Product operators must know their index data

 

Article reprinted from the public No.  PB Products PB product, author Soso

        

Data indicators, most of the time can really measure the performance of a product is good or bad, only that can help us find the problem and avoid racking our brains type of thinking patterns.

Recently often encounter some colleagues to ask questions about the data type of the index, so the current journal first to tell you about some of the knowledge-based indicators point data.

We basically can be broken down into user behavior index the following types:

Click Browse categories of indicators

User Activity class index

The final conversion metrics like

Our basic indicators can be broken down into the following types:

1, click Browse categories of indicators

We know that clickstream data derived from a lot of behavioral indicators, such as: frequency of visits, average length of stay, click on browsing behavior, information interactions, content publishing behavior. But these indicators are too complex and not conducive to our quick analysis of the user, then the how these indicators effective and simple and division, thereby helping us to quickly analyze a user to operate it?

Speaking generally, or click Browse tend to think that a certain type indicator associated with the page. Different pages will be used in the analysis of different indicators and methods.

Page can usually be divided into three categories:

- shunt type (Home, Landing Page, etc.)

Page layout used for drainage users pay more attention to page layout information architecture, placing the effectiveness of the module, with the highest click conversion output target.

- functional class (page flow, page Submit etc.)

Page Layout greater emphasis on the successful completion of the main flow, does not distract the user, it is often to improve the conversion process, reducing the intensity of the user fee goal.

- a list of categories (product list, search lists, etc.)

Page layout used for display or product orders, product content layout, copywriting and guide the effectiveness of promotion, of course, is a measure of the important indicators page, but at the same time bring the real deal page subsequent transformation is equally important. Description by browsing the list of classes page allows users to really find the products they need.

1) shunt category pages most commonly measured indicators: click conversion rate (CTR)

This indicator is the most simple and direct measure of page layout and information architecture attractive promotional copy.

Click molecules is relatively easy to understand, when the user clicks one will Click + 1.

But when PV as the denominator, it will be understood as when the page is loaded once the browser how many clicks will produce , if re-entered each time the page will be executed PV + 1, then, while the page is not the current page jump click (usually called as an anchor point), the ratio should be less than 1.

But as the APP will encounter different ways to calculate PV, after leaving the home page click on the page, then click on the other pages returned back home, after some PV companies will not return as the PV + 1.

But I personally feel that no matter when and where as long as the page is re-opened should PV + 1, because as a user, he once again see the page, click on the likelihood of behavior they may have on the page. Such a ratio is relatively fair.

But sometimes we also use UV as the denominator, meaning it will change, visitors visit each page, click generated . Of course, this ratio is certainly greater than Click / PV, as a visitor browsing typically produce a plurality of number of pages, and therefore should be less than UV PV.

Once the time before the revision, we met before and after comparison of what should be revised with PV or UV as the denominator.

We can see that this revision will be the most important thing is the head of the TAB A program to switch to delete, merge into a unified structure of the page B program.

As an impartial attitude towards A and B, it should be clear that the sum of A click of TAB1,2,3, divided by three TAB sum of PV. However, due to the TAB click will result in subsequent pages PV, click on the page itself, but did not bring traffic distribution. Therefore, a departure from our click conversion goals.

Scheme below, assume that we need to count C click function, and returns to the page Click / PV, is easy to see by analyzing the equation will vary.

After re-think deeply, we find that using the Click / UV is more just for the page, but here's the UV, not simply UV1 + UV2 + UV3, because three plus there will always lead to a simple UV access at the same time the situation of those who visit TAB1,2,3 occurs, UV double counting, so we need to do here is to re-UV.

与此同时,在分子Click中,我们需要删除TAB的点击,因为TABclick对于页面点击分流并没有带来贡献,如果加上TAB的点击显然对于方案B是不公正的计算。再套回刚才的场景,Click(去除TAB)/UV(去重),结果都等于1。

通过以上深入分析,不难发现一个页面需要通过仔细的琢磨,选择正确的指标才能得到更为合理和公正的结论。

最近,频频被提起的还有一个类似点击转化率的概念,就是点击曝光比。

这个指标,如果用来比对同个页面中的区域或者内容的转化,则会有和点击转化完全不同的目标和结论。

曝光量分为页面曝光,区域曝光和内容曝光。其实对于页面维度来说曝光量应该基本等于PV,这个数值之间相差量应该微乎其微。

区域曝光是指该区块大于1/2的内容在用户可视范围内。对于首屏而言,由于不同机型和分辨率差异,用户的可视化区域会有所不同,故首屏越靠上的区域的曝光会越接近页面PV。越往下会越呈现下降的趋势。那么当页面中的不同区域进行点击曝光比时,通常可以看出每个区域的转化效率。我们可以通过调整区域的位置,将页面曝光量最高的给到产出点击更多的区域,那么在这个调整中的依据,就可以使用点击曝光比。

举例来说:首屏有某一个区域的点击曝光比非常低,三屏理财区域点击曝光比却很高。我们是否可以考虑将首屏这个区域和理财来进行对换,来观测点击曝光比是否都有提升。

内容曝光则指区块内的具体内容是否在用户可视范围之内。

针对某个位置会有不同客群,展示不同内容入口。当这个时候,这个指标就显得格外重要。因为我们可以通过点击曝光比,来比较内容入口对于客群的吸引力。如果效果不佳,可以调整内容的同时也可以同样考虑改变客群精准或者减少客群的比重,来增加页面的分流效率。

2)分流类页面另一个重要指标:访问深度转化率(DV Rate)

访问深度转化率。是我在eBay时使用最频繁的另一个点击类的数据指标。网站访问深度就是用户在一次浏览你的网站的过程中浏览了你的网站的页数总数,简称DV(Depth View)。那么访问深度转化率DV Rate=DV/Click。举例来说,如下图,是两个用户通过点击Banner产生的网站访问行为,需要计算这个Banner的访问深度转化率。

 

小A点击Banner之后产生了5个页面的浏览,而小B产生了4个页面浏览,所以DV:5+4=9。而Banner的点击次数之和为2,因此DV Rate=9/2。

如果说CTR可以作为页面分流效率的基本指标,那么访问深度转化率就更为综合。因为点击之后,在网站中产生更多的浏览,这并不是一个Banner的设计,或者模块入口的设计能够做到的。而是需要的整个APP的体验,商品的吸引力,包括页面中的互链、流程等等都非常重要。

3)流程类页面:漏斗转化率

针对流程类页面,最直接的做法,就是直接用后一个页面UV去比前一个页面UV,就好像漏斗漏到最后完成页的比率,来观测流程页面。

 

通过漏斗转化率,我们不仅可以知道,从开始流程到完成页我们的转化比率是多少,我们可以知道每个步骤的转化。对于流程分解的比对是可以了解每个转化环节的问题。

举例:

这个流程是养车洗车类的APP其中某两个流程的漏斗转化。通常来说洗车的转化成功率相对其他产品的提交转化会高出许多,但是通过前图,我们会发现保养手册到选择服务页的转化却是漏斗转化中最高的,甚至超过洗车详情页至提交成功页。但是选择服务页到提交成功页的转化却低了很多。漏斗转化低的环节通过更进一步的调研和分析之后,我们了解到真正原因,产生在选择服务页中所推荐的产品,不是用户真正需要的。后期改版中,将推荐逻辑进行调整之后,在观测前后的漏斗转化产生的变化,就可以更容易了解流程页面改版的好坏。

2、用户活跃类指标

看到这个指标,大家一定异口同声说,不就是月活或是日活么。但是月活和日活却各有各的侧重点。

日活和月活

DAU(Daily Active User)日活跃用户数量。DAU通常统计一日之内,打开APP(或者登录)的用户数(去重)。

MAU(Monthly Active Users)是指APP月活跃用户数量(去重)。

日活,决定了你的生命基础;月活,决定了潜在生命力。如果针对的用户需求是日常的,比较高频的需求,那么就应该看日活,如果使用频率相对降低,那么更适合看月活。如果月活在增加而日活在减少,就可能是用户的重复使用率在降低,或者新用户的留存度和活跃度都不高,需要改善。运营策略和产品迭代的反映最先都是出现在日活上,各种决策也需要以日活指标来调整。

产品运营被赋予的任务之一,就是必须将月活/日活比减小,最大程度提升——目标朝向1:1(当然是不可能的)。通过对比同行业的月内使用周期就可以知道产品的极限所在。汇总各领域的标杆APP数据,还可以了解各类市场的人群大小和需求频次。下图可以看出支付宝月活/日活将近4,而微信最小,接近1.5。

图片摘自互联网

除了日活和月活外,另外一个指标也同样重要。

用户留存率

这个对于一个APP来说是一个非常重要的数据指标。统计的是一个用户在下载首次打开APP过后,间隔一段时间之后仍然打开APP的比例。通常有次日留存,7日留存,14日留存和30日留存。这也是一个统计用户粘度和流失的重要数据。一般来说一个APP会随着时间越久留存会渐渐变少。

以次日留存为例,次日留存=第一天的新用户中第二天仍然打开APP的数量/第一天的新增用户总数,依次类推。

留存率当中最常用的就是时间分组,下图就是留存率数据表现:

图片摘自互联网

我们需要了解新用户的留存现状进行分析,通过对新访问用户和全部访问用户的留存曲线对比分析,我们会发现新用户的留存明显低于全部用户,那么从这个角度来说,新用户的留存就是很大的一个增长点。

对于银行信用卡用户来说,最开始激活的环节非常重要。能否让用户在第一次使用产品时就能迅速低成本地感知到信用卡的价值, 决定了新用户的激活率和首刷率。

图片摘自互联网

所以如果要提升新用户的留存,一方面要降低激活的使用成本,另外就需要让用户明确知道自己通过激活首刷后可以获得的好处(而这个需要创造触达用户的渠道,短信方式显然是不够的),用微信的简单途径促进用户的激活,不仅可以降低用户激活的使用成本,同时还可以额外增加触达用户的途径,其实还是很不错的选择。

3、最终转化类指标

订单/Buyer转化率:对于OTA也好,还是一个B2C的企业来说,订单转化率一定是最重要的指标。

携程酒店产品的核心指标CR,其实也是订单转化率。

但是也有例外,在eBay的时候,并不看订单转化率,看的是Buyer转化率。对于C2C企业,并不关心究竟成交的订单单数多少,也不关心客单价,最关心的是在平台上进行交易的人数,因此更关心的是Buyer转化率。

除此之外电商或者OTA比起转化还关心收益,是否能产生盈利。因此还有一个指标叫GMV。常常用GMV代指网站的成交金额,主要包括付款金额和未付款的。

通俗来说,我们平时网购时会进行下单,产生的订单中往往会包括付款订单和未付款的订单,而gmv统计的指标就是其二者之和。利用GMV可以进行交易数据分析,虽然GMV不是实际的购买交易数据,但同样可以作为参考依据的,因为只要你点击了购买,无论你有没有实际购买,都是统计在GMV里面的。所以,可以用GMV来研究顾客的购买意向,顾客买了之后发生退单的比率,GMV与实际成交额的比率等等。

我们为什么要了解最终转化类指标?

《彭博商业周刊》曾经刊登过一篇文章,叫做沃尔玛与亚马逊的电商之争,文章中提到沃尔玛在被亚马逊打的节节败退,虽然主要原因是错失了在线商城的先机,但是另外一个很重要的原因就是对待盈利问题的高层态度。亚马逊在与沃尔玛开战的最初就已经表示,他们只在意占领的市场份额,而并非利润和收益。但是反观沃尔玛要承担很多的盈利压力,因此在两强争斗的过程,一个就好像不怕死的拼命三郎,另一个则瞻前顾后,无法全力以赴。最终错失良机。

Therefore, a company's core indicators tend to guide the future strategic direction of the company and, we, as members of the company, the company must understand the core of the final conversion metrics, and by means of product operations of companies to help complete this target.

As a product operator, the data is the basis for product design. No data are based on product design in bullying.

Welcome to public concern number:

        

Welcome to communicate -

 

Published 137 original articles · won praise 93 · views 160 000 +

Guess you like

Origin blog.csdn.net/livan1234/article/details/100588398