How to use data to drive operational growth

Maybe we won’t get a satisfactory result immediately after data-driven operation , but if we don’t even make optimizations and improvements, we won’t even see any good results.

I introduced the development of novice operations work. In summary:

1. First understand the product, the division of labor and status of people
2. Clarify the content of your work and make a good plan
3. Cultivate the use of operation tools and shape operation thinking
4. Practice small skills and prepare for the transition

Once you know the content of the work, you need to carry out the work based on the data.

Operations involve dealing with data on a daily basis. KPIs are set based on data, and operational strategies are adjusted based on data; therefore, understanding data is a basic skill.

Refined operation is the core thinking of operations in the current Internet environment. It is a method based on data to achieve better results with less cost. So how to achieve refined operation at different stages of the product?

1. Define your data baseline

Product data is divided into basic data and operational data;

• Basic data refers to the data generated by the product every day, such as DAU, UV, clicks , transactions, online time, etc.
• Operational data refers to the data generated when operations take certain actions on the product; for example: activity data, activation conversion data, order volume optimization data, etc.

As an operator, you must first understand the basic daily data of the product; only when you know the DAU and transaction volume first, you will have a clear idea when you want to evaluate the effect of an activity.

For example: Generally, the number of participants in an activity must be two to three times the DAU to pass; to pass, does it mean closing the activity on time? Or do secondary diffusion? If it fails, will it be taken off the shelves for adjustment? Or finish the summary? If you don’t have any benchmark data to refer to, such as DAU 300 and number of activity participants 310, and you still think they are good, then there may be something wrong.

I will write a separate article to introduce some data baselines for each industry.

2. Find the core indicators of different stages of the product

2.1 Product Primary Stage

Business model verification: In the start-up phase, an Internet product is more likely to solve specific problems for a specific group of people and conduct market verification of its own business model . In many cases, users do not know their own needs. Internet products do not provide products, but more of a way to solve user needs.

Take Taobao as an example. It not only moves products online to meet users' greater shopping needs, but also provides a complete model from product selection, online payment to logistics. Imagine if it only allows you to buy something that you can't buy locally, but the logistics takes 1 month, would you still choose it? Therefore, in addition to the product itself, what is more important is the way to solve user needs, that is, the business model.

At this stage, the data you are looking at may not be how much revenue there is, but the feedback from vertical groups on core functions. We should pay attention to some core indicators related to growth, such as daily/monthly active users and retention . The purpose of these indicators is to measure the current performance of the product and also to provide benchmark data for future growth.

Product function verification: An APP carries the company's business model, and the functional design of the product itself directly affects the operation and channel operation, so every product operator needs to know the basic data of his own products.

For example: when the product retention is poor, people will definitely not put up advertisements ; if there is no stickiness, then burning money for growth will not really lead to growth. Because the loss rate exceeds the growth rate. We should focus on polishing our products to increase activity and retention.

On the contrary; if the retention rate is higher than the industry level, you should go ahead and do the campaign, use earlier data as a benchmark, and always pay attention to the growth momentum. If the performance of a certain month is lower than the historical benchmark data of the product, it means that there is something wrong with the product's actions.

2.2 Mid-stage product

As the product functions and experience become more and more perfect, the user base is relatively stable and has accumulated a certain number of users. At this stage, if the operation does not understand data-driven and relies on intuition to compete with peers, it may work once or twice, but no one can enter a casino and win dozens of times in a row.

Therefore, operations need to be able to quickly optimize each node and improve the conversion efficiency of key nodes.

Through the continuous improvement and accumulation of conversion efficiency, user loyalty will naturally increase, and eventually become the core competitiveness of the product and form a moat.

For example, increasing activity interactions, sign-in points, medal tasks and other gameplay. The extension of these scenarios is more about promoting user retention or further improving core conversions.

At this stage, what needs to be measured is whether these extensions have a promoting effect. In order to improve the user experience as much as possible, you need to understand user needs and even provide product designs that are tailored to each individual.

2.3 Mid- and late-stage products

The user growth rate at this stage is relatively slow, and the product already has a relatively stable and large user base. If the early and middle stages of the product are about finding vertical groups and segmented user groups in order to meet the needs of different groups of people, then the later stages must introduce user stratification operational thinking and strategies - because at this stage, no matter the user size, the complexity of the business, and the company's core demands, you need to be able to operate holistically and efficiently.

You need to clearly define: which ones contribute the most value to the product, which ones need to be stimulated, and which ones are about to be lost. It is very necessary to provide different reach strategies for different levels of users - because the core indicator in the later stage is business monetization; business monetization requires a certain base of active, paying users .

Generally, Internet products have a number of highly active users with good experiences, who will be converted into paying users. It is like a funnel that needs to be screened continuously, and the efficiency of operations is what matters here.

For example: the conversion funnel of e-commerce users is generally: visit - register - search - browse - add to shopping cart - payment , etc. This is a very long funnel. To do a good job in data-driven operations, we must pay attention to every link of the funnel and continuously track these data. This conversion efficiency can be achieved through marketing methods, product improvement methods, and even customer operations methods. A small improvement in each link together will result in an exponential improvement.

3. Several key points for implementing data

Data operations is the process of operations personnel converting data results into operational strategies. People are still the main productive force, so I think people's data awareness should still be required:

1. When using data to make decisions or formulate strategies , you must know what the data results you have currently analyzed can prove, and also know what the data cannot do; personal cognition and experience are limited, and you cannot be too exaggerated or too radical; this is the kind of thinking that should be avoided, and you must learn to share and discuss with the team.

2. The effective use and analysis of data are closely related to operators and teams . Top-down advocacy and initiation are the best results. If the top management has the strategy and awareness of data-driven operations, the management has guiding experience in data-driven operations, and the executive level can implement data-driven operations, then the entire system will be successfully implemented.

However, although operations personnel consciously want to make use of it, they may be unable to push it forward due to limitations on the team's strength, leadership awareness, etc., so they must learn to compromise and stop complaining. For things that cannot be pushed forward, start by analyzing a small node, and let the team develop a habit of thinking with data and using data results to judge and summarize. Later, they can receive specialized systematic training or recruit data professionals to promote the entire data standardization.

3.Finally, data analysis tools and various models are used . This is also the basic requirement for data analysis skills. You must understand multi-dimensional analysis, cross-analysis, pirate model, user stratification model, RFM model , 90-10-1 model, AB testing, etc. You may not use them all depending on the product status, but you must be clear about which one is used to analyze what. Otherwise, after talking about data operations for so long, you get the data but don’t know how to analyze it, what to analyze, and what to use for analysis; isn’t it just empty talk?

Of course, there are many articles on the Internet about the specific techniques and methodologies of data analysis, which are suitable for beginners to understand. For more professional core content, it is recommended to read professional books. Here I mainly emphasize the cultivation of data thinking awareness.

4. Data analysis starts with a purpose, and the purpose should be lean

Product operation strategies include: attracting new users , retention, activities, push, marketing, maintenance, etc.; it is impossible to analyze all users every time, as this is a waste of resources and costs; because you cannot satisfy all users in one way, nor can you do your best in one way; there are differences between users, and these differences need to be divided out by operations; and compensated by refined operations.

If you have 100,000 users and you want to analyze their consumption, you should first emphasize your purpose again and think clearly in your mind; which level and which month's user consumption do you want to analyze?

To reduce team costs, the precise purpose is to break down the target into finer granularity and analyze the users in March; there are new users of the month, and users registered in January who have settled into old users in March; e-commerce companies sell promotional cosmetics, and based on applicable user portraits , the target population should be selected based on age group, city, job occupation, etc. If the purpose is divided finely enough, the goal will be obvious. Precision is a way of thinking about data analysis and also a means of operation.

Final Thoughts

1. Maybe we won’t get a satisfactory result immediately after the data-driven operation, but if we don’t even make optimizations and improvements, then we won’t even see any good results. Excellent operations will not be complacent with a good data result, but think about whether this is the best I can do now, so is there a better possibility that I haven’t thought of? It is both the end and the starting point. It is also self-iteration and the core of ability.

2. If you want to operate your work efficiently, you cannot do without data. However, many operations are actually trapped by various data models and theories, or they have no data awareness at all. Work should start with optimizing a certain node, and then explore the data system of the entire product; and then have a baseline awareness of the product data of the entire industry. Like many investors , they know how big the market can be just by listening to you talk about the product. There are also many experts who can tell whether a product is good or bad and whether to leave or stay just by looking at the current product data.

This is off topic, but I hope we all have the time to understand it all.

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

Product promotion services: APP promotion services Advertising platform Longyou Century