This article is compiled by the editor of Programming Notes# for you. It mainly introduces the basic concepts and operation methods of data operation. I hope it will have certain reference value for you.
Baili wants to do data analysis.
The following are some things learned by Fanruan data operation officer and data analysis
1: Data analyst direction:
1: Business type: business is the foundation of data analysis, Only by thoroughly understanding the business can we analyze the problem. The business type has very high requirements for the analyst’s business understanding ability.
2: Data research and development type: This development direction requires you to be very skilled in technology and be able to analyze data through technology.
3: Algorithmic type: Algorithmic type requires you to have good mathematical skills, and the research ability of the topic is very strong.
2: What should data analysts do daily:
1: Daily/weekly/monthly;
2: Temporary data;
3: Modification of routine work; (change the current work status to increase production capacity)
A: Why read the daily newspaper:?
Through the daily newspaper, you can understand the business status most clearly. ; There may be a situation where the business data does not move every day.
If the data remains the same, it means that there is a problem with the company’s business. At this time, the data analyst should point out the problem .Reflect your own value
B: Cultivate your own data sensitivity through daily reports !
By checking business data every day, paying attention to business data fluctuations, and cultivating your own business sensitivity.
Providing development for business Suggestion, because when the data fluctuates, it is necessary to find the reason for the fluctuation, why it rises and falls. Over time, you will find that the data will rise and fall after making any adjustments to the product or operation, and then you can provide more reasonable development suggestions for business development.
C: Understand the essence of analysis through daily/weekly/monthly reports.
Data Sensitivity and business development suggestions are continuously accumulated from the analysis and summary of daily/weekly/monthly reports. For most companies, daily/weekly/monthly reports are just routine business analysis, listing numbers, and many times Lost the essential meaning of data analysis. The important point is to understand the reasons for fluctuations. The reasons for fluctuations are pain points.
D: Temporary data processing skills
Get the data business caliber (know where the data comes from), and display the result data through sql.
In principle, do not use a data extraction machine. For business orders Dian Weitao. Tracing back to the source, establishing an analysis framework for this business category, and thoroughly solving this type of problem from point to surface.
E: Skills for routine work
The first tip is about daily daily processing. If it takes a lot of time to analyze every day, it is better to hand it over to a machine. For example, build a daily analysis framework through Python to reduce work pressure.
The second skill is that because analysts are busy every day, there will be a large number of business needs that need data support. At this time, it is necessary to keep a clear mind and be able to Reasonable priority scheduling is required.
The third skill is to communicate with your leaders in a timely manner when encountering problems that you cannot solve. Problems can often be solved easily.
Three: Thematic Analysis
Thematic analysis tests the analyst’s thinking, technology, time management, speech ability, etc.
So it is not easy to do a good thematic analysis, which can be divided into 4 steps:
1: Interpretation of requirements;
2: Establishment of logic tree;
3: SQL data extraction and analysis;
4: Three suggestions for writing reports.
1: Demand Interpretation
We know it will cost 20% of the time is spent on communication needs
/span>
Original needs–>Understand needs–>Essential needs
Original needs: target appeal needs, asking for needs is a bit Go to the surface. What data is the demand composed of. I hope to preview the results obtained.
Understand the demand: what kind of help to the demand after completion, what related production capacity has been improved, and what is the impact on daily activities Help, and enlarge the current operation problem.
Essential needs: how to optimize. Which functions are not satisfied, how to meet
2: Create Logical Tree
Establish a logic tree
The purpose of establishing a logic tree is to make the thinking more concise and clear. at the same timeTraffic and revenue are included before the event. Taking traffic as an example, traffic includes the traffic effect every day before the event, because there is often some warm-up before the event, so there will be traffic fluctuations.
At the same time, in the event, it also involves traffic and income, such as whether DAU/MAU has increased, the situation of attracting new users, the situation of old users waking up, the situation of continuous visits during the event, and the distribution of online frequency ,etc. At the same time, it is also necessary to pay attention to the horizontal comparison of various activities, so that optimization suggestions can be made for the business, rather than just staying on simple numbers.
How about the precipitation effect of attracting new users after the event? What is the performance of inactive old users in the event and how active they are in the future? As well as summary data and review data analysis.
Facing the requirements, we must establish a reasonable logic tree. Only with the logic tree can we know what specific points need to be analyzed.
Content before using the effect—->effect after the demand is solved—>subsequent content precipitation (function review)
微信扫一扫打赏
支付宝扫一扫打赏
