7 essential skills for operators: Data analysis ability is the watershed of future operations

The era of traffic-centric, barbaric operations has ended. The next era will be an era of refined operations based on scientific data and centered around users.

Data-driven operations are the trend of future operations and a watershed for us operators. As the slash-and-burn era of operations has declined, refined operations have become particularly important. Data-driven decision-making is a challenge that we operators must face and a skill that we must subconsciously learn.

But it is also a headache for many newcomers who have just entered the operation field, because it involves data analysis methods, methodology, logical analysis capabilities and the use of some tools , and a pile of data is also something that many operations personnel are reluctant to face. In this chapter, we will start from how to obtain data, how to analyze data, and which data dimensions a product focuses on .

1. Where to get the data?

Before we analyze data, we must have data for analysis, so we have to get the data. How do we get it?

There are two main sources of data:

• Proprietary data analysis system - the company's own data is the most qualitative data, and is also the most reliable and comprehensive. Generally speaking, internal data is used as the basis when conditions permit;
• Third-party data analysis tools , which use external tools to obtain data.

Here are five main data analysis tools:

1. Umeng

Support iOS and Android application data statistical analysis

2.growingio

The power of growingio lies in the fact that it can obtain and analyze comprehensive, real-time user behavior data without embedding points, so as to optimize product experience and achieve lean operations.

3. Application Radar

For iOS only, view the App Store overall and category rankings. Check the product’s search score in the App Store, which is one of the criteria for judging ASO effectiveness.

4. Baidu Mobile Statistics

Support iOS and Android platforms. In addition, after embedding the statistics SDK, developers can conduct more comprehensive monitoring of their own products, including user behavior, user attributes, geographical distribution, terminal analysis, etc.

5. Cool Transmission

Only supports Android platform application monitoring. Developers can view data such as app downloads, rankings, ratings and comments, keyword rankings, etc. in mainstream markets, and can also systematically compare data with similar competing products.

Of course, there are more than five data analysis tools. If you are using others, that’s fine too. Using the analysis tool we can get the following:

Record click information , including information that does not interact with the website; can directly generate link percentages, click distribution maps and heat maps; can count user hovers and visualize user potential behaviors

There are actually many ways to obtain data. The key is that as an operator, you must understand what kind of data is important and the relationship between these data. This is an interconnected process, not a single behavior.

Now that we have these data, how do we analyze them? Which ones can be used by us and which ones can be eliminated.

2. How to analyze existing data

After obtaining this data from a third-party data analysis tool or your own analysis backend, how should you analyze it? I believe that many operators don’t have much idea when they get the data. Either you try to do everything at once, or you don't know where to start. These are all manifestations of a lack of analytical thinking, which requires guidance from both macro-methodology and micro-methodology.

In the previous articles, in Yilin Xiaoyu’s article "The Methodologies of Data Analysis in the "Operational Methods in the Post-Product Era", the methodologies that are often used when we conduct data analysis are listed. These methodologies play a macro-guidance role when we conduct data analysis. Therefore, when we conduct data analysis, we should first find a methodology that suits us for guidance. The main methodologies used are:

1. PEST analysis method: used to analyze the macro environment, including political, economic, social and technological aspects.
2. 5W2H analysis method: Why, What, Who, When, Where, How, How much.
3. Logical tree analysis method: list all sub-problems of the problem in layers.
4. 4P Marketing Theory: Analyze the company's overall operations, including four major factors: product, price, channel, and promotion.
5. User behavior theory: mainly used for website traffic analysis, such as returning visitors, new visitors, churn rate, etc., select some applicable indicators from many.
6. A AR RR (Pirate's Rule of Growth Hacker): An important framework for lean entrepreneurship , growth from the five links of acquisition, activation, retention , revenue and referral.

There are many methodologies for data analysis, which cannot be listed here one by one; there is no best methodology, only the most suitable one. Below I will introduce the AARRR methodology in detail. This methodology is very suitable for issues such as lean operations and business growth.

For Internet products, users have obvious life cycle characteristics. I will explain this using an APP as an example.

First, acquire new users through various online and offline channels and have them download and install the APP. After installing the APP, users are activated through operational means; for example, first order free, vouchers, red envelopes, etc. Through a series of operations, some users are retained and generate revenue for the company. During this process, if users think the product is good, they may recommend it to people around them; or encourage sharing to their friends circle, etc. through incentives such as red envelopes. It should be noted that these five steps are not necessarily in the order above; operations can be flexibly applied according to business needs. The five links of AARRR can all be measured and analyzed through data indicators to achieve the goal of lean operations; the improvement of each link can effectively grow the business.

When using these data analysis methodologies, it is important to be clear about their role:

• Sort out the analysis ideas and ensure that the data analysis structure is systematic.
• Break the problem down into related parts and show the relationships between them.
• Provide guidance for subsequent data analysis.
• Ensure the validity and correctness of the analysis results.

For example, when we analyze the data dimensions of an APP, we will use trend analysis, because trend analysis is the simplest, most basic, and most common data monitoring and data analysis method. Usually we create a line graph or bar graph of data indicators in data analysis products, and then continue to observe, focusing on outliers. In this process, we must select the first key indicator and not be misled by vanity indicators.

If we take the download volume of the APP we analyze as the first key indicator, we may go astray; because a user downloads the APP does not mean that he uses your product. In this case, it is recommended to use daily active users as the first key indicator, and only users who have initiated and executed a certain operation can be counted; such indicators are of practical significance, and operations personnel should focus on such indicators.

3. What data dimensions does a product focus on?

We all know that operators deal with various types of data every day. So what data dimensions of a product do we often analyze?

The data indicator system of a product (especially APP) can generally be divided into: user scale and quality, channel analysis, participation analysis, function analysis and user attribute analysis.

1. Analysis of user scale and quality: including total number of users, number of new users, retained users, and conversion rate . User scale and quality are the most important dimensions of APP analysis, and their indicators are also the most compared to other dimensions. Product managers should focus on the indicators of this dimension.

2. Channel analysis mainly analyzes the changes and trends in the relevant channel quality of each channel in order to scientifically evaluate channel quality and optimize channel promotion strategies . Channel analysis should be given special attention, because cheating in the mobile application market is an open secret in the industry. Channel analysis can compare the effects of different channels from multiple dimensions of data. For example, users from different sources can be compared from the perspectives of new users, active users, next-day retention rate , single-time usage duration, etc. This way, you can find the channel that best suits you based on the data and achieve the best promotion effect.

3. Engagement analysis mainly analyzes the user's activity . The analysis dimensions mainly include launch times analysis, usage time analysis, page visits analysis and usage time interval analysis.

4. Functional analysis mainly includes:

• Functional activity indicators: active users of a certain function and usage; function verification; data analysis of product functions to ensure the rationality of function selection.
• Page access path: the page access and jump status of each step in the entire process from user opening to leaving the application. The page access path is a full statistics. Through path analysis, we can find out the diversity of user types and the diversity of users' purposes for using products, and restore user purposes; through path analysis, we can segment users; and through user segmentation, we can return to product iteration.
• Funnel model: It is used to analyze the conversion rate of key paths in a product to determine whether the design of the product process is reasonable and analyze user experience issues. The analysis of user conversion rate focuses on analyzing the reasons for loss at each level of the funnel. Focus on the conversion rate of each step in the app and its impact on revenue levels by setting up custom events and funnels. By analyzing event and funnel data, we can optimize the steps with low conversion rates in a targeted manner and effectively improve the overall conversion level.

5. User attribute analysis Whether in the early stages of our product launch or in the adjustment of our strategy, analyzing user portraits is of great significance. For example, before designing a product, we need to build a user portrait to guide design, development, and operation; the product iteration process requires collecting user data to facilitate user behavior analysis and link it to the business model , etc.

User attributes generally include gender, age, occupation, location, mobile phone model, and network usage. If you are interested in other attributes of the user, you can go to the background of your own WeChat public account or other backgrounds such as Toutiao, UC, etc. to see what dimensions the user attributes include.

Mobile application product promotion service: APP promotion service Qinggua Media advertising

The author of this article @艺林小宇 is compiled and published by (Qinggua Media). Please indicate the author information and source when reprinting!