequency of Internet use, and so on, which is the focus of this article.

Typical user-centric access to the Internet using data by monitoring sample populations (panel) and statistical methods. The most famous of these is the combination of ComScore and Nielsen Netrating and Mediametrie. But there are other, more flexible ways of collecting information, compared to the panel-dependent approach. For example, the well-known Google Adplanner (now DoubleClick adplanner) and Quantcast.

Coincidentally, I believe that most readers may now think of web Analytics in China, a product introduction of the article, a hung Ming statistics. The information collection method for such tools is a lot of flexibility, for Google's products, it is easy to think of Google's grasp of a large number of various account information, search information and the information collected by the tool, while Quantcast is mainly dependent on their partners and user network, Provides statistical information about the site's audience. If from his partner digging down, in the final analysis, whether it is e-commerce sites, social networks, network providers or webmaster networks, or ultimately through the way of anonymous tracking, the use of user registration information. Take a closer look at the information provided by Quantcast, from age, gender, income to family size, and number of children, to information that is difficult or virtually impossible to discover through Internet technology. As for the hung Ming statistics, through my test results, the information given is still very diverse (but for less than 100 of the traffic they can also give a demographic distribution, this is not to be Khanyi). As for the data sources and information processing methods, many people should ask for an account. believe that when the product is officially released, they will give the introduction.

Through the User-centric method can give us the website visitor Demographics Information, thus provides the guidance for the marketing strategy and the activity strategy formulation, especially to the advertisement sale provides the basis. But whether it's a way to collect information through a panel or to collect information through multiple channels, they have a short board: they can't "position". That is, I know that 20% of the site's audiences are women between 20-30, but I don't know who they are or how to personalize their "communication" approach.

In the marginal area of User-centric and site-centric, behavioral orientation (behavioral trageting) gradually developed. Behavioral orientation to people as the core, using a similar Site-centric method to track netizens, while synthesizing other channels, such as CRM information (User-centric method), to a visitor portrait (principle see here); Then the design of the behavior and attributes of the pattern to match, and finally based on matching results to achieve personalized communication purposes. So, to a certain extent, behavioral orientation is a "person" as the core of the site analysis technology, it and User-centric,site-centric formed a very good complementary.

In addition to behavioral orientation, in the "people" as the core of the site analysis (very reluctantly called the Web Analytics bar), we also have more accurate positioning methods-customer data integration (CDI, customer information integration) and identification technology. Friends who engage in direct marketing (directly marketing, please separate from direct selling) should know a lot about this piece. This article mainly talks about two points related to the website analysis: expands the knowledge and the accurate on-line direct marketing to the website registration user.

Do customer data integration companies have many, in the field of marketing, the most famous should be Acxiom (Prudential) it, in addition to the European market Sogec (FR), WDM (FR) and other companies. Each company has its own demographic database, such as Acxiom's personicx, which divides all consumers (households) into more than 70 groups, with 24 million French households in France alone. They collect the data through a variety of ways and partners: Post offices, newspapers and magazines, E-commerce and social networking sites, enterprise CRM data, surveys, telecoms companies, etc.

Personicx

Their business of course also a lot of, which and the Web analytics have links, for example, which day I semwatch (http://semwatch.org) Open registration, I collected a group of opt-in e-mail address. But I just have a bunch of email addresses. To make good use of these email addresses, such as selling soy sauce and doing real estate, you need to know the personal information of the owners of these email addresses. What about->?

At this point I can find these companies, compare my email address with their resources to enrich my own customer information. By comparison, I can give a part of the mail address plus the Post Office address, plus the owner's age and income situation, plus family conditions, plus interests and hobbies ... (In some legal circumstances, this approach is illegal or strictly restricted, so the business is geographically; What is the current situation in China?) But after such a lap, spent countless money, only to find that there are no more than 10 e-mail addresses on hand, and two are registered on their own test. What about->?

At this time I can still find them, to buy their opt-in address of the right to use (note: Only anonymous, can not have), carry out my marketing campaign. I can add a lot of detailed requirements when buying the right to use these addresses. For example, if I were to sell powdered milk, I can ask them for information about these people: at the same time, with e-mail address and post office address, young mother, age not more than 30 years old, home has 1-2 children, family income in 50,000 yuan-100,000 yuan, home address in the coastal line, second-tier cities, Like to browse a variety of women's websites, not Nestle or Danone's Mom club members.

Whether through behavioral orientation, or through customer data integration identification, is ultimately a Web site analysis data and other consumer-related data mining and application, and then to achieve accurate and one-to-one marketing purposes; At the same time, when doing web analytics, on the one hand should pay attention to collect this information, And with the database or Third-party applications, in order to obtain a comprehensive portrait of the site users, on the other hand, in the analysis, should be used to obtain the classification of information, the different customer base to do a detailed analysis and comparison.

———-here for a short period of time so that my friends can tell me that it's too long to write ——-

Please continue to focus on "lifting her hijab"-2 "continue to discuss the methods and possibilities of analyzing people-centred Web sites."

This article starts Semwatch, reprint please indicate the source

Original address: Http://semwatch.org/2010/05/user-centric-web-analytics-1