How to make users willing to pay? Xunlei's 7 effective strategies to convince 5 million people to pay
|
There are two things about the Internet that I find difficult: The first thing is to put the user's money into the boss's pocket; the second thing is to put your ideas into the boss's mind. Unfortunately, these two things happen to be what I spend most of my working time on. It's really not easy to support a family. A few years ago, if my boss had told me that Xunlei 's main revenue came from charging individual users, I would have thought he was joking. You know, one of the spirits of the Internet is free. Free, cracked, pirated, users have a lot of alternatives, how can they pay? Take me for example, if I see a software or service that is very good but requires payment, I will immediately lose interest and don’t even think about taking a penny from me. With the joint efforts of friendly companies and peers, paid services are becoming more and more abundant, and user acceptance is gradually increasing. Many services are actually not expensive at all (for example, Thunder membership is 15 yuan/month), and they provide great services. While sometimes you can find free alternatives, I feel like paying is a way to support their hard work. So, how do you convince your users to pay? First, you have to build confidence - it's not that difficult to get users to pay. If you are discouraged, no god can save you. Secondly, the main task of persuading users to pay is to tell them why they should pay, why it is better than free, and to continuously strengthen this reason with products and operations . Slowly, more and more people will pay. Isn't this just beautiful nonsense, a correct but useless theory? Don’t worry. If my eight years of front-line experience is just like this, it would be too superficial. So, what are the specific (preferably systematic) measures? Many people may think that the operational measure to charge users is to hold activities, hold different types of activities, hold activities at different time points, and hold a series of continuous activities. Then make people who see it think "this thing is such a great value, I must buy it", thereby completing the KPI of the number of paying users . That's not the way I wanted to convey it. I have repeatedly stressed that operations require fundamentals, strength, and hard work that takes time and patience. The technical content of the activity is relatively low and everyone can do it. If we start from the fundamentals systematically, the technical content will be relatively high. What are fundamentals and strength? In simple terms, the product you operate is good enough. The work of operations is to make your already good product even better, make more people know about it, and generate greater value. What? If my product is good, why do I need you? This is a misunderstanding of many people, who believe that operations have the ability to turn decay into magic, and that good operations can make ordinary products exaggerate and sell well. What operations need to do is to stimulate users to use products frequently and on a large scale. It is difficult to sustain long-term success by just relying on temporary deception. There are now countless Internet entrepreneurial teams. Being able to discover capable teams and join them is itself a reflection of operational capabilities. What if the product isn’t good enough? Before your product fails to impress you, please improve it more and promote it carefully. If you spend money in advance on promotion and use facial recognition to increase traffic , users will just come and leave, wasting manpower, material and financial resources. I divide the transition from free to paid users into the following seven stages: (1) Not aware of your product at all (2) Known but never tried (for free) (3) Tried but didn’t buy (4) Just bought it not long ago (5) Bought it for a while, or bought it repeatedly (6) About to expire or expired not long ago (7) Churned These 7 stages form a complete life cycle. We have different operation strategies for different stages. The operation strategies for the 7 stages are summarized as our operation battle map, as shown in the table below: Table: User life cycle operation strategy table You see, after we developed such a battle plan, the operation work became systematic. During the entire life cycle, the ideas on what to do and what not to do at each stage are very clear and easy to see at a glance. Below, I will take the Thunder membership operation system as an example to systematically explain how we use this method to make Thunder's 5 million members pay. Stage 1: No idea about your product At this stage, the main focus of operations is to reduce the cost of acquiring user traffic . If others spend 10 yuan to get a user, and you can find a way to do it with 1 yuan, that is a huge advantage. This huge advantage does not allow you to save money, but allows you to invest boldly and reap the rewards quickly. Some reference experiences: l Don't fight for the most popular position, find the traffic entrance that others ignore; l Exchange traffic instead of buying traffic; l Imitate your peers who are better than you in conversion rate , imitate first and then innovate; l Concentrate resources on one channel to go deep and thoroughly, skimming the surface is a waste of money and has no effect; Fortunately, in the operation of Thunder paid membership, Thunder client has accumulated a large amount of free traffic, and what I do is to make good use of this traffic. This is much easier than many other products that do not require quantity. In this regard, there is a very popular term recently called growth hacker , which you can learn about. Back to my own case, 99.999% of Thunder users actually don’t know about Thunder membership. Although we don’t have money to advertise on CCTV, we have a huge number of free users of Xunlei, with hundreds of millions of downloads every day. Then we will expose it below each download. How to let free Xunlei users know about paid membership functions Through such an exposure portal, we have made tens of thousands of users aware of the “acceleration” function by just clicking. Stage 2: Knowing but not trying it for free Today’s users have become very picky. Sometimes they may not even bother to pick up the free money they get when they sign up. So what can we do? (1) Find the right scenario. When you are hungry, giving you a free sample of a newly released cake will have a higher conversion rate. However, if you have just finished a meal and are a little thirsty, asking you to try cookies is not a good scenario. As for the grasp of the appropriate scenarios, it is called "operational feeling" in a more mysterious way, or in plain words, it means whether you often use your own products and whether you are a hardcore user of this product. Let’s take the Thunder download that I am familiar with as an example. When the user’s download speed is slow, it is a good opportunity to open a membership. No matter how good pop-up ads or banner ads sound, people probably won’t read them carefully. At a time when download speeds are slow, we need to provide users with some differentiated experiences: trial. High Speed Lane Trial Button Changes (2) Lower the threshold for trial. Sometimes when I try other companies’ products, they set up a lot of obstacles. They ask you to register, fill in a verification code, and bind your phone number. I very much understand that we hope to obtain some information about users through the trial to facilitate future contact. But don't rush. If your experience threshold is too high, people are so impatient nowadays that they might just turn around and leave. Taking the high-speed channel trial of Thunder members as an example, when users see the button change, they naturally click on the trial. You can experience the member acceleration effect directly during the trial, and the bandwidth can be fully utilized instantly. This experience is very pleasant. No registration, no extra operations, just one click to complete. (3) Leaving money (information) for trial use is not charity . The ultimate goal is to hope that users will pay for the official version after using it. The traditional method of paid software is to try it for 30 days, let you develop dependence, and then pay. This method is very effective. Here are a few more effective ideas: l Provide more trials if costs are controllable to leave a deeper impression on users; l Provide active trial paths for users who have tried the product (e.g., trial by registering an account, trial by binding a mobile phone, trial by helping you attract new users); l Provide payment guidance for users who have good trial results; and provide more trials for users who have poor results; Taking Thunder membership as an example, we have trial durations ranging from 60 seconds to 120 seconds, allowing users to see the acceleration effect and the sudden drop in speed after the trial. There is a limit on the number of trials per user per day (I can't remember if it's 5 or 10 per day). If you register an account and bind your phone, you can get a full 3-day trial. If the trial speed is increased by more than 50% (it has been changed several times in the middle, meaning that the results are good), you will be guided to pay, and if the results are not good, nothing will be said. We have millions of users trying out our membership services every day. The number of members activated at this time every day accounts for more than 50% of the new members added that day. This can be said to be a very efficient entry point. It looks like a button and a function, but the operation behind it is a lot of tricks: Do you think just adding a trial function is enough? If you want to find files that are large and download slowly, a trial of 60 seconds or 120 seconds will significantly speed up the process. You need to find ways to improve the search speed so that the trial button appears within 1 or 2 seconds, otherwise users will minimize the window. You need to find ways to pop up a window in the lower right corner when users minimize the window, but reduce harassment. You need to find ways to expand the number of files that can be tried. At first, our daily trial volume was only a few hundred thousand, but later it expanded to millions. How should the content of the prompts given to users change before, during, and after the trial so that users can feel the real acceleration effect? The number of trials a user has per day should not be too many or too few, and we should keep experimenting with this degree; we should encourage developers to make the trial computer room independent, so as to reduce the impact of the increase in the number of trials on paying users; Stage 3: Tried but didn’t buy It’s finally time for the final push. At this stage, the most important task of operations is to “provide a reason for users to pay.” Common reasons include: (1) Paying for a product gives me more features, and I think the extra features are great, so I buy it; (2) Paying for a product allows me to break through limitations, and I happen to be stuck at the free limit, so I buy it because I need it; (3) The trial worked very well, and I think it’s useful, so I buy it; (4) The price is cheap, there are discounts, and there’s a promotion, so I buy it on impulse; (5) It’s sold together with other things I want to buy, so I feel like I’m getting a good deal, so I buy it; (6) I actually want to pay but don’t want to spend money, and then they ask me to do something, so I do it; Reasons 1 to 3 are that users have needs and you happen to be able to meet them, which is the most natural way to activate it. Reasons 4 and 5 are the "superficial work" of operations that we may see the most, doing various activities. If you are interested, you can collect QQ membership, Thunder membership, Youku membership, iQiyi membership, YY membership, and various online game activity pages, which can help you learn a lot of activity forms through imitation and plagiarism. Let me expand a little on reason 6. Taking Thunder membership as an example, users usually do not purchase after several trials, and are still high-active users who frequently download BT and eMule. We will push "free" or "bargain" messages in the form of advertisements. The free services here are usually paid for by other manufacturers. For example, during the financial management war that was particularly popular some time ago, many financial management websites proposed the condition that "Xunlei will give xx yuan for bringing in a registered user and xxx yuan for bringing in a purchasing user." For us, we would rather return this money to users in the form of Thunder membership. Although users are cheap and unwilling to pay, they can get membership "for free" in this way and then enter into our membership operation system. Once he gets used to high-speed downloading, it will be painful for him to go back to the previous slow downloading speed, so there is a possibility of renewal later. With this trick, we can attract more than 100,000 "stingy members" every month. In stages 1 to 3, more work is done on traffic and promotion operations. After the user pays, the work shifts to product operations . Some operations colleagues think that stages 1 to 3 are their main work, and just leave them alone after collecting the users’ money. Because they think that whether or not they can renew their subscription depends mainly on the quality of the product. From a certain perspective, user renewal does play a major role in determining the quality of a product. If I spend money and find out your product doesn't work, why should I renew my subscription? When analyzing user churn and renewal, we found two characteristics: (1) Users paid for a certain function, but renewed their subscription because they had many more functions; (2) For users who paid for a one-month subscription for the first time, the likelihood of renewal the next month increased by nearly 1% for every additional day of usage; In the next few stages, the focus of operations will shift to: 1) enhancing the value of paying users; 2) stimulating users to use your product frequently and on a large scale. These all require close cooperation with the product to be achieved, which is one of the reasons why we often say that product operation is inseparable. When communicating with my colleagues, I often say that operational sense is very important. Some people can quickly extract the user's psychological behavior and think of functional points that will impress users, while some students just imitate the functions of competing products or send out a questionnaire to let users vote on which functions they want most. Feeling is a very mysterious thing. I later observed several people who were particularly good at operations, and combined with my own summary, I found that they had the following common points: l Be curious. I often use new stuff, and it’s basically stuff I haven’t heard of before; l I like to recommend things to others. The process of recommendation is to summarize and think about how to make people understand it quickly; l Sensitive and high frequency. They are better at finding weird questions than many of their test classmates. And it is I discovered it naturally during extensive use, and I didn’t deliberately use it to be picky; When using it, it feels like I have schizophrenia: one normal self is using the product, while another virtual self is observing the "self using the product", recording what the "self using the product" feels good or bad about, what psychological activities occur, and any misunderstandings or things that almost go wrong. When you take a few moments each day to do this kind of “product meditation,” you’ll continue to feel better. Stage 4: Shortly after purchase Back to the stage of "soon after purchase", the two most common states of users are excitement and confusion. I just paid for a membership, and all the functions that were previously unusable are now open. All the functions that were previously restricted are now available. My first inclination is to use it vigorously, download it as much as possible, and waste it as much as possible. If nothing goes wrong, everyone is happy. However, if there is even a little problem at this stage, such as network delay causing function failure, frequent operations causing submission failure, submitting hundreds of BT at a time exceeding the background settings, or your account being hacked shortly after payment, the user will first be confused, then extremely sensitive, and finally burst into extreme negative emotions such as "I paid and this is it", "You cheated me of my money and run away", etc. Operations must do a good job of managing user expectations. Reduce possible problems in advance and ensure smooth communication after problems occur. Just like people who are in the early stages of a relationship, you need to be gentle and patient with him/her. Here are some examples: (1) Accounts that have been paid for are prone to being stolen, and users complain that “my account was blocked as soon as I paid for it.” In fact, someone just changed the password. Because people don’t pay much attention to protecting free accounts, but once the value of paid accounts increases, they will be targeted by “bad uncles”. Therefore, after the user successfully pays, we will remind everyone to bind their mobile phone, email address, and change their low-strength password to reduce the risk of account theft. What if the user is too lazy to change it and it gets stolen? At first, we provided password retrieval services through a dedicated customer service hotline for members, but we later found that the speed was too slow as the excitement of purchasing a membership had already passed. We will just ask you to provide a screenshot of your payment record and we will help you recover it. This greatly improves the password recall experience for paying users. (2) What should I do if the acceleration doesn’t work after I buy it? This problem was particularly troublesome when I first became a paying member. It is clearly a problem with the user's bandwidth network, but most people don't understand and think that if they pay you, it should be fast. Later we changed our operation strategy: 1) For small operators and campus network users, remind them that the effect is not ideal and they should be cautious when opening the network; 2) Targeted marketing to users who download BT and eMule frequently, because the speed-up effect is most obvious for these users; 3) If the trial speed-up effect does not meet the standards, it will not be guided to open. In this way, user expectations can be controlled before activation. If we don’t want to do a one-time deal and hope to have long-term and continuous payment, it is very important to manage user expectations; (3) The server is down. This is also a common problem. Even if you have 99% reliability, the tolerance for failures between free users and paying users is completely different. What's more, in many cases, reliability is far from that high. At the beginning, we tried to hide our fault, saying it might be caused by the user's own network environment. However, once everyone on the forum gets to know what is going on, hiding things will only lead to more suspicion. Later we adopted a unified solution: if the server was down for a few hours, we would compensate for 1 to 3 days; if the server was down or upgraded for more than 1 day, we would compensate for 7 days. In this way, if the user cannot use the service, the time he or she has paid for will not be wasted. Later, some people with a not-so-sunny attitude emerged, saying that their membership was about to expire and they tried their best to "curse" us into having problems, hoping to be compensated for a few days. (4) Multiple people share one account. After new users have paid for the service, they particularly like to share it with their friends because it makes them feel proud and they can’t use it all themselves anyway. This kind of sharing not only brings about a significant increase in costs for the company (but users don’t care), but also has the problem that you don’t know whether your buddy will share it with his buddy, or his buddy’s buddy. This makes the entire sharing become larger and larger, and when I spend money to buy what I want to use, my account is kicked out because it cannot be online at the same time. Later we came up with two strategies to manage user sharing. One is to allow limited sharing with a few friends, but if the number exceeds a certain number we will prompt your account may have been stolen and you must change your password before continuing (soft restriction on sharing). The second is to open a special sub-account for sharing. The parent account is for your own use, and the sub-account gives you face. We wanted to do more with Alphabet accounts in the future, but unfortunately, due to cost control after the listing, it became just talk. Newcomers often know nothing about the product's functions, rules, and background restrictions. Good operational guidance and expectation management can greatly reduce the tendency to break up during this sensitive period. When users get used to your rules and understand how to use these paid features, what works and what doesn't, the operation work will move to the next stage. Stage 5: Bought for a while, or bought repeatedly After using it for a while, I have become familiar with your products and accustomed to the highlights you provide. If they don’t see any product improvements for a long time, if there is no freshness, and if they don’t feel that “the money is really worth it”, their payment life cycle will usually not be too long. In order to extend the payment period, it is very important to "enhance the value provided to users." This reinforcement is not just about spiritual incentives such as telling users after each use how much effect you have gained, how much better you are than free users, and how much time you have saved. It also requires real material incentives - improvements in product functionality. When users purchase a Thunder membership, they are actually purchasing the service of "let me see good movies faster." They want speed and pleasure. The faster the movie is downloaded, the better. 1M/S and 2M/S are too slow. They wish they could have 100m fiber optic cable in their homes. If you observe yourself or your friends (if possible) carefully, you will find that you often use the "fast forward" function when watching movies. Don't get the wrong idea, I'm talking about the past memories in the cartoon. If you don't like this part, fast forward to the next part. If you don't like this episode, skip to the next episode. If this movie doesn't live up to its name ( clickbait ), switch to another one. A movie may only take 3 to 5 minutes to watch. Let's abstract this problem and look at it: the time users spend relaxing their minds = 15 minutes looking for N movies + 2 hours downloading N movies + 10 minutes skipping between N movies. When we want to make it faster for users, the biggest bottleneck is the time it takes to "download N movies in 2 hours." If we can eliminate this time, it will be more effective than increasing bandwidth, servers, or improving background algorithms! If a method of online playback is provided instead of downloading, then a great deal of time can be saved in this regard. That’s the feeling of operation. But this is absolutely impossible to be solved by product operation. If you have a little understanding of online playback, the format of online playback is very different from the format of downloading and watching. The solution proposed by our smart engineers is to use a high-performance server to quickly transcode the film (transcode once, benefit for life), and then play it as an online video. We call this feature "Cloud Broadcast". With the advent of "cloud broadcast", the agonizing wait of several hours has been reduced to 3-5 seconds of buffering and immediate viewing. In subsequent version improvements, fast forward, progress bar dragging, progress bar screenshot preview, and screenshot preview speeding up from the original real-time screenshot to pre-screenshot have been added, so that you can quickly understand whether the plot of the film is what you want. Later, we found that a large number of users watch more than ten or twenty movies a day, and many of them are stopped within one minute. Through a series of functional improvements, we have further strengthened the value of "paying faster", which has significantly increased the renewal volume. Stage 6: About to expire or will expire soon Everyone has procrastination problems. Why do you ask me to renew my subscription when it is working fine? According to our backend statistics, the vast majority of users renew their subscriptions on the expiration date or within the golden 72 hours after expiration. I have four words to summarize the operational strategy at this stage: coercion and inducement. Intimidation means telling the user what benefits they have received and what they will lose after downgrading. Inducement, which is to give discounts (usually to buy more time) and guide users to use automatic renewal (usually the price of automatic renewal is cheaper than buying separately); Thunder members are not very successful in this regard. The main reason is that people only renew their memberships when they have downloading needs. If I have less time to prepare for an exam this month due to a business trip, I will not continue to purchase. I'll just buy it next time I need it. We borrowed a practice called VIP level system from QQ members. When you renew your subscription continuously, your VIP level will get higher and higher, and users with higher levels will enjoy more privileges (for the same amount of money). When you stop renewing, this level will decrease over time. The next time you renew your subscription, your level will be downgraded. This strategy stimulates a small number of users who care about the vanity of their rank. Later, I discussed with several colleagues who were doing functional payment, and overall this strategy had little impact. It still depends on solid differentiated features to impress users and keep them paying. Stage 7: Churn If users feel that your service is worthless and useless, they will leave. Relatively speaking, it is quite difficult to get lost users back. Everyone knows the value of your product. So what does it matter how good your advertising sounds? At this stage, there are two opportunities that can easily impress them and recall users to renew their subscriptions. One is that the user's own situation has changed, and the need for this paid function has increased; The other is that we proactively enhance new features and new values that may be attractive to users; I continue to use the example of Thunder membership. We have lost a considerable number of users because their bandwidth is not fast enough. Free p2p acceleration can already fully utilize the bandwidth. Even if you activate membership acceleration service, if the effect is not obvious, you will certainly not be willing to renew the subscription. So you have to continuously analyze the user's usage environment. When we find that the user's access bandwidth has increased and ordinary downloads can no longer fully utilize the bandwidth, it is a very good marketing opportunity. Give them a longer trial period on a "targeted basis" to allow them to experience the acceleration in a new environment. The speed has been substantially improved, and coupled with exclusive discounts for old users, the possibility of impressing users will be greatly increased. Sometimes one trial may not be enough to impress users. You can provide multiple full experience memberships in a rhythmic manner over a period of time. This will allow users to feel the long-lasting speed-up and increase the possibility of recall. Don't you have insufficient bandwidth? Then we can help you increase your bandwidth. Will this impress you? In fact, ADSL in many places already supports higher-speed access, but not many people actually use it, either because users don't know about it, or because it's too troublesome to go to the business hall to apply, or because there's no demand for it. We later worked with telecommunications companies to offer enhanced physical bandwidth as a value-added service bundled with Thunder memberships. When users’ download speeds were slow, we provided a trial of dual acceleration of physical bandwidth + server acceleration, which was immediately “amazing”. Many novice users who don’t know much about it asked in the forum what kind of black technology this is that can actually break through their own physical bandwidth limits and truly increase speed. After we launched this "black technology", we conducted several rounds of marketing targeting former members and successfully recalled hundreds of thousands of old users who were difficult to convert. I later chatted with some members, and they said that because of this feature, they renewed their membership for another year. Some people say, isn’t this the job of the product? Only when the product is improved and the functions are increased can we attract more users to pay. That’s right, stages 4 to 6 require product cooperation, and the principle of “product and operation are inseparable” is clearly reflected in these three stages. Without product improvement, no matter how much you hype it up, users have already experienced and understood it, and it is difficult to have a substantial effect; Without the value delivery of operations, there will be a loss of cohesion no matter what to change, what functions to add, which users to target, and when to deliver value. Summarize In fact, if you cover every fundamental solidly and continue to do so every week, month and quarter, you will reap very generous returns. Later, after this trial system was perfected, we found that our classmates had nothing to do and just lay there every day counting how many people came and how much money was collected. This kind of operation requires real effort and is the fundamental basis for the continued growth in user numbers and revenue. Compared with those short-sighted, profit-seeking, impetuous, superficial and exaggerated methods, this kind of operation allows me to sleep peacefully. So the actual operation often looks dull and nothing special. Sticking to it for a month or two may not produce any results, and may even be worse than those revenue-generating tricks that deliver a lot with a little effort. If it can be sustained for three to five years, the effect will be apparent. Mobile application product promotion service: APP promotion service Qinggua Media advertising The author of this article @黎晨 is compiled and published by (Qinggua Media). Please indicate the author information and source when reprinting! Site Map |
<<: Some tips on server hosting?
>>: How to write a good article title? Learn these 3 points!
Recommend
Introduction to Moutai: How to avoid malicious clicks on advertisements when promoting corporate websites
For most companies or products, they would like t...
Renyifei-Low Buying Ideas 1PDF
Ren Yifei - Low-Buying Ideas 1 PDF Resource Introd...
Five elements of user reach: materials, channels, and objects!
My definition of user reach is: sending specific ...
How much is the price to join the Tangshan Tea Mini Program? Tangshan Tea Mini Program Franchise Price Inquiry
In order to better penetrate into various industr...
Ji'an SEO Training: How to increase the click probability of your website? Develop a comprehensive click strategy for new users
How to trigger more users to click on the website...
How to do a good job of Zhihu marketing for a brand?
Zhihu's commercial marketing system is curren...
7 key points you must know when starting to operate Douyin
I have summarized and sorted out some universal a...
Analysis of fresh food e-commerce operation strategy
Due to the epidemic, the fresh food e-commerce ma...
How to achieve effective drainage? Share 2 strategies!
There are only two truly effective traffic-genera...
How to write a short video script? These templates can be applied directly!
Although a short video is only 60 seconds long, e...
5 steps to operate Internet users!
As the link closest to users, user operations are...
Which industries are suitable for Momo information flow? Promote industry analysis!
Mobile Internet is in a state of vigorous develop...
How to operate an event well?
This article is divided into three parts: the val...
How to give full play to the SEO function of the website?
1. SEO parameter settings Website title keywords:...
How do physical stores operate private domain traffic?
Private domain traffic is the life-saving straw f...