Shulou(Shulou.com)06/02 Report--

This article is to share with you about how to analyze the website. The editor thinks it is very practical, so share it with you as a reference and follow the editor to have a look.

I. where should website analysis begin?

A complete website analysis application process is like this: analysis target → interpolation → analysis → action, but in fact most people can not participate in this link, so the understanding of the data is not deep enough, the analysis is easy to produce deviation. In order to make it easier to understand, let's briefly introduce the principles of website analysis.

The mainstream website analysis tools basically use the page marking method, that is, embedding a small piece of code that can be executed by the browser on the front-end page (such as JavaScript) is equivalent to installing a camera on the page, recording the user's every move on the page, and then transmitting it to the back-end server.

A tragic fact that must be clear is that the website analysis does not have accurate data, which may be due to the fact that it is too late to load the collection code, network transmission and other reasons. after all, what is collected is the terminal behavior, and what is the specific situation of the real person behind it? we have no way to know whether we have been browsing or going to the toilet and then coming back.

Data collection is the starting point of website analysis, and it is also a very tedious work, especially at the beginning, there is no specification for the page and promotion, and the collection code can only be customized, which is quite troublesome, even if the specification is established, the operation and development are often in a hurry to go online and have no time to follow, the flow of personnel also affects the implementation of the specification, and often finds out where it is not deployed when using the data.

There are some methods, such as solidifying the collection specification into CMS, promoting the process, monitoring the abnormal collection data, and so on.

Website Analysis Cross Proverbs

Enter the theme, how to make good use of website analysis data, the key is to grasp ten keywords: transformation, trend, subdivision, comparison, traceability, website analysis to see what, look at two: transformation, trend, what do you think? There are only three words: subdivision, comparison and traceability! These words are not invented by the author, but summed up by the elders. The author only talks about his personal understanding.

1. Transform

How many people come to the store, how many of them buy, this is "transformation". When we are in operation, we usually design the behavior path of some users, for example, we want users to visit →, visit the baby page →, enter the purchase process, →, fill in the information, → to confirm payment, or scan → to enter the activity page → to participate in activities, and so on.

We want to see whether the flow of users in each link is in line with our expectations, corresponding to those "transformations" that do not meet our expectations (the indicators involved are "wastage rate", "bounce rate", "departure rate", "withdrawal rate", if we examine these concepts, there will be a lot of differences, and there will be no introduction here), we want to know where the loss of users has gone and what is the reason.

The word "transformation" is not only the ultimate goal of operation, but also the core of website analysis. All the analyses are carried out around the word "transformation" (those who lose, jump out, exit, etc., can be regarded as bad "transformation"). The key to doing a good transformation analysis is to accurately grasp which are the key links. different paths are designed for different scenes to form a transformation funnel diagram.

two。 Trend

As mentioned earlier, the website analysis data is not accurate, but why is it meaningful? because the accuracy of the data is stable (although it is not accurate, the same monitoring method has been used all the time, the error and the deviation direction of the error are the same). Although the specific value is not absolutely accurate, it can accurately grasp the trend of operation.

For the operation, what is more important than the current is whether the traffic and transformation of the website have been improving on the right path, whether the important links, paths, page experiences continue to improve, whether the operation has hit the ceiling, and so on. Grasp this trend, clear whether they belong to the initial period, rising period, mature period or encountered the bottleneck of development, and different operational strategies should be adopted under different trends.

3. Subdivision

One of his predecessors said well: "without subdivision, I would rather die." For example, tell you that the number of website visits has increased by 10%. What's going on? You can subdivide through the web page and find that most of the web traffic has not changed, except that an active page has increased by 200%, or through the breakdown of traffic sources, you can find that a certain promotion source has increased a lot, so it is clear that this is just a simple example.

Without subdivision, there will be no truth. It is very important to choose the right subdivided dimensions and the division of dimensions. generally speaking, the division of subdivided dimensions should follow the following principles: determine which are stable and which are unstable, and pursue the unstable ones.

For the whole site, direct access and SEO sources are relatively stable in a short period of time, while promotion is unstable. As for the purchase conversion rate, it may be subdivided in terms of traffic source, purchase path, product and other dimensions, because different user quality varies with conversion rate, purchase path is different, and products of different specifications are different.

Sometimes, some relatively stable subdivisions have big fluctuations, which should be carefully analyzed. It may be that there is a major problem with the website or a change in the whole Internet habit.

4. Contrast

If you find a problem, start with the comparison. And their own past ratio, horizontal ratio, and industry benchmarking ratio, and target value ratio, and their own brain data comparison, subdivided and then compared. Only by comparison can we find the problem, and only by comparison can we find the direction of improvement.

With the same data, different people will have different effects. For example, the conversion rate of a whole station only knows that it has increased by 10% compared with the past, while B knows that this is close to the highest level in the industry, so there is no need to make efforts here. The empirical data stored in each person's brain is different, the data that can be compared is different, and the conclusions drawn are different.

5. Trace to the source

Sometimes no matter how subdivision, comparison can not find the problem, what to do? It is necessary to go back to the source, examine the original data records, reflect on the user's behavior and the process of data collection, and in addition, in-depth communication with operators, it is possible that some of the actions they have done, but the data has not been accurately restored.

Third, the author's original "point-face-line" website monitoring and analysis system.

How to establish an analysis system for an e-commerce website? The application of the above-mentioned cross maxim must first look at the overall transformation and trend, and then according to the subdivision principle, it is divided into two parts: daily operation and sales promotion monitoring.

For daily operation, it can be divided into three key elements: "point", "surface" and "line", and their transformation and trend are studied.

Points: for example, advertising space, QR code, navigation, search, customer service, recommendation, etc., these are the dots that drain to important pages, and their drainage rate and second-hop rate are very important for the conversion of the whole station. For example, generally 25% of orders come from on-site search, and 15% of color 25% orders contribute to on-site customer service.

Face: it is the core page, such as home page, landing page, baby page, activity page, etc., which are the key to retain users and facilitate transformation. These pages play different roles, so it is necessary to build corresponding analysis and evaluation models, such as the Heart model available on the home page.

Line: refers to the core processes, such as ordering, payment, registration, etc., whether these processes are smooth, in line with user habits, affecting the closing of the deal.

Analyze and refine these three key elements, transform the whole station into natural ascension, strengthen the internal power, and the pulse-type promotion flow can be accepted.

Thank you for reading! This is the end of the article on "how to analyze websites". I hope the above content can be helpful to you, so that you can learn more knowledge. if you think the article is good, you can share it for more people to see!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.