This is the 31st day of my participation in the August Text Challenge.More challenges in August

Buried point analysis is basically all about the number of points will be mentioned, after all, is one of the important sources of web data. In general, I will talk about the concept of buried point analysis, several ways of buried point analysis, and the whole process of buried point. But how to more accurately design a site buried sites, because buried sites everywhere will lead to too many useless buried sites. And if it is the front end buried point, how to verify the correctness of buried point?

Buried point type

Buried bill basic can be divided into visual buried point, full buried point, code buried point, three forms

Visual buried point refers to the use of third-party tools directly on the front page box to select the location of buried point – simple – can not collect invisible behavior, no interaction behavior

Full burial point refers to the burial point in all places that can be buried, on demand query – efficient – too much data is prone to collapse, data need secondary processing

Code buried point is divided into front-end buried point and back-end buried point. The front-end buried point collects comprehensive user behavior data — no request to the server is required — the back-end buried point collects interactive data information, real-time collection, precision — multi-party cooperation, low efficiency — invisible, non-click behaviors can be collected

Burial site design

As I understand it, buried design starts with clarifying business needs and goals to be achieved with the product manager. Figure out the user’s behavior path according to the business process. Then from the user’s behavior path to find the core node, the core node buried, although some users will do some operations, but for this demand may not be very important. For example, in order to collect the basic indicators of users, such as user PV, UV and activity, there is no need to bury the event of user ordering and purchasing.

  • Demand disassembly – activity sorting – key nodes for buried point design. In short, business first, according to the business needs of the buried design

How do you verify the correctness of the burial point?

Front-end buried points can cause data loss due to delays and other reasons – data collected at the front end is within the acceptable range of 5% and should be avoided as much as possible prior to buried points

  1. Buried design documents should be standardized, naming and definition should be clear, do not cause ambiguity
  2. According to the module division of buried point, buried point document in the design should pay attention to the document written by the module, so that the back and forth end developers buried point by the module when buried point, to avoid omission

So how to judge the correctness of buried point after buried point?

  1. On the Web side, the developer tool of the browser can check whether the corresponding network request is successful and whether the data field returned is complete and correct
  2. Client authentication – Monitors the reporting of buried points of clients
  3. Server verification – The buried logs of all clients are collected, and the server performs follow-up buried point verification. – The buried point logs to be verified need to be obtained. – The logs are reported with characteristic marks