Get started quickly in 5 minutes. How can operations achieve useful and real target data?

Today I will talk about three particularly "uninteresting" topics:

What is user operation ?

Why do we need to do user operations ?

How to do user operations.

1. What is User Operation?

Like this picture, the project you operate is a castle.

Why do users come in? This was acquired by your company's business and marketing departments at a high price, so the baton was passed to user operations.

When a user comes in, the operation should try to prevent him from leaving, drive him into the castle and block the exit. This is called retention .

No matter whether it is an event or anything else, the operation must make these users have fun in the castle and make them active.

We need to activate dormant users, turn newly registered users into active users, and bring back lost users. The goal of all user operations is to make users active.

Only when users have fun can we consider the final monetization.

So what exactly does operations do? Actually, it’s just about doing things in the painting. When your PR and marketing attract users, you need to make sure they have fun and don’t run away. Users will give you money if you have fun playing, that's all.

2. Why do we need to do user operations?

1. Data

Why keep users active? First of all, we need operational data and user data, which are the top level, and we need to proceed step by step. Our operations must first collect operational data, acquire new customers, and retain them.

2. Target data

When you are doing operations, each project is different, and you need to figure out what the target data for yourself or your department is.

Each company is different at different stages. For example, when an App is just starting out, it is concerned about attracting new users; after an App has been around for a period of time, it cares about user retention.

If you know what your current target data is, you won’t do things blindly.

3. Real target data

A common mistake our operations colleagues make is that they often fantasize about the future too beautifully and ignore the cruelty of reality. I think this thing is right and I hope all my users are like this, but often the opposite happens.

At this time, you will find that the actual target data is a world away from the target data you imagined. You are fantasizing about your customers and you are also fantasizing about yourself.

4. Useful and real target data

I often go to some friends who are engaged in app development and ask them what data they are paying attention to now.

I wanted to ask you, how much retention rate do you generally pay attention to? They talk about one-time retention, seven-day retention, monthly retention, and three-month retention, and they all pay attention to them. I say that if you can pay attention to all of these and do them well, you will be the next Tencent.

It's a pity that he spread his energy so much that he will find that in the end he didn't do anything well.

He said that all the data are related, it is a funnel effect, there is a bottom only after the top.

I'm absolutely right, but can you do all the nodes well? You have to figure out what is useful and real.

What is the real data?

If it is a small and beautiful project, you need to consider how to make customers pay and how to keep the team alive when operating users;

If your App requires a high investment to operate at the beginning, and the BOSS hopes to gain more venture capital attention and platform-based business through the number of users, then your operation must pursue new users.

Therefore, we need to consider what kind of target data the team and company want, and what use such real target data will have for the project.

At that time, Xiao Enai was very concerned about retention, next-day retention and monthly retention. We are not very concerned about six-month retention, why? One possibility is that they have broken up, and the other is that they have passed the stage of being sick of each other every day and just need to express their feelings normally (using WeChat ) without having to use the private function to call "dear" when you wake up in the morning. So we don’t need to pay too much attention to the six-month retention rate, it is constantly declining and it has nothing to do with the product.

We pay special attention to the next-day retention. Once the next-day retention drops, the first thing we question is: users come and leave the next day, why? There must be something wrong with the product experience. Our app requires two people to register at the same time to use it. Since there is such a binding relationship and they leave after only one day, when this happens we will direct the conflict directly to the product.

We break down all the data and focus on one data at a time node, ignoring other data. After we get one data right, we think about how another data is related to this data.

Let me share a real case with you. We used to have a forum on our app, but because our user base may have reached a clear bottleneck, we cannot compete with social apps.

So what should our forum do? Where is our real and useful target data? We thought, since we can’t win by data, we can only win by quality.

Let me tell you some of the weird rules we follow when running our forum.

First, what topics do couples talk about when they show their affection? Topics like who gave me a gift, what gift they gave me, and sex are unavoidable topics between couples. If it appears, all little affections will be banned.

We work 24 hours a day, and our colleagues at night delete the posts manually. We have three levels of review, so we won't see any posts about sexual topics.

The second post you can’t see is the breakup post. What with the partner running away and no longer believing in love? These are normal expressions of human emotion. Why can't they be put on our forum?

The reason is simple: we understand what useful data we need. In the future, we cannot win by quantity. We need to create a positive environment and introduce positive big-brand collaborations to realize profits. Big brands value the advertising environment rather than just the number of advertisements.

We have 350,000 people on the forum and 27,000 posts every day. We have advertisers such as McDonald's , Pizza Hut, and Chow Tai Fook. The reason they pay is because they think the data is useful.

If we don’t interview the previous operating methods, our data volume will definitely become 700,000 or even higher, but who will come?

3. How to do user operation

Don’t think that operation is a trivial matter. Don’t think that everyone talks about KOL or big IP. In fact, you are an IP yourself. Our WeChat and Moments are things that can be operated. Everyone is doing operations every day: some people operate very well, and some people don’t.

Don’t think of operations as something particularly complicated. What everyone can do every day is the circle of friends.

If you are an operations manager and you have 500 friends, and you write an original article for your circle of friends, which is very well written, but when you post it, you find that only about 5% of people like it. You feel particularly lost.

This is the same as making an app. You do really good activities and promotions , but in the end not many people are attracted, so it feels boring. How can this be done?

Let’s look at the reasons first. The number of people in your circle of friends who you communicate with frequently generally does not exceed 10%. The vast majority of them can be called "knowing" but cannot even be called acquaintances. Now I’m used to adding WeChat wherever I go, just like exchanging business cards in the past. The same goes for the apps in your phone. How many of them have not been opened in more than a week?

So if you want everyone to like and forward your article, what can you do?

First, set clear targets. Get 500 people to like it? impossible. If your boss asks you to do this, let him do it himself. We hope to have more than 50 likes and more than 20 reposts.

Next, segment your users. 10% are the ones you communicate with frequently. These are the people you interact with every day, so you don’t need to worry too much about them. The key is what to do with the remaining 450 people.

After you split up the 450 people, you find that these people care about different issues and have different hobbies and inclinations.

After you categorize the 450 people, you find that most people like a certain topic. So you think, if you do operations for everyone at the same time, you won’t be able to do any one well; it’s better to focus on the group of people who like a certain topic. Be the first to like, comment and post pictures. Summarize and understand the active time of this group of people.

If you take a series of actions and some of them never respond, it means that these are silent users. The same goes for apps. After using a series of targeted operational methods, you find that some people are still indifferent. These users are about to become lost users. Losing users is not a big deal as long as it is controlled within a certain proportion.

Many partners said they wanted to activate new users, and in the end a lot of people participated. Have you broken it down and found how many new and old users there are? How many active and inactive users are there? Only by breaking it down can you know whether your operating indicators have been achieved.

You can implement this once in your circle of friends. This is a very good way to practice.

Finally, a few suggestions for operators:

1. Respect the user's "real" feelings, think from their perspective, and don't take your own experience as the user experience ;

2. Get closer to the users. I work at Xiao En'ai. All employees, from the vice president to the beautiful front desk lady, work as customer service staff for the first 15 days after joining the company. They are responsible for two tasks: chatting with VIP customers in the group and resolving complaints from VIP users. You can only enter this company if you pass the customer service assessment.

Only by getting closer to users can you know what they are thinking.

For example, everyone knows that the WeChat chat experience is currently one of the best in the market. One thing that makes Xiao Enai better than WeChat in terms of chat function is the speed of picture transmission.

We tested 20 social software on the market that can send pictures, and only Xiao Enai could surpass WeChat.

Why can we do it? We talk to customers every day and find that a large number of county-level users and users in the western region are very fond of posting photos, but these areas have not had 4G networks in the past few years.

In order to meet the needs of this group of people, we spent five months building a special team to increase the speed of image transmission.

Our transmission speed under 4G network is not as fast as WeChat, but under 3G and 2G networks, our transmission speed is the fastest among more than 20 social software.

These information products are unknown and can only come from the operations side.

3. You are the user, but the user is not you.

We once launched a fortune-telling product that allowed couples to calculate whether they were compatible with each other. We found two companies, one that does horoscopes and tarot cards, and one that does feng shui and eight characters. Only one person could go up at that time, and no one could convince anyone else.

In fact, we all think in our hearts, who among those born in the 90s and 95s believes in fortune telling? But I thought about it and realized that we cannot make choices on behalf of users. I said we should do a test, using both together and see what the final data looks like.

The final result surprised us. The number of users who used Bazi calculations and paid was nearly 13 times the number of users who used Tarot.

Later we found about 1,500 users to do a survey: The charges for horoscope tarot reading are cheaper than horoscope reading, so why don’t you read your horoscope?

Users say that everyone is doing horoscope tarot and they have read too much of it. The horoscopes online are very expensive, so I'll give it a try.

That's all.

This time it was a slap in the face for us. We have been doing this for so many years, but we have forgotten that you are the user, but the user is not you, and you cannot represent the user.

4. Know how to interpret real data and segment data based on user dimensions.

Understand what your real data is. 20,000 people participated, it was very cool, so many prizes were given out, so many people shared it on WeChat Moments... This kind of rough operation will be replaced by detailed operation.

Regarding user operations, I have summarized three keywords :

The first is feelings. You need to understand the user’s feelings.

Second, the data must be real data. You know what data is real and reliable.

Third, I hope that you can understand the strategy and why you are doing operations. Don’t just do data for the sake of doing data.