How to use data to drive product operations at different stages of product development?

Operational positions in Internet companies are generally divided into event operations, user operations , and product operations . Different companies will further subdivide according to specific positions, and can also be divided into new media operations , event operations, data operations, merchant operations, category operations, game operations , network promotion, online store operations, overseas operations, content editing, copywriting planning, etc.

Different people will give different explanations about operations. In my opinion, the purpose of operations is to better realize the unity of business value and user needs, so that the product can live better and longer. Correspondingly, the three indicators of most operational work are: attracting new customers, retention , and promoting activation (conversion). That is, with the lowest budget, find the best path, and take the most effective approach to more efficiently bring new users or traffic to the product, so that users stay and actually use it, extend the user's stay time, or promote the conversion of active users into paying users.

Different companies have different business demands, and their operational focuses will also be different. The focus of information product operations is mainly to increase user retention time so that information flow ads can get more exposure and thus obtain more revenue. Internet consumer finance focuses on increasing the number of consumer loan personnel and consumption quotas and reducing bad debt rates. The assessment standards for operational work are very different from those for information products. Similarly, the operational focus of social products will be more finely divided according to the positioning of different social products. Regardless of the mode of operation, the purpose is to maximize business value without harming users.

Product operation is a major branch of operation and accounts for a large proportion of operation work. The work of product operation mainly involves analyzing product and market data, and using this as a basis to promote product improvement, making a product better and more dynamic. Product operators need to always maintain a keen sense of users. So how should product operations be carried out? How to use data to drive product operations?

Based on the different stages of the product, I would like to explain to you some of my ideas and commonly used tools for product data operations.

1. Product primary stage: focus on core appeal points

The product form in the initial stage is relatively simple, often only realizing one function which may not be the core function. For example, Chunyu Doctor wants to provide online Q&A services for doctors, but they did not launch this function in the previous versions. Instead, they used the online registration function as the entry point. This has a lot to do with the characteristics of their product. The prerequisite for launching the question-and-answer function is that the platform must have enough doctors who can answer users' questions. Chunyu Doctor did not have this condition in the early stage of the product launch, so it turned to the online registration function.

In the initial stage of product operation, the product needs to be verified in the market. At this time, the focus of the product is relatively simple, mainly in terms of user experience. There is less data available for analysis at this stage, but it is not without rules. Before conducting data operations, we first concretize the term user experience in order to identify the data available for operations.

User experience can be simply viewed as the feeling that users have when using a product. Taking tool products as an example, users use tool products mainly to meet their own task needs. User experience is mainly related to task completion, which can be roughly divided into: product response time, user stay time, steps required to complete the task, and final task completion. Users use tool products mainly to save their time and complete tasks more efficiently. At this time, the product needs to respond quickly. If the product responds too slowly, it will cause user loss. If users stay too long, it means that users are not better satisfied in the product. If the steps to complete the task are too cumbersome, it will also cause user loss. Whether the task is finally completed is the determining factor of the success or failure of the product. If the user's needs are not ultimately met, no matter how smooth the process is or how beautiful the interface is, users will abandon the product.

After this analysis, we can determine the direction of data analysis : user stay time, steps required for the task, and the conversion rate of each step. These data are recorded in the server log. We can record these data in Excel and build a funnel model based on them, as shown in the figure:

By observing the conversion funnel and data analysis table, we can derive the corresponding operational strategy. The usual approach is to reduce the process required for the task on the product side, provide users with product usage prompts, establish an error correction mechanism, allow users to re-enter the input after an error in the current interface without having to return to the initial page, provide direct options for users' common problems, and so on.

Simply put, the role of data operations in the early stages of product operations is to help operations personnel conduct product market verification, provide direction for product operations, and promote product optimization and improvement.

2. Mid-term product stage: focus on the optimal path

In the intermediate stage of product operation, the product's functions become increasingly complete, the product enters a relatively stable iteration cycle, and the user group is also relatively stable. If the initial stage of product operation is extensive operation, then this time it involves refined operation of the product. In the mid-term stage of a product, the functions of the product are basically complete. At this time, the main goal is to expand the number of users and increase the number and frequency of use. The product demand generated at this time is mainly external demand and new demands that need to be accessed to meet the needs of different types of users.

Each stage of the product life cycle has corresponding tasks. In the mid-term stage, we need to focus on the optimal path, that is, the optimal path to increase user registration and the optimal path for agile development. In the era of mobile Internet, traffic costs are getting higher and higher, and it is becoming increasingly difficult to acquire new users. There are two things that product operators should do most. Externally, they should observe the input-output ratio of each delivery path, and find the most cost-effective conversion path for delivery. Internally, they should understand each product form, clarify whether the developed products can be reused, and whether the product logic can be applied to different aspects of the same product, so that when new demands are received, they can be met quickly and at low cost.

In the mid-stage of the product, it is necessary to continuously make product corrections based on user feedback. At this time, users have reported a large number of demands. In the face of a large number of demands, it is necessary to use statistical data for comparative screening and analysis in order to divide user needs. We only need to focus on users' core needs for us and find the best path to meet these needs. In terms of data analysis, we can use server logs or JavaScript logs. The timeliness of server logs is poor. If you want to understand the real-time behavior of users, you can use JavaScript logs. Several aspects of data analysis are related to the demands of the product at this stage.

Driving products with data is not just about analyzing data for the sake of analyzing data. What is important is to find data related to the product, clarify the direction of product optimization and use data to evaluate the results of product operations. The role of data should not be exaggerated. Do not use a single data to evaluate the overall situation, do not exaggerate the role of accidental events, avoid using conclusions to deduce causes, and ultimately avoid data-only theory.

3. Mid- and late-stage product development: Finding the most effective way

The user growth rate at this stage is relatively slow, and the product already has a relatively stable and large user base. Most companies will conduct commercial monetization at this stage. At this time, product operators must understand the product goals, that is, the company's business value, and choose the most effective way to rely on the product to realize commercial value without harming the user experience.

This stage mainly involves refined product operations. At this stage, we need to provide personalized operations for users to give them a five-star experience in the product. Taking information products as an example, at this stage, on the product side, we can provide data support for operations through statistics and analysis of user behaviors within the product. Through these data, we can push our content to users according to their preferences so that everyone feels that the content is actually pushed based on their own preferences, rather than conventional content recommendations. At the same time, more in-depth operations can also be carried out based on the user's behavior in the long-tail channels within the app.

The mid- and late-stage product operations mainly involve refined operations. Server logs and JavaScript logs can be used for data analysis. These two data recording tools have their own advantages and disadvantages. The best way is to combine the two to provide more support for refined product operations.

Finally, using data to drive product operations is only a more reasonable way of product operations. Data is objective, but the people who interpret the data are subjective. Only by correctly understanding the data can we correctly use the data. When doing data analysis, we must look at the data beyond the data. We must have a verifying mentality when dealing with data, and always bear in mind the purpose and significance of data analysis, to ensure that the data used can provide the correct direction guidance for the product to the greatest extent.

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

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