# 在分类和标签层面追踪页面浏览量和独立用户数

**URL:** <https://meta.discourse.org/t/tracking-of-pageviews-and-unique-users-at-category-and-tag-level/113469>\
**Category:** Data & reporting\
**Created:** [2019年四月4日 19:24 UTC](https://meta.discourse.org/t/tracking-of-pageviews-and-unique-users-at-category-and-tag-level/113469 "2019-04-04T19:24:32Z")\
**Posts on this page:** 20\
**Page:** 1

<div class="post-metadata">

**Author:** ![Curtis\_Kephart](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/curtis_kephart/32/89070_2.png) [@Curtis\_Kephart](https://meta.discourse.org/u/Curtis_Kephart)\
**Post date:** [2019年四月4日 19:24 UTC](https://meta.discourse.org/t/tracking-of-pageviews-and-unique-users-at-category-and-tag-level/113469/1 "2019-04-04T19:24:32Z")

</div>

I am curious if you have advice on tracking community health metrics at the category and/or tag level.  
For example, I’d love to have a clearer sense of the daily number of pageviews and user visits (both known and anonymous users, and ideally by specific group-members) on topic-threads under each category and tag.

In `/admin/dashboard/reports`, new topics and new post have reports that can be filtered to the category level, user profile\_views can be filtered by group, and there are sitewide metrics on page views and other useful info.

I suspect some of this can be set-up somehow in Google Analytics, but that’s not obvious to me since it’s not obvious how to associate topic urls with tags and categories in google analytics. I also suspect this may be done via the `/admin/plugins/explorer`, but I haven’t had any success.

Your advice would be greatly appreciated.

---

<div class="post-metadata">

**Author:** ![codinghorror](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/codinghorror/32/110067_2.png) [@codinghorror](https://meta.discourse.org/u/codinghorror)\
**Post date:** [2019年四月15日 22:56 UTC](https://meta.discourse.org/t/tracking-of-pageviews-and-unique-users-at-category-and-tag-level/113469/3 "2019-04-15T22:56:54Z")

</div>

Do you have any advice here @hawk? It seems like a reasonable question and maybe new dashboard / reports covers this a bit?

---

<div class="post-metadata">

**Author:** ![HAWK](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/hawk/32/86627_2.png) [@HAWK](https://meta.discourse.org/u/HAWK)\
**Post date:** [2019年四月15日 23:11 UTC](https://meta.discourse.org/t/tracking-of-pageviews-and-unique-users-at-category-and-tag-level/113469/4 "2019-04-15T23:11:21Z")

</div>

It does seem like a reasonable request, yes. At the moment the reports on the dashboard aren’t segment-able but if we can work out a way to do that I think it would be an excellent addition to v2. We’ve also talked about exposing approved [Data Explorer](https://meta.discourse.org/t/32566?silent=true) queries on the dashboard.

cc @j.jaffeux – do you have plans to revisit the dashboard any time soon?

---

<div class="post-metadata">

**Author:** ![j.jaffeux](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/j.jaffeux/32/60297_2.png) [@j.jaffeux](https://meta.discourse.org/u/j.jaffeux)\
**Post date:** [2019年四月16日 08:31 UTC](https://meta.discourse.org/t/tracking-of-pageviews-and-unique-users-at-category-and-tag-level/113469/5 "2019-04-16T08:31:36Z")

</div>

Yes, it will be step by step but might come to something close to this. I don‘t want it to become too complex though and it will never be a “combine any segment you want”, there’s sql (and data-explorer) for this. But will probably try to provide this on a per report basis where we think it makes sense.

---

<div class="post-metadata">

**Author:** ![HAWK](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/hawk/32/86627_2.png) [@HAWK](https://meta.discourse.org/u/HAWK)\
**Post date:** [2019年四月17日 01:22 UTC](https://meta.discourse.org/t/tracking-of-pageviews-and-unique-users-at-category-and-tag-level/113469/6 "2019-04-17T01:22:09Z")

</div>

This query is specifically about _pageviews_ per category, which I don’t think can be handled with [Data Explorer](https://meta.discourse.org/t/32566?silent=true) can they?

---

<div class="post-metadata">

**Author:** ![skepto](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/skepto/32/117985_2.png) [@skepto](https://meta.discourse.org/u/skepto)\
**Post date:** [2019年四月25日 10:17 UTC](https://meta.discourse.org/t/tracking-of-pageviews-and-unique-users-at-category-and-tag-level/113469/7 "2019-04-25T10:17:13Z")

</div>

Yes, I would also be interested in a report about the unique users on my site.

---

<div class="post-metadata">

**Author:** ![j.jaffeux](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/j.jaffeux/32/60297_2.png) [@j.jaffeux](https://meta.discourse.org/u/j.jaffeux)\
**Post date:** [2019年四月25日 15:38 UTC](https://meta.discourse.org/t/tracking-of-pageviews-and-unique-users-at-category-and-tag-level/113469/8 "2019-04-25T15:38:10Z")

</div>

Yes this specific report is using ApplicationRequest table, which doesn’t have any notion of category. So not possible through reports, data-explorer or raw sql.

---

<div class="post-metadata">

**Author:** ![Dawn](https://avatars.discourse-cdn.com/v4/letter/d/71e660/32.png) [@Dawn](https://meta.discourse.org/u/Dawn)\
**Post date:** [2019年六月7日 20:04 UTC](https://meta.discourse.org/t/tracking-of-pageviews-and-unique-users-at-category-and-tag-level/113469/9 "2019-06-07T20:04:59Z")

</div>

Can we pull a report on the number of anonymous users instead of the number of pages viewed anonymously?

---

<div class="post-metadata">

**Author:** ![HAWK](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/hawk/32/86627_2.png) [@HAWK](https://meta.discourse.org/u/HAWK)\
**Post date:** [2019年六月9日 22:49 UTC](https://meta.discourse.org/t/tracking-of-pageviews-and-unique-users-at-category-and-tag-level/113469/10 "2019-06-09T22:49:53Z")

</div>

Do you mean the number of people that have used the anonymous posting feature? They all get unique anonymous usernames so it will be very easy to pull a report from [Data Explorer](https://meta.discourse.org/t/32566?silent=true)

---

<div class="post-metadata">

**Author:** ![Dawn](https://avatars.discourse-cdn.com/v4/letter/d/71e660/32.png) [@Dawn](https://meta.discourse.org/u/Dawn)\
**Post date:** [2019年六月10日 14:21 UTC](https://meta.discourse.org/t/tracking-of-pageviews-and-unique-users-at-category-and-tag-level/113469/11 "2019-06-10T14:21:11Z")

</div>

I’m trying to find out how many people are visiting the site who have not logged in at all and do not have an account.

---

<div class="post-metadata">

**Author:** ![Krischan](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/krischan/32/47290_2.png) [@Krischan](https://meta.discourse.org/u/Krischan)\
**Post date:** [2019年十月14日 14:12 UTC](https://meta.discourse.org/t/tracking-of-pageviews-and-unique-users-at-category-and-tag-level/113469/12 "2019-10-14T14:12:26Z")

</div>

也很想看看这个！关于如何构建查询有什么建议吗？谢谢！

---

<div class="post-metadata">

**Author:** ![mrjn](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/mrjn/32/121540_2.png) [@mrjn](https://meta.discourse.org/u/mrjn)\
**Post date:** [2020年七月7日 17:28 UTC](https://meta.discourse.org/t/tracking-of-pageviews-and-unique-users-at-category-and-tag-level/113469/13 "2020-07-07T17:28:06Z")

</div>

我们还需要这个数字（每个类别的页面浏览量，排除爬虫活动）。是否可以通过数据探索器获取？

---

<div class="post-metadata">

**Author:** ![HAWK](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/hawk/32/86627_2.png) [@HAWK](https://meta.discourse.org/u/HAWK)\
**Post date:** [2020年七月7日 19:58 UTC](https://meta.discourse.org/t/tracking-of-pageviews-and-unique-users-at-category-and-tag-level/113469/14 "2020-07-07T19:58:02Z")

</div>

我不这么认为。我们不会像那样细致地追踪页面浏览量。你需要接入你的 GA 并使用它。

---

<div class="post-metadata">

**Author:** ![mrjn](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/mrjn/32/121540_2.png) [@mrjn](https://meta.discourse.org/u/mrjn)\
**Post date:** [2020年七月7日 21:06 UTC](https://meta.discourse.org/t/tracking-of-pageviews-and-unique-users-at-category-and-tag-level/113469/15 "2020-07-07T21:06:40Z")

</div>

@HAWK：那么你认为这不可能实现吗？[Counting Pageviews for users (non-staff, non-crawler)](https://meta.discourse.org/t/counting-pageviews-for-users-non-staff-non-crawler/157071)

---

<div class="post-metadata">

**Author:** ![jrgong](https://avatars.discourse-cdn.com/v4/letter/j/c57346/32.png) [@jrgong](https://meta.discourse.org/u/jrgong)\
**Post date:** [2020年十二月1日 22:55 UTC](https://meta.discourse.org/t/tracking-of-pageviews-and-unique-users-at-category-and-tag-level/113469/16 "2020-12-01T22:55:42Z")

</div>

有人找到统计特定分类下所有帖子页面浏览量的解决方案了吗？用 GA 可以实现吗？

---

<div class="post-metadata">

**Author:** ![jericson](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/jericson/32/116215_2.png) [@jericson](https://meta.discourse.org/u/jericson)\
**Post date:** [2023年五月5日 23:01 UTC](https://meta.discourse.org/t/tracking-of-pageviews-and-unique-users-at-category-and-tag-level/113469/17 "2023-05-05T23:01:59Z")

</div>

您可以通过汇总类别下所有主题的浏览量，从数据浏览器中获取\_总\_类别浏览量：

```plaintext
SELECT 
       c.id as category_id, 
       SUM(views) as "total views"
FROM categories c 
JOIN topics t ON t.category_id = c.id
WHERE read_restricted is false
GROUP BY c.id
order by sum(views) desc

```

按标签获取总浏览量的方法类似：

```plaintext
SELECT tags.name,
       sum(views)
from topics t
     join topic_tags tt on t.id = topic_id
     join tags on tags.id = tt.tag_id
group by tags.name
order by sum(views) desc

```

还有一个 `topic_views` 表，理论上可以用来按用户和日期划分浏览量。然而，我发现这个表不太有用，因为当主题浏览量很大时，查询会超时。它还显示每个主题的浏览量少得多，因为如果我理解正确的话，它不计算匿名浏览量。我还认为 `topic.views` 可能不仅包括匿名浏览量，还包括机器人流量？这比我在 GA 上看到的要多得多。

说到 GA，它拥有您需要的所有数据，但很难按类别或标签进行分组。我能做的最好的事情是尝试解析 `pageTitle` 来查找类别。我正在使用 [googleAnalyticsR](https://cran.r-project.org/web/packages/googleAnalyticsR/index.html) 在 R 中执行此操作。请遵循手册中的说明授权您的 Google 帐户并获取您想要的指标。请务必将 `pageTitle` 包含为维度。我的 API 调用大致如下：

```plaintext
ga_this_year <- ga_data(ga_id, 
                        date_range = c("2023-01-01", "2023-05-30" ),
                        metrics=c("screenPageViews","averageSessionDuration", "sessions"), 
                        dimensions = c("pageTitle", "deviceCategory")
                       )

```

理解下一部分的关键是看到 `pageTitles` 符合此通用格式：

```plaintext
Topic title - Category - Site title

```

如果主题未分类，则“Category”将缺失。还有很多实用页面（“Latest topics”、“New topics”等）没有类别。（我没有将“Latest [category] topics”视为类别的一部分，尽管包含它们可能更好。）最后，主页使用 `Site title - short site description` 作为页面标题。然而，我们不关心其中任何一个。所以，我使用的正则表达式是：

```plaintext
str_extract(pageTitle,
            str_glue(".*? - (.*) - {sitename}"),
            group=1)

```

将其组合成一个按类别分组的函数，我得到类似这样的结果：

```plaintext
category_views <- function(data, sitename = "Meta Jon") {
  data %>% 
  mutate(category = str_extract(
                                pageTitle,
                                str_glue(".*? - (.*) - {sitename}"),
                                group=1
                                )) %>% 
  group_by(category) %>% 
  summarise(views = sum(screenPageViews)) %>% 
  arrange(desc(views)) #%>% head(20)
}

category_views(ga_this_year)

```

（显然，请将“Meta Jon”替换为您自己的网站名称。）

不过，我目前不知道如何根据标签提取 GA 数据。

---

<div class="post-metadata">

**Author:** ![Jagster](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/jagster/32/192154_2.png) [@Jagster](https://meta.discourse.org/u/Jagster)\
**Post date:** [2023年五月7日 11:17 UTC](https://meta.discourse.org/t/tracking-of-pageviews-and-unique-users-at-category-and-tag-level/113469/18 "2023-05-07T11:17:02Z")

</div>

这些总观看次数是否包含机器人，也就是匿名用户？

---

<div class="post-metadata">

**Author:** ![jericson](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/jericson/32/116215_2.png) [@jericson](https://meta.discourse.org/u/jericson)\
**Post date:** [2023年五月8日 00:46 UTC](https://meta.discourse.org/t/tracking-of-pageviews-and-unique-users-at-category-and-tag-level/113469/19 "2023-05-08T00:46:59Z")

</div>

GA 根据[其文档](https://support.google.com/analytics/answer/9888366?hl=en)排除了机器人流量。我找不到 `topic_views` 的文档，但[代码](https://github.com/discourse/discourse/blob/main/app/models/topic_view_item.rb#L13)中的一条注释说：

```plaintext
# 每天每个事物每个（用户 || IP）只存储一次视图

```

我不确定，但 `topic.views` 似乎显示了机器人的页面浏览量，因为它显示的浏览量比 GA 同一页面的浏览量多得多。

---

<div class="post-metadata">

**Author:** ![Jagster](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/jagster/32/192154_2.png) [@Jagster](https://meta.discourse.org/u/Jagster)\
**Post date:** [2023年五月8日 06:32 UTC](https://meta.discourse.org/t/tracking-of-pageviews-and-unique-users-at-category-and-tag-level/113469/20 "2023-05-08T06:32:57Z")

</div>

是的，但使用数据浏览器时来自 Discourse 的计数是多少？

早些时候 GA 也这么说过，当有合法的机器人时，这可能是真的。但大多数调用是由恶意机器人、行为不端的 SEO 机器人、敲诈者等引起的，然后一次访问规则就会被打破。如果我诚实地说，我就是这么认为的。

无论如何，我正在使用 Matomo。我不是那种需要 GA 的大人物。

---

<div class="post-metadata">

**Author:** ![JammyDodger](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/jammydodger/32/254611_2.png) [@JammyDodger](https://meta.discourse.org/u/JammyDodger)\
**Post date:** [2023年六月8日 06:56 UTC](https://meta.discourse.org/t/tracking-of-pageviews-and-unique-users-at-category-and-tag-level/113469/21 "2023-06-08T06:56:53Z")

</div>

> [@jericson](#):
>
> 它还显示每个主题的浏览量少了很多，因为，如果我理解正确的话，它不计算匿名浏览量。

仅供参考，这实际上是因为 `topic_views` 表限制为 ~~每天只计算一次新浏览~~ ，而 `topics.views` 字段允许每 8 小时（默认情况下，但可以使用 `topic view duration hours` 更改）计算一次新浏览。它确实包括用户和匿名用户。👍

**更新**  
事实证明，`topic_views` 表甚至不计算每天一次的新浏览……它实际上计算的是 **某人第一次浏览某个主题** ，之后不再计算。因此，每个用户或 IP 每个主题只会有一个记录。

[下一頁](https://meta.discourse.org/t/tracking-of-pageviews-and-unique-users-at-category-and-tag-level/113469.md?page=2)