Mouchangqing: Talking about the biggest functional defect of 51LA statistic

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

I should be the most loyal users of the 51LA statistics, starting from 05 to do the site until now, I have every site must use 51LA statistics. May be many years of use has developed a habit, plus 51LA some special practical features, is I to today also very highly praised 51LA reason.

I've written a few good articles about statistical tools. Every time you mention 51LA statistics, more praise than criticism (Baidu news search "51la" can find a lot of references to 51LA articles are written by me). A friend to do statistical tools even asked me, is not good relationship with the river, so the site is used 51LA statistics (in the eyes of competitors, will always be their own products better). I and arjunolic on the network did not say a word, not understanding, not to mention a good relationship. So in my previous article, when it comes to 51LA statistics, there is no false ingredient, just because you like it.

But today's article, not to continue to praise 51LA statistics how good, but to criticize. The purpose of this article is to hope that we can learn more about the usefulness of statistical functions, and also hope that 51LA statistics can do better.

1: Why do you think of writing this article

Last week I volunteered to find a webmaster class site friends, Exchange text chain ads, I hope on his website to recommend my new site. On his website, there are 2 locations that I can choose from. One is the site's first screen text chain ads, but this position is JS call, not an effective outside the chain. The other is the site at the bottom of the Friendship link location, if this position can add a high quality of the chain, to improve the PR value of the site to help. However, the flow effect may not be the first screen text chain position.

I weighed for a long time, finally chose the other side of the first screen text chain, feel that the flow is more affordable. Later on the other side is also in that position, to hang me a text chain advertisement. The 2nd day, I look at 51LA source statistics, found that the effect is not bad, came a good dozens of IP, at that time fortunately, I chose the right. But when I looked at one of my other statistical tools to see the source conversion analysis, I found that my choice was probably wrong. At the same time I also think of such a mistake, may be a lot of webmaster friends will encounter, so decided to write to remind more webmaster friends can also notice. Why wrong, please take a closer look at the picture below.

2: Source conversion analysis, the basic function of statistical tools

New Station (chaxun.la) one day traffic routing situation (figure)

Through the above figure, you can see that the site's largest 9 antecedents, respectively, resulting in the follow-up effect. You can see the "number of pages per visit" "New visit percentage" "Average stay Time" "bounce rate" and so on. One of the 7th antecedents is the amount of traffic from that text chain AD. Other travel time is the lowest 1 minutes or more, and the 7th of the route, site stay time on average only 2 seconds. Basically open the page has not looked, it closed, per capita PV is only 1.09 is the lowest, the bounce rate is also the worst.

Such an experience of the antecedents of friends, a look at the false flow of knowledge, if the real user access, it is not possible to such a low conversion rate. Conversion rate of the highest number of antecedents, in addition to direct input, the remaining 4 are their Webmaster class site, because the site type counterpart, are the real user clicks, so the conversion rate is relatively high a lot.

A closer look at the friends will find that Baidu and Google's antecedents, conversion rate is particularly low. Why is the search engine's conversion rate so low? Because this part of the routing, although the user is true, but not the corresponding target user groups. I this site is to do webmaster tools, but also be Baidu Google has included a page of the W, it will inevitably have some users search some and webmaster tools not related to the keyword into the site. It is as if the user is looking for beauty map, the results opened a look is a webmaster tools, will be close to the door. The reason why there is still 1 minutes of stay, because there are some from the search engine to the keyword traffic is also targeted users, so the overall conversion rate led up.

I used to deliberately optimize the blog "Erotic May days" The word, then found that search the word into the blog, stay time is basically only about 3 seconds. Without subsequent source conversion analysis, it is not easy to know the true effect of these flows.

51la has not yet launched this function, I would like to their technical experience, certainly not the reason for the technical threshold, it is not possible to have not been thought of, did not launch may have unknown consideration. Now the statistical market competition is fierce, if 51LA can fill this defect, I believe that 51la in the market share of the occupation has a very big help. Conversely this function has not been, gave 51LA loyal users, use other statistical tools reasons.

3: Other minor flaws

Another 51LA has been more proud, IP statistics is very accurate, will not repeat the statistics of an IP. This is also what I used to appreciate, because with 51LA statistics are often less than other statistical data, I have been like statistical data less statistical tools, because the flow is low, more can promote me to do promotion. But the IP statistics also has the biggest flaw, is the IP is easier to do the false, compared with the UV (independent visitor) makes the false more easy. Now most cheat software is for IP.

The last small flaw, is familiar to most stationmaster. Is the occasional loss of data statistics, pause data statistics, basically every month will come to this time. To my own small site to no matter, less to see the day's data has little impact. But for the regular website with large traffic, the daily statistics are very important. As far as I know, some of the advertisers to do traffic transactions, but also used 51LA to statistical advertising data (because of the low 51LA statistics, especially the Ads to buy window ads, like to use 51LA statistics) if suddenly a day of statistical loss, this part of the population is still influential. Because 51LA this flaw, also causes some websites not to continue to use 51LA the reason.

4: The final conclusion

Today, the ultimate goal of writing this article is to remind those who have to do traffic cooperation, advertising to buy webmaster friends. As such cooperation, do not only 51LA a statistic. Because there is no website source analysis, encountered some false cheat flow, it is difficult to identify, and thus economic loss. My own site, as long as the more dedicated to promote operations, I do not have to put 51LA of this statistic, will be used in other statistical tools to assist use. Another source analysis of this function, also can not guarantee 100% will be able to identify, false cheating traffic, because some cheating means also improve, more need webmaster own experience accumulation, to analyze.

If you are interested, there will be a chance to continue to write about traffic statistics articles, do network promotion, website operation, the most important is data analysis. and data analysis is based on having a good statistical tool. Especially E-commerce site, the demand for statistics is higher, need to understand is not only the flow rate of conversion analysis. E-commerce website, pay more attention to CPA,CPS and other effect conversion statistics, this need more powerful statistical tool to support.

Mouchangqing (mcq0544) Stay

2010-1-27

This article first address mouchangqing Network promotion blog: http://www.muchangqing.com (reprint please keep)