Big Data Era: A Summary of Party B Party data differences

Today, the big data era quietly, many people are hung mouth to create value, data mining and some hot words. The internal companies are also gradually realized the importance of data, have set up the data sector, looking forward to the data can really serve the business. In addition, some third-party companies specializing in data services are constantly emerging, hoping to help Party A which generates the data share the burden of data analysis and dig out more valuable regularities to help Party A continue to improve its business level and continuously find out Problems. From this perspective, the original intention of Party A and third-party data service company is the same.

So in the end in the third party to do data services and data services company do have any difference? Combined third party companies in recent years to Party A to do the data experience, the difference between the two data made a simple Summary.

1, the pursuit of different

There are several types of cooperation modes between third-party companies and Party A about data services:

1) long-term monitoring of traffic data - regular daily newspaper, weekly, monthly;

2) Temporary Project - Upon receipt of the demand of Party A, initiate the project and summarize the project research results in the form of a report within the prescribed project cycle.

No matter what kind of cooperation, third-party analysts in the analysis process, always lonely, always more dependent on the data. Because not in the analysis of the environment, do not know what the operation is doing recently, do not know what kind of product plans, in a word: do not know where the other side of the point of concern.

Party A to do the internal data, in fact, the way of cooperation with the above listed two similar. Only with the details of different:

1) Before the research, we will learn more about the background of the project and the confusion of the product or operator.

2) The study encountered data can not explain the problem, you can always find the relevant personnel feedback. Analyze the cause of data anomalies with business colleagues.

3) After the study, the entire research report will be reported and business colleagues discuss the next improvement strategy and plan, agreeing on the next research point in time.

So, subject to conditions, although the original intention is consistent with the company, but because of inaccessible business, so for the ultimate goal can only stay in the completion of a rigorous, professional data analysis report. As for the follow-up, Party A how to use this report, how to improve the business, whether to improve the business and a series of follow-up work is unknown. Therefore, the output of third-party companies can not always form a closed loop in the entire business chain.

However, this is indeed an unavoidable thing. I believe most of the current companies, especially Chinese companies, are more concerned about the privacy of the company's internal data. Their attitude toward third-party companies will not be completely open. Therefore, cooperation between the two parties is limited to relatively shallow cooperation. Third-party companies want to really finish the business closed-loop, from the current stage is completely impossible.

Over time, third-party data analysts get accustomed to making a good report as the ultimate goal.

Attached: the true feelings of the transitional phase

Just entered the Sina microblogging from the third-party company to do the first project, the project reported the same day the product was affirmed. I was very happy at the time and felt that the work was recognized. Product colleagues have agreed that, sure leadership will feel pretty good. However, the result is exactly the opposite. The leader's classmates were very puzzled by the fact that I did not have any follow-up plan for the follow-up of the results. From our serious conversations, I deeply understand that the data in Party A is really for product improvement or operation. If you only find the findings in the summary and finishing problems, it basically only made 50 points. Equivalently, the doctor took the x-ray only for the patient, then ignored the patient.

2 . Different forms of presentation

For the cooperation between the third-party company and the first-party company, the value of the project is reflected in the report, so the production of the report should be beautiful and make people feel "excellent value for money." Only a few pages of PPT must not be used as the final output. If you can in the study, through the establishment of a complex model to help illustrate the research results will be more valuable.

Within the company, everyone urgently wants to know what I can do to see the results of this research. If this model is complex to the product staff can not understand, or overwhelmed, it is pointless. Instead, it is a study of a specific problem, even if only a few pages of paper, a few data, will make the product staff very excited.

For example, occasionally see some of the data from the phenomenon or problem, then do a simple order, print a few pages of paper you can go to the product, the operation of the students to talk about. To see the business front-line colleagues how to treat this phenomenon, is not some important business strategy affect some of the data results, resulting in abnormal data results, and not really a problem. If there is no other abnormal factors still appear this phenomenon, then we really have to project the following to take the time to find the cause of the problem.

3. Achievement reflects different

Third-party data companies, the core business is the data collection, analysis, and therefore responsible for the output data report data analysts, in contrast, the results of the work can easily be seen. Therefore, also received the company's attention.

In Party A, data services are a functional line that serves product and operations personnel. Perhaps the manifestation of the value of the work only after the completion of the business chain can be reflected. Even taking the whole business chain and assessing the role of data in the whole process is a difficult task. However, being a data analyst can see his analysis, helping to identify problems with the product or operation, improving the product experience or operating mechanism, and does this sense of accomplishment inspire analysts or will they be enthusiastic Rushed to the next project.

4. Different grasp of industry standards

Third-party data companies Because of multiple companies within the service industry, the same business modules of the same industry companies are put together to an industry standard. Provide industry-standard data to clients without revealing the trade secrets of their clients. So that everyone can clearly understand their position in the industry and understand what the data looks good on the surface. However, compared with the industry standard, the situation is actually not optimistic. I think this is the biggest value for third-party data companies.

However, the status quo is that in China's Internet industry, everyone is still very tight on the data and is unwilling to allow third parties to intervene too much. The result of this is that we do not have an industry standard. The quality of the data can only rely on experience.

In summary, the differences between the two data analysis work, to my experience is: data analysis is not focused on the data, but rather how to really solve the actual problem. The ultimate value of data analysts is not how many statistical tools will be used, how many data models can be excavated, but really understand the business. See the results of the data to know what business problems, and see the business problems and can clearly know by analyzing what data can be informed of the reasons for the problem. Of course, if, on a large scale, companies are willing to work with third-party companies and establish industry standards through third-party companies, that will eventually make the data more valuable.