Building a data operation system from scratch

building data operation system is the end, and is the starting point, this process is iterative, is the core of the system.

 

Before

knew the data operation, did the operators have the following problems:

different channels, the effect is good or bad,

?What’s the cause of the decrease in

activity,

?

how about this campaign?

released a version that users like or not,

we always say how big it is to spread and spread,

These are

products and operations every day, every hour, and every minute will encounter problems. Data operation, practical to solve these problems as the root. It has never been the exclusive BAT, nor is it the only pet of big data. Every Internet Co has the right data to operate on the soil.

data operation system is the collection and application of data analysis. It is also the strategy of data first. It is not only the work of operators, but also the common vision of products, markets and R & D. From the management point of view, is from top to bottom promotion, if the leadership does not pay attention to, then the executor data is used better, but also half a leg walk.

how to build a data operation system? Here is my summary of thinking.

I divide the data management system into four layers, each of which is evolving and interdependent, and each layer is indispensable. The four layers are data collection layer, data product layer, data operation layer and user touch layer. It is a framework from the perspective of operators.

data collection layer

The bottom of the

data operation system is data collection, which is the oil in the system.

 

The core of the

data collection is to collect as much data as possible, and it has two principles:

sooner rather than later: mean products from the creation stage, you need to collect data to be conscious, rather than wait for the company to B, C to gather round round. The data operation carries out the whole product stage, and has different operation methods in different stages.

should be less appropriate: refers to only inappropriate data, and no rotten data. Data such as historical data, changes, records, or details is valuable.

, for example:

has a financial product, its credit reporting system will record user behavior in detail, users upload security information when borrowing, will record the user’s operation steps and time in these pages.

there is an assumption that ordinary people must guarantee the information upload is cautious, if this step is completed very smoothly and fast, is likely to default and debt to the crowd: you are so cool is not operating, >

Leave a Reply

Your email address will not be published. Required fields are marked *