How to analyze the website data-the idea of clearing the website analysis

Intermediary transaction SEO diagnosis Taobao guest Cloud host technology Hall

How to analyze the website data? This is another question that the audience raised before sharing the traffic, and here's the thing to sort out.

The diagram below is a life cycle diagram of a website analysis that, after confirming that the requirements are analyzed and the data we need is collected (emphasizing that it is important to make clear analysis requirements, which can be avoided for analysis), we can fully use the various reports of the Web Analytics tool to analyze the data.

But the amount of data in the Web Analytics tool is so large that you may accidentally drown in the ocean of data, and you need to have a clear idea of what reports to use and what reports to see to help you quickly find the answers to your questions. The following is a Web site analysis of the data modules involved, which also provides a Web site data analysis of the general idea.

Based on the contents of the above data module, in the website Analysis report I usually to the following listed the plate and the index carries on the concrete analysis (listed below is in the website data analysis Some I personally felt the more important plate and the index, the different website important data index may differ):

Basic situation:

What is the level of traffic on the website? Compared with the data of similar websites in the industry, according to our market position, how much do we need to improve our traffic in the future?

Where do visitors come from? This is used to confirm that the marketing strategy we have developed matches. If there are a large number of users from some small language countries, do we have to consider building a multilingual site?

What kinds of devices do visitors typically visit the site? Today, as the mobility is becoming more and more popular, do we want to build our own mobile site or develop our app?

How about the proportion of new and old users? Are we able to keep old users while expanding new users?

Flow source

What are the main sources of traffic from the site, SEO, SEM, EDM or social media? What are the types of traffic sources we can expand?

What are the priorities of these traffic sources, what are the sources of traffic that we rely on most, and what are the highest conversion rates?

SEO flow level, how to upgrade?

How is the use of EDM, social media marketing, and conversion rates?

Website content

What is the page classification of the website?

Product page, catalog page, marketing topic pages and other types of page traffic and Transformation Performance (page value) How is the situation?

What are the most common landing pages on the site? is the content on the page exactly what we want the user to browse?

Is there a problem with the user's access path and are we directing the user to the main marketing topic or product page? For an analysis of access paths: You can refer to how the user's access conversion path is parsed.

Does the user have an interaction with the site that we expect to participate in, watch videos, download, vote, subscribe, or order?

Product Sales

How much is the order conversion rate and customer unit price of the website?

Shopping Cart Conversion Funnel data How, the user in which step of the waiver rate is high, shopping cart user experience can be optimized?

What kind of products are best sold?

How many times do users usually visit the site before buying or how long will they have to take into account?

It should be noted that when analyzing the data, if there is a problem (such as the conversion rate of the shopping cart is particularly low) but baffled, you can personally experience the Web site's access process, to see if there is a barrier to accomplishing a specific goal or task, perhaps you suddenly find the root of the problem.

Case：

When you need to do a comprehensive analysis of the site, you can use the contents listed above to the site's various data modules are analyzed systematically. But each marketing channel's website Analysis demand is various, the different demand analysis method also different. and encountered some indicators submitted by the Channel Department of data Abnormal analysis needs, we can flexibly handle.

Here are two simple cases.

Question 1: The daily average of an electric dealer's web site is $80 million, but one day it suddenly drops to only $ million.

Analysis：

We can follow the contents of the inside step-by-step analysis, to find out the root causes of sales anomalies, but if you are familiar with the business operation of the site, in such an unexpected situation we can sharply find the root cause of the problem, so as to quickly correct the problem to restore the normal sales of the site.

or in accordance with our habit of thinking. We are used to the relationship between sales and traffic, when sales problems, we will habitually check the site traffic. Is the flow down? About the flow change here are two possible ways:

Flow also has a similar range of decline = "traffic source out of the problem?" = "Traffic flow source (SEO, CPC, EDM, user area) for analysis =" Page traffic analysis (whether there is a change in commodity concerns)

No significant drop in traffic è is there a problem with the order conversion rate? = "The sales of products for analysis, the conversion rate of some products or almost all of the product conversion rate decreased?" = "to analyze the page flow of the product or to analyze the route Is it because the focus of the products with higher conversion rates has decreased, or is the purchase of the website leading to a poor user experience, or even a shopping cart system in a certain period of time can not access?

Analyze the data in depth from the start of the flow until the root cause of the problem is found. In addition, in the analysis of abnormal index data, some additional factors such as special days, major events, seasonal also need to take into account, such as "Double 11" others are in full swing in the big promotion, and you did not go to the fun, this period of sales may become more bleak.

The problem 2:EDM partners sent 500,000 messages to their North American users (the link to the GA UTM logo), but did not contribute to the growth of the site's sales.

Analysis：

Web Analytics System from the EDM traffic data How many = "This part of the flow from which areas, really North America?" = "What is the access path of this part of the visitors, there is no way to enter the shopping cart =" Finally have no order

Analysis results show that the EDM only brought a small amount of traffic, and the number of visitors in Hong Kong and Southeast Asia, did not bring any sales, it seems that the partner did not practice the content of the contract, the next time do not find their cooperation ...

Google Analytics Smart Alerts

In addition, in the analysis of the site index data anomalies, it is recommended to fully use the smart GA intelligent alarm function, which can significantly reduce the workload of your site. When the data is abnormal, it can give you the exception data indicators, and will list the reason for the abnormal data.

There are two types of reminders in GA: Custom reminders and automatic reminders. Automatic reminders are reminders that Google Analytics generates based on its algorithm. That is, every day the GA's smart engine examines the following dimensions (including, but not limited to) metric values to confirm that they have changed significantly:

All traffic

Visitor type (new visitor and return visit)

City

Area

Country/Region

Advertising Series

Keyword

Source

Media

Referral path

Landing page

Exit page

Click-Through (AdWords)

In addition to automatic reminders, you can also set up custom reminders to monitor your site's operating data. You can set a reminder standard for any metric and apply it to any dimension, and you can even apply reminders to advanced, subdivided access groups. We can set the site traffic and sales peaks and troughs to alert, so that when the main indicators of the site are abnormal, these custom reminders can be sent through the mail function in time to notify the relevant personnel. Only custom reminders can now be used to automatically send messages.

The website analysis does not have the fixed step and the method, when you are very familiar with the use of web analytics tools and to analyze the business of the site, you can not rigidly adhere to the above mentioned ideas and methods, but the purpose of web analytics must be clear: reduce costs, improve efficiency, analysis of the optimization work is not missing.

Original link: http://www.analyticskey.com/web-analytics/