The original meaning users portraits, to help businesses find the target user , clear out their likes and dislikes , and thus optimize product features and services , ultimately creating a more business and social value.
Users portrait of the development so far, it has been further enhanced usability. Of course, it is still the foundation of user tags , but under a large data addition, many companies able to extract more user base characteristics and behavioral characteristics from the mass of data to enrich user tags, which allow users to become more three-dimensional portrait and real.
Basic features , refer to the user's basic information, that is, the user attribute information, such as age, gender, consumption levels and other information.
Behavioral characteristics , referring to the statistical analysis of the data behind the behavior and user behavior on the site or App, to arrive at the online behavior of the user preferences, such as an airline customer service for white collars, preferences morning office like the use of medical and health class App.
Combination of the two, formed the cornerstone of the user's portrait, the portrait has become a bottleneck in the system user foundation Bo Hou.
The richer user tags, the user full portrait. The portrait of the plump user, the more support APP can make the right decisions.
The question is, how APP filling the fullness user picture it, can not make bricks without straw, rice where to look?
Housewife is not difficult to find rice
First, we have to be disassembled for user portrait, put this thing down into the following steps:
From the chart, we can see that the user has a portrait label three sources :
Based on the business scene from the abstract , particularly before the APP on the line, we can not get into the user's behavior data, through research on the business scene, you can tease out the initial user tags;
Based on existing data from precipitation , the APP on the line, with the development of the business generated a lot of business data and user behavior data, refined user can continue filling the portrait;
基于决策模型的数据归因验证而来，决策模型输出的决策会不断产生新的数据来验证与完善用户画像，所以引入模型是提高用户画像置信度的有效手段。
有意思的地方来了，在产品设计、规划、冷启动，乃至产品萌芽期，APP开发者需要锁定产品的目标人群，从而让产品功能更加聚焦，达到最小损害、最大收益的目的，如果仅通过用户调研与场景分析构建用户画像，因为缺乏了数据验证，画像的置信度自然会被降低。
不管是初创公司还是成熟大厂，在打造一款新产品的时候，总是要经历这个从无到有的过程，这也就意味着，我们总要经历产品数据的积少成多。当数据不足以验证画像置信度的时候，怎么办？亦或是产品相对成熟，数据有所积累，但终究尺有所短、寸有所长，想进一步做画像补全，怎么办？
正所谓是：巧妇想做饭，米从何来哉？
此时，找一个像个推这样的优质第三方画像服务供应商成了最能保障效率的选择。
个推深耕于开发者服务领域多年，沉淀了维度全、覆盖面广的深厚数据资源。个推面向开发者推出的用户画像产品——个像，全面覆盖用户属性、兴趣、行为、场景等各维度细分标签，能够帮助APP开发者360度勾勒立体用户画像，深入洞察用户，精准把握受众。
适用的第三方用户画像服务，满足了我们找米的期望。在期望之外，像个推这些做用户画像的企业，还提供了可参考的标准标签与策略模型，甚至考虑到了应用场景。这样一来，相当于不仅给了米，还给了口锅，并附带了将米加工成美食的菜谱。
引他山之石，切合自身业务特征，低成本、快速建立用户画像体系来攻本山之玉。这么看来，巧妇找米并不难。
产品腾飞的助跑器
很多企业都在推进自己的用户画像系统建设，但常常没有把用户画像的战斗力发挥出来，甚至沦为了摆设，所以这里有两点需要强调一下：
用户画像需要范围边界，精而专就好，大而全反倒是浪费资源，动辄成百上千个维度只会增加使用难度；
用户画像需要评判标准，用户画像的决策模型是产品与运营的战略资源，能否具备可用的决策模型也就成了评定用户画像优劣的重要标准。
当我们打造出符合自身业务特点的用户画像体系，收获的季度就来了。
业务与产品验证
从点子到产品，APP开发者不断地设想着自己的目标用户群体，从最初的调研访谈到用户画像系统的建设，都是为了达成这样的目标。
在用户画像体系相对完善的基础上，开发者可以根据条件的组合，筛选出对应的用户群体，用目标用户的标签数据去匹配现有的业务数据，可以验证业务方向是否符合原有预期，更可以在业务推广时，进行推广质量评估。
另外，用户画像可以辅助产品设计，按照用户群体特征进行用户分组，根据用户分组数据验证产品功能的使用效果，用数据量化产品功能，找到产品改进关键点，从而保障产品设计不偏离既定航道。
精细化运营
不少人认为产品相对成熟之后才需要考虑精细化运营，从而在产品萌芽期采用了相对粗放的运营策略。但是，如果可以定点射击、百发百中，为什么还要漫天扫射、浪费子弹呢？
在产品启动的那一刻，引入第三方的画像工具，快速构建用户画像体系，通过自有产品数据进行修正，不断提高用户画像置信度，利用推荐算法辅助产品的营销推广，这不仅可以缩短用户的选择路径，提高用户转化率，还能实现“千人千面”的精准营销。
当画像内容积累到一定的程度，使用算法对已有的海量数据进行处理，构建出智能服务模型，当数据产生“思考”的那一刻，用户画像所带来的想象空间将变得无比辽阔，不管是消息推送、个性化推荐还是精准搜索，都会变得更加善解人意。
In practice, with the message push service started a push , in order to provide a more extreme push service to APP, giving users more intimate quality experience, has achieved self-portrait of the user with the message push functionality between the two products get through . APP can help a user to push a rich portrait label dimensions, the user cluster analysis, the development of more refined user models, and insights can be applied directly to the user's portrait message push to further enhance the accuracy of contact information, effective to enhance user stickiness, thus forming a virtuous cycle can continue to improve the degree of "fine operation" APP.
data service
Recently, more and more hot data sets, user portrait system as an important part of the data in the table, is an important user data query platform, but also user data visualization platform, the user is presented the results of the analysis platform.
More importantly, in the context of big data, user portrait no longer static as in the past to present, but you can get real-time dynamic display of user data, which have the ability to support data services marketing decisions.
Moreover, the data service with the maturity, user portrait system big data and algorithms can provide is no longer confined to showing data visualization and analysis of the results can even be further achieved in forecasting demand, to insight into the market and the incremental space the development trend of existing products.
The increasingly fierce business environment for enterprise product a higher liquidity requirements, companies not only want to live, but also want to live well. Opened to door to the future, companies need to pay a lot of effort, and user portrait is one of a stepping stone, but the door open to help sail voyage breeze.
Good wind power, send me on, user portrait as yes.
Author: Zhang ink, Internet practitioners, product manager of Wen Qing children, micro-channel public number: Moonlight tank (moontank1918)