For most websites and apps, visitors go through a long dating process, not a one-night stand: they rarely convert after the first visit. Most commonly, the first visit is used togauge credibility。随后的访问可能涉及更复杂的相互作用,并且可能是转换事件,可能发生weeks or months after the initial visit。Thus, to better support our users, we need to understand the actual dating process for our site visitors. Do people use our site every day or even multiple times per day? How many visits precede that final conversion? Does this behavior fluctuate based on the time of year?


Definition:Frequencyof site visits indicates the overall number of visits made by each user on your site. This metric allows you to assess the percentage of new users on the site as well as the familiarity level of all returning users.


Combined, frequency and recency measure how “sticky” a website or app is — do people return regularly to it or not? Analyzing the behaviors of loyal customers can lead to valuable insights — for example, you may find pages or tools used frequently by returning users, but ignored by occasional visitors (particularly “siteseers” who only visit once). When you uncover such trends, you can then adjust the design of the site to increase the discoverability of those features that your repeat users find valuable.

Frequency: How Many Visits?

When using analytics tools such as Google Analytics or Adobe Analytics, frequency is described by a histogram of the number of visits. (A histogram is a type of bar chart in which each bar shows the number of data points that have a certain value or fall inside a range of values.) In Google Analytics, this histogram is called the会话计时报告和in Adobe Analytics it is thevisit-number report

这样的报告不仅可以衡量用户是新用户还是新用户,还可以让我们更准确地了解用户的熟悉程度:有多少访问者对网站非常熟悉(比如说,有过6次访问)thvisit or more) versus how many are relatively fresh to the site?

Count-of-session (or visit-number) reports are usually defined for a specific period of time T (e.g., August) and they tell us, for each possible number of visits (i.e., count of sessions) n, how many users have had their nth visit within the time period T. So, for example, for count of sessions n=2 and the month of August, the histogram would tell you how many users had their second visit in August. (Note that some of these users may have had their first visit before August and some may have had it in August — they are all included in the count of sessions for n=2, as analytics tools report visit number as a lifetime value for each user.)

Screenshot of frequency report in Google Analytics
Google Analytics count-of-sessions report for all site visitors during the month of August: The first row shows that 225,966 users had their first visit during this time window, the second row shows that 47,859 users had their second visit, and so on. Note that the bins in this histogram are not equally spaced: the ones towards the bottom include bigger and bigger ranges instead of a single number n.

用户可以包含在几个会话计数器中:例如,在8月份的前两次访问的用户将包括在n = 1的会话的数量中,并且对于n = 2。如果有更多的新访客而不是返回,你会期待第一个酒吧(对于n = 1)比其他酒吧更大。

为了获得有关特定类别的用户的更详细数据,我们需要创建一个特殊的segment(a subset of the data obtained by applying a chosen filter). For example, to see the frequency graph for those users who visited only one time in a given month, we would specify the rule “Sessions =1” as a filter to create a segment.

Screenshot of frequency report for a segment of users in Google Analytics
此Google Analytics报告显示了在一个月内完成1次会话的用户段的会话计数。在此期间有191,800名用户首次和唯一一次访问;在此期间有9,106名二次访问者,只有一次访问,等等。您可以看到,尽管所有这些用户在给定的时间范围内完全访问了该网站,但他们根据他们过去的其他会话进行了不同的会话垃圾箱。

In isolation, looking at a frequency report does not reveal much because the rows generally just decrease in number as the count increments (not particularly surprising — as you increase the loyalty level, you can expect fewer and fewer users at that level). The real value emerges from looking at the segment of visits where important events (such as conversions for a certain goal) occurred and determining how many other visits preceded the event of interest.

Screenshot of frequency report for a segment of sessions in Google Analytics
此Google Analytics频率报告仅显示了一段时间的会话段(例如,产品)在指定月份完成了一定目标(例如,销售)。因此,我们看到307个用户在他们的1st在第二次访问中转换后,231名用户访问该网站,等等。

通过揭示用户在转换之前的典型访问数量,我们可以更好地为用户的实际旅程设计。例如,如果在用户5上发生大部分转换th或6.thvisit, we can implement tools to remind people what they were doing in the past to help them pick up where they left off, email users with only 3 or 4 visits to persuade them to revisit the site, and so on. We could alsoplan a usability studydiary study要了解转换的障碍可能是什么:也许需要这么多访问转换,因为用户难以将我们的产品与竞争对手的产品进行比较,或者因为他们认为它需要太长并开始注册并开始。

新闻: When Did People Last Visit?


在Google Analytics中,对于新的用户来说,即新的重新值(天自去年会话) is recorded as 0 days —the same as for users who have visited twice (or more) during the same day. Thus, when looking at Google Analytics recency reports, it is helpful to filter new users out by applying a segment including only multisession users. In Adobe Analytics, the return-frequency report already excludes first-time visitors. (Other tools may report this type of data using either method, so be sure to check how your analytics tool functions.)

在Google Analytics中的2个细分报告的入门报告屏幕截图
此Google Analytics报告显示了在指定月份访问该网站的所有用户自上次访问以来的天数。报告显示两个用户段的数据:所有用户(蓝色酒吧)Multi-session Users(orange bar). The latter 仅包括在给定的时间范围内有超过1访问的用户。新用户(没有以前的访问)被安置在0箱,与同一天有两次访问的人一起:这就是为什么All Usersbar is so much larger than theMulti-session UsersRENGY值为0 - 新用户往往是大多数人。


For example, in the screenshot above, we can see that most multi-session users return to the site within the same day as the previous visit (the longest bar is the one in the row天自去年会话= 0) or just 1 day later. Notice also a small bump at the天自去年会话=6. In this case, the bump corresponds to a weekly email newsletter.

Segments can help uncover desirable usage patterns — in particular, those of frequent visitors vs. those of occasional visitors. If people who return every day or every other day tend to convert more, whereas less frequent visitors end up never converting, then you should investigate what other behaviors are different between the two groups. Is there a feature or tool that more-engaged visitors use and that you could make more discoverable for everybody? Are frequent visits triggered by certain email promotions? Do frequent visitors have a different experience due to being kept logged in to the system? Different times of year can also be compared to uncover usage trends based on seasonality or specific marketing efforts.



Frequency and recency data are helpful to better understand thecustomer journeyof your users, as well as their needs and behaviors. They can help create or保持你的角色, and also discover how to better support not only your visitors, but also your business goals. How many times do people visit before they finally convert or take some other desirable action? Do those visits occur within a short time frame or are there large gaps between sessions? Such knowledge can be used to turn visitors into valuable, engaged users.

Learn more about using analytics data for UX insights in our关于分析和用户体验的全日制培训课程