Web page metrics and segmentation

When we use some web analytics tools, we find that general reports are divided into three main modules: User access, content browsing, and traffic sources. Each category consists of a variety of analysis metrics that make up a variety of presentation reports, starting with a description of the various metrics under the Content browsing module (primarily the page view of the Web site), and what subdivisions we can implement based on these metrics.

The definition of some common web analytics metrics can refer to my previous article-the basic metrics of web analytics, which are listed below for some of the metrics of the page:

Page Views (pageviews)

The number of times the page was opened or requested.

Unique page Views (unique pageviews)

This is a Google analytics above the use of a metric, mainly to avoid repeated loading and refreshing of the page caused by pageviews virtual high situation, so in the same visit repeatedly open the same page, the unique pageviews of the page is always recorded only 1 times.

Number of visits (Visits)

The number of visits to the page, if the independent page to calculate the visits of each page, in fact, the results and the above unique pageviews is consistent, so many Web sites can analyze tools inside there is no unique pageviews, and directly using visits to measure the page's unique browsing volume. However, it is important to note that visits is also often used as a metric for the entire Web site or for some content subtotals, in which case the total visits of the site and the total unique pageviews are inconsistent, such as visit a visited the A-B-A-C4 page, and visit b only access to B, then for a page, pageviews is 3,unique pageviews is 2,visits is also 2, but for this site, the Unique pageviews is 5, and visits only 2.

Number of unique access users (unique visitors)

This should be easy to understand, that is, the number of different IP or cookies entering this page.

Page stay (time on page)

When a user browses a page, the time when the page is to be visited, when the page is measured, it is more about the average number of users, that is, Avg. Time on page = total number of users browsing the page when all users browse it.

Direct jump out of access (bounces)

Perhaps you can see a lot of explanations about this definition, and the general web Analytics tool will be called bounce from the page and go directly to the site. And this metric is more in the form of bounce rate, that is, the number of accesses directly from the page/access from the page.

Number of entry and departure (entrances and Exits)

As the name suggests, the number of accesses to and from the page is typically entered in the form of enter rate and exit rate, the number of accesses from that page, the total number of visits to the page.

There are other page metrics, such as new Visits, target value (Goal value), and so on.

We will note that the results of the general web Analytics tool will have many different reports, and each type of report will be repeated or not repeat the various measures, of course, each report may have its different presentation angle and provide different data analysis purposes. But maybe we don't pay attention to all the metrics on a daily basis, and generally focus on some of the metrics that are closely related to the site's operational status, depending on the characteristics of the site. Several reports on Google analytics content modules typically show:

Pageviews, Unique pageviews, Avg. Time on Page, Bounce Rate,%exit (departure percentage) and $index (target value index)

Here are a few metrics that I personally think are important for evaluating site pages, or those that individuals prefer and care about:

1. Number of visits (Visits) or unique page views (unique pageviews)

These two metrics have been described above, and they are the same for a single page and will only be different when calculating totals. So why choose the amount of traffic, not the number of page views or the number of unique users? My understanding is that in the same visit, if users browse the same page more than once, it is likely that users will like to refresh or reload the page, or users are accustomed to reverse the operation and repeat the shuttle to your site, so these repeated browsing for the evaluation of the merits of the page is not meaningful; and if it's a different visit, Even if the same user to browse the same page, then we can more understand that the user for the last time to see the content is still not done, want to carefully study the end, this kind of page browsing for the evaluation of a page is effective, so I prefer to select the number of pages visits.

2. Failure rate (Bounce Rate)

Well, whatever you call it, you have to admit it's fascinating, the site has a lot of related articles, and before my blog I wrote a question about the definition of bounce rate.

For its value I understand so, users will jump out, nothing more than 3 kinds of situations: 1. The content is too boring; 3. The page to enter is also the export of the website. So pay attention to bounce rate when it is necessary to carry out special circumstances special analysis, such as the home page of the bounce relatively high is understandable, because as the front door of the site, there may be a considerable number of users mistakenly entered; and the site's underlying content Details page bounce Rate relatively high there is a possibility that the user directly into the page, found the required information and left, that is, the 3rd of the above mentioned situation. Excluding these two special accesses, if the other page's bounce rate is high, then the page has a problem.

3. Average page stay time avg. Time on Page)

This should be easier to understand, users need to get some information on your page will need a certain browsing time, of course, according to the content of the page provided by the length and complexity of the average stay time there will be a certain difference, but we do not need to pay attention to such situations, We just need to find those pages where the average stay is short enough for the user to be able to respond effectively to the content of the page, and there is no doubt that the pages are problematic.

Based on the metrics described above, we can select any single or multiple to subdivide the page, and I think of the following methods for subdividing the results:

Based on pageviews, Unique pageviews, uniquevisitors distinguish hot pages and unpopular pages;

Based on Avg. Time on page distinguishes between interesting content and tedious content;

Based on bounce rate and the diversity of the next browsing page to distinguish sick pages and pivot pages;

Identify portal and exit pages based on enter rate and exit rate;

......

Of course, you can also think of a number of other ways to subdivide, as long as the results of the segmentation is valuable to your site. Here's a simple example of using multiple metrics to subdivide, and the metric to choose is the 3 page metrics I think are worth looking at. Since there are 3 metrics involved, most of the planar coordinate charts are not well displayed, and perhaps you have the same idea for a chart-bubble chart, and here's a simple show:

Horizontal coordinates identify the number of visits, ordinate identification bounce Rate, the radius of the bubble R identify the average length of the page, we can divide the page into 4 categories, corresponding to 4 quadrants, including:

First quadrant: Higher traffic and bounce Rate, so the page is more popular but not to attract all the users, perhaps your site's home page or landing page is in this quadrant, you may need to optimize the SEO keyword or buy more relevant ads or keywords, You can also optimize your home page and landing page to make them more attractive to retain users;

Second quadrant: If your site has a lot of pages in this quadrant, then your site is in a bad situation, and often in this quadrant the bubble will be smaller;

Quadrant three: In fact, some pages in this quadrant can increase their access by raising the exposure rate, because they are not less attractive from bounce rate, of course, if certain content is directed at a specific group of people and others are not interested in them, then it is a normal situation in the third quadrant. ；

Quadrant Four: These pages focus on some of the gold content of your site, and see where they attract so many users and let other pages learn about them.

Maybe you can think of other more interesting page segmentation methods and welcome to share the conversation with me.