Tips for breaking 1 billion GMV in live streaming!

Many people may think that a set of gameplay must be very complicated. In fact, it is not. A good gameplay respects the rules and keeps things simple.

If the overall logic of a game is very complicated, it is definitely not a good game.

What people often call gameplay failures are mostly due to a lack of understanding of the underlying gameplay and a lack of detail processing.

The 30,000-word series will combine the personal practical experience of 100+ live broadcast rooms, and deeply analyze the natural flow gameplay from the bottom up and details. This article is the upper part, focusing on the underlying logic.

The underlying motivation for live streaming product collection

In people's usual impression, the live broadcast rooms of natural flow will have money for attracting traffic, money for welfare, and money for profit, and none of them can be missing.

There is no problem logically. The traffic-generating funds are separated into recommendations, the welfare funds take over the traffic, and the profit funds are tested and orders are placed.

But in fact, different live broadcast rooms have different model combinations and product quality, and the most important factor associated with them is the anchor.

We often find that some live broadcast rooms, even if it is the first broadcast, do not have any traffic-generating or welfare-oriented products, and directly offer prices at the regular price to hold orders, and they can still complete the cold start of the live broadcast room.

We also found that some live broadcast rooms did use traffic-generating funds for their first broadcast, but without any welfare funds, the anchor was able to complete the cold start.

For a live broadcast room that does not rely on any low-cost traffic, the gameplay is secondary, and the host’s ability to transfer funds is the key.

In front of a super anchor, there is almost no need for any traffic-generating money. Even if the cold start audience is very low, the anchor can almost understand the users point-to-point in terms of value presentation, atmosphere creation or rhythm control, and use this to shape the interaction and transaction data of the live broadcast room and achieve healthy growth of the live broadcast room.

For the anchors who are a little lower in the ranking, it may be difficult to place orders directly for the full-price items, but by combining some traffic-generating items and understanding the user's psychology, and making good connections between the traffic-generating items and the full-price items, as long as the atmosphere is created right, placing orders for the full-price items will also be a natural result.

When it comes to ourselves, we have to find ways to arrange the funds according to traffic, welfare and profit. Why?

Because even if we can see such an excellent live broadcast room, most of them do not have such excellent anchors. Regarding the relationship between the anchor and the product, it can be summed up in one word.

All the product models are designed to make up for the lack of anchor capabilities.

This is the current situation of many live broadcast rooms. If you are a novice team and there are no mature anchors, I do not recommend playing the natural flow.

It’s not to say that it will fail 100%, but with so many ways to play such as short videos, store broadcasts, Qianchuan, content live broadcasts, etc., there is no need to spend time on the way to eat the traffic of the anchor.

You don't have a choice? Yes or no?

The so-called "I have no choice" on many levels basically comes from two points. The first is self-destruction. I chose a track that is extremely difficult to match in terms of gameplay. I have the ability to build a bicycle, but I chose to build an airplane. The second is the information gap in the industry. People work in isolation and study one set of gameplay to death. But many times, if they open their horizons, they will find that there are many roads leading to Rome.

Many times, the so-called failure of gameplay is really just the mismatch between one’s own abilities and gameplay.

I often say that compared to capital costs, the team's time trial and error costs are the most valuable. I think my GMV is higher than most people who read my articles, but I am often worried about failure in the future. I have also reminded the person in charge of the live broadcast room in many meetings -

Every extra day we make in the live streaming field means we are closer to the dividends dissipating.

Path planning for natural flow patterns

There is no right or wrong way to play, only pros and cons. The conventional order-holding strategy has the disadvantage of a short cycle, but it has strong explosiveness. The progressive strategy focuses on long-term operations, but it has weak explosiveness and strict requirements for explosive products.

When you think that AB chain can no longer be played, there are still people who are thriving in the live broadcast room. When you think that altitude sickness can no longer be played, there are a large number of teams applying the delivery-type gameplay.

Many times people lament that a certain gameplay is not playable. In fact, it is not that it is not playable, but that their own level is not able to master the gameplay.

The gameplay derivatives of each team are based on the people, goods and places in the environment at that time, as well as internal operational adjustments. This also means that most of the time, imitating the gameplay is tantamount to imitating others without success.

Rather than putting the blame on the non-disclosure of gameplay, I am more concerned about issues at the execution level. The theoretical level is full of confidence, but there is a disconnect at the execution level. This is the fundamental reason why most teams cannot play TikTok well even if they are given the gameplay.

From the perspective of a new team, I encourage novice teams to try short-term methods first rather than progressive methods, because with immediate results, they can collect execution-level results faster and gain data perception from the practical level.

Compared to keeping the idea in your mind forever, if you do it, there is hope; if you don’t do it, you will still be in despair.

There are many ways to get the gameplay. Be aware of all the training bloggers. Don’t get it for free and think that you will be ripped off if you spend money. You should know that compared to trial and error by yourself, spending money to buy a plan is a very cost-effective thing.

With feedback from the data layer, it is equivalent to measuring the deficiencies in the fundamentals. If the goods are not good, change them; if the scene is not good, change them; if the people are not good, optimize them. In short, the fastest way is not to verify your own strength, but to find out where the deficiencies are.

Is there anything wrong with short-term play? That's right, if you run one account hard, you can make 300,000 yuan, and 10 accounts can make 3 million yuan. How many live broadcast rooms can achieve such investment? Live broadcasts are done with the mentality of short-term profits. If it were me, I would do the same, but this is not me.

In the matter of Tik Tok, it is impossible to keep making noises and making noises. Those who only paddle on the shore cannot learn to swim well.

If you want to get involved in the long-term game, you must be prepared to endure. Things that lack dopamine always make people lack passion, but from the perspective of the future, only long-term effects can leave long-term assets.

The progressive method is considered a long-term gameplay, and other forms also exist. If you don’t know them, it means your circle is not big enough.

To evaluate the long-term gameplay, please read the article "20,000 words long article, 2022 live broadcast room algorithm disassembly", that is, look at the algorithm structure at the bottom. The growth of a single algorithm indicator often has a shorter-term gameplay, while the long-term gameplay pays more attention to the balance of live broadcast room indicators.

An abnormal traffic cold start model

In the common impression, for a live broadcast room with natural flow, the first thing to do is to attract traffic and open the scene, followed by interspersing welfare money, washing labels to undertake price gradients, and testing products and placing orders with profit money.

This is true for the idealized model, but in practice, problems arise when dealing with the details.

We can see that many live broadcast rooms rely on traffic-generating funds to open recommendations in the initial cold start, with the amount ranging from tens of thousands to hundreds of thousands. Then they try to use welfare funds to clean the labels and then fantasize about turning them into profit funds.

And the problem lies here. At the beginning, you use a very low-priced product to attract a lot of traffic to the live broadcast room. At this time, if you really want to remove the label and sell products at full price, you will find it very difficult.

Because your front-end crowd base is too large, the data eventually spirals downward, the live broadcast room drops in level, and the number of viewers drops to thousands.

A new account can open natural recommendations through interactive indicators in the early stage because from the perspective of platform assessment, as long as it can meet the popularity indicators, it can get natural recommendations. However, as the scene increases, the algorithm assessment becomes more diversified and will cover the transaction level from the popularity level.

This is also the point where most accounts die. They have completed the increase in popularity, but cannot reach the transaction level. So, how can they successfully cross the transaction level?

To solve this problem, we need to understand the impact of traffic scale on label washing. As a new account, there is almost no large-scale transaction data, the users are general and their behaviors are shallow. This type of user is extremely inaccurate in itself. As the scene expands infinitely, the difficulty of washing labels will naturally increase.

We can think of traffic and labels as a formula, with labels as the numerator and traffic as the denominator. Since we cannot increase the size of the numerator on a large scale, we must reduce the size of the denominator.

This is also the principle of progressive gameplay cold start design. At the beginning, you don’t need a very large field. Maybe ten or twenty thousand will be enough. In this case, you can introduce trading indicators to increase data and wash labels simultaneously. You will find it much simpler.

When starting an account, traffic is the starting point, taking over is the process, and explosive sales are the result.

For any set of gameplay, when formulating the first step, you must be able to imagine how to proceed in the second step. Each stage of the live broadcast affects each other. For example, being able to circle the size of the population with general traffic in the cold start stage will itself reduce resistance in the label washing stage, and vice versa.

On the price anchoring of the lead generation model

Assuming that we have accepted the "influence of traffic scale on label washing", there will be a phenomenon in operation, that is, the original design only opens natural recommendations at the beginning, but "inadvertently" expands the field of view.

I don’t want to do this, but the streaming is coming and I can’t stop it.

Don’t make excuses for the flaws in the traffic-generating models you designed. Looking at most of the live broadcast rooms that can tear up a large audience in a short period of time, the traffic-generating models all have one thing in common, that is, absolutely low prices.

In a live broadcast room where regular-priced products cost 200 yuan, it is not difficult to attract hundreds of people online by selling a style priced at 30 yuan at a loss of 9.9 yuan. But if you sell a style priced at 150 yuan at 59.9 yuan, try to attract hundreds of people online.

We need to establish a mindset that the steps of label washing start with the selection of drainage models.

The influence of drainage styles on user labels plays a very important anchoring role. One is style, and the other is price. Many people pay attention to the tone of the style because the style can anchor the basic population portrait and lay the foundation for the user's consumption attributes.

However, it is overlooked that only price can anchor an in-depth population portrait and lay the foundation for the user's consumption level.

For a cold-start live broadcast room, it is not necessary to use an extreme traffic-generating product to hold orders. On the contrary, you can use a high-order traffic-generating product.

For a live broadcast room that is in the cold start-up phase and does not need to be viewed by a large audience, adopting an absolutely low price will only create trouble for label washing in the later stage. What is the purpose of setting such a low threshold for attracting traffic? Will it be recommended naturally if it is opened?

At that time, I thought that my high-priced products would not attract so many viewers, but that was not the case.

The premise for products with high price anchoring is that the value of the product itself is very high. As long as the language and atmosphere of the live broadcast room are well-shaped, natural recommendations can also be enabled. The only difference is that the interaction index caused by the relatively low price compared to the absolutely low price is lower, and the scale of the audience is smaller, but the crowd is more accurate.

And the cold start itself does not require a large-scale viewing.

A breakfast shop with a daily average order of 9.9 yuan may place a 1.9 yuan egg at the door to attract customers. Even if the customer is not full, he or she may order a bowl of noodles for 10 yuan. A high-end tea restaurant with a daily average order of 599 yuan may place a 1.9 yuan egg at the door to attract customers. But after sitting down and seeing the expensive menu, the user will only order a glass of boiled water.

Cold start uses high customer orders to attract traffic, the essence of losing money is to pull vertically, and pricing often determines the traffic circle.

Another factor that turns low prices into regular prices: atmosphere

Using welfare funds to raise the price gradient and carefully clean labels are common means of converting to regular prices, but it is often encountered that the regular price cannot be converted even though the welfare funds can be taken over.

Many people overlook an invisible and key factor in live streaming, namely the atmosphere of the live streaming room.

The reason for this phenomenon is that most people can only remember it literally, but still cannot realize mentally that Douyin is essentially an interest-based e-commerce.

In interest-based e-commerce, the common consumption behavior of users is not to compare prices from different stores like searching on Taobao, but emotional consumption under the influence of the atmosphere.

Breaking down the specific consumption process, most users of Douyin buy something not because they need it, but because they are attracted by the content of the live broadcast room after entering it, and after seeing the product, scene, and the host’s explanation, they find that this thing is not bad, and thus place an order.

In fact, in most of my food and department store live broadcast rooms, there are almost no welfare items. Why?

Because a person who can afford food that costs 9.9 yuan can definitely also afford food that costs 29 or 39 yuan. If the user does not buy it, it is obviously not the price.

But the emotion is not there yet.

A live broadcast room with high turnover must be one that is extremely good at playing with user emotions, from the scene, product display, anchor's words, tone and rhythm, to the live broadcast screen, every detail must be grasped firmly.

Even as a live streamer in the same industry, I couldn't help but exclaim when entering a competitor's live broadcast room. The atmosphere is so fucking awesome!

Of course, this does not mean that welfare models are not needed. For a live broadcast room that sells T-shirts for 39.9 yuan, it is understandable to attract traffic at 9.9 yuan, but for the same summer T-shirts, which have a high customer order value of 300-600 yuan, lack of welfare models will cause problems because the crowd is stratified.

Secondly, another factor should be considered. The lower the price of the drainage fund, the more tiered prices need to be designed for the welfare fund.

Routine is an indispensable marketing "art"

Most of the live broadcast rooms with natural flow are inseparable from one word: low price. I analyzed the existence effect of low price in detail in June last year.

Because being greedy for bargains is a common trait among most people.

Once it involves giving the other party an advantage, routines are inevitable. Why do we have routines? I would ask in return, why can’t there be routines? Do you want to give users a bargain for nothing?

What is a routine? There is an offline barbecue restaurant with a sign at the door saying five dishes for 30 yuan, all you can eat. But when you enter the door you find that there are only five dishes, but what they are and how big the bowls are are two different things.

What is a routine? All the marketing methods that can be seen in the live broadcast room are routines.

The low average order value is used to hold back orders, 50 users are deducted, and only 20 orders are actually placed. Is this a trick? It was promised that the order would be placed immediately, but the anchor was reluctant to place the order. Instead, he chose to hold back the order and then bring the order online. Is this a trick?

The price was clearly 9.9 and was to be released to users, but when it was finally released, they took advantage of the herd effect of high traffic and directly released a product for 59. Is this a trick?

If you want to play the natural flow and raise an account, you must have routine thinking.

If you don't have routine thinking, and you bring a low-priced item to me, I'll give it to you, and then you'll want to buy the full-priced item from me, how is that possible?

The trick is not to sell counterfeit products, but to create a live broadcast room atmosphere and enhance the sales promotion effect of the live broadcast room through seemingly unreasonable means.

On the contrary, tactics that harm the quality of consumer products will not last long and are not encouraged.

But most novices do not have routine thinking, or are not good at using routines. If you don’t know how to play tricks, go and hang out in the live broadcast rooms that broadcast from 7 am to 10 pm, follow the dark horse-type natural flow accounts, and see how they play tricks in their live broadcast rooms.

If you are willing to spend a week setting up 5 to 10 live broadcast rooms every day, you can master at least 5 to 6 routines in a week.

The minimum threshold for a routine is your psychological threshold, and you must use the routine; the second is the disassembly threshold, and you must have the ability to disassemble the pattern; the third is the usage threshold, and the routine can run smoothly in your live broadcast room.

The mechanism of chasing hot products under the traffic scale

There are basically two ways to obtain high GMV through natural flow gameplay. The first is to constantly cater to the platform algorithm, stimulate traffic recommendations, and obtain high GMV through high viewership. The second is to wash the live broadcast room into a precise model after a cold start. The audience is not large, but with the mechanism of testing explosive products, explosive products drive high transaction rates, which directly increase the audience.

In the first half of last year, I generally adopted the first approach. Regardless of whether it was the Doujia account creation method or the Lucky Bag account creation method, the essence was to collect data for each game, increase traffic for each game, and then use the game observation to obtain a high revenue scale.

However, there is a problem. Data increment can be achieved in the short term, but maintaining high flow rate in the long term requires too high requirements for data incremental modeling. As long as there are a few drops in data and the account level drops, it will be difficult to pull it back to the initial traffic level.

If you want the natural flow gameplay to be long-lasting, you still have to go back to the mechanism of chasing hot products, that is, do not rely on the scale of traffic to generate revenue, but test hot products through live broadcast rooms. After using hot products to drive high GPM, the algorithm will continue to push traffic to continue transactions, and ultimately eat high weight to obtain periodic traffic.

What is an illusion that often occurs when creating an account? You may feel that the live broadcast room is popular because the audience is big, but it is not true.

If an account wants to pursue hot-selling products, it does not need a high-level view at all. Why?

For a live broadcast room, if you can ensure vertical traffic, for example, if you start broadcasting for 3 to 4 hours, not to mention 100,000 or 200,000 traffic, even 10,000 or 20,000 traffic will be enough for you to test for a hit.

Ten or twenty thousand traffic can ensure at least dozens of people are online during a four-hour live broadcast. Is it low to have less than 100 people online? Not low. Looking at Qianchuan’s current investment, there are only a dozen people online on average, while the daily transaction volume is basically over five figures.

To do Douyin, first you need traffic, then conversion, then explosive traffic

According to the experience of progressive gameplay, by tearing off natural recommendations in small steps in the early stage and then taking over through the process of welfare funds, the scale of the audience can basically reach over 10,000, and the traffic can be relatively accurate. At this time, as long as a hot-selling product appears, the live broadcast room will be blown up.

There is one benefit of promoting hot-selling products. As a large number of regular-priced products are sold, the weight of the live broadcast room will directly cross to another dimension.

From the perspective of traffic, the scale of natural recommendations is increasing. From the perspective of quality, labels are more accurate. In the past, 10,000 traffic could be sold for 8,000 yuan, but now 10,000 traffic can be sold for 12,000 yuan because the UV value has increased.

On Algorithm Structure and Script Arrangement

Live streaming is like a game of levels, merchants are the players, algorithms are the rules, and weights are the rewards.

A live broadcast room that can meet the interactive indicators is equivalent to proving its popularity to the algorithm, and the platform will give traffic recommendations. In order to achieve this point, in the new account stage, low-price orders must account for the majority.

Attract user interest at low prices, and obtain user behavior by holding back orders, and the whole connected together is the live broadcast script.

For a new live broadcast room, it is not difficult to create a script. A five-minute loop script is sufficient.

Because any user who enters the live broadcast room will not be able to stay in your live broadcast room for more than 5 minutes.

Since it is impossible to retain users for a long time, we need a set of five-minute user scripts so that every user can understand what is happening in this live broadcast room, what they can get, and how to participate.

What the 5-minute script does is to stay, interact, like, and comment, and then cooperate with the horse racing mechanism. As long as you can outperform your peers, you can get traffic rewards from the platform, thus completing the cold start of the live broadcast traffic.

If five minutes is a round, then each live broadcast is a theater.

The opening of a live broadcast is accompanied by extremely fast traffic, and popularity ranking is very important. In order to meet the algorithm assessment, single items are held back to take on traffic, increase the popularity of the live broadcast room, and crush peers in terms of stay, online, comments and other indicators.

After the extreme traffic, in order to maintain a balance between popularity and transactions, we began to make up for the intensive transactions, and cycled through welfare payments and low-price profit payments, using high conversion rates to make up for the transaction data.

For a live broadcast, the hour before the broadcast is crucial. As long as the algorithm's assessment of the data in the previous hour is met, more sustained streaming will be achieved later.

On the contrary, if a live broadcast is slow in the first hour, especially the first half hour, then after one and a half or two hours of broadcasting, you will find that the overall flow rate becomes slower.

After an hour, the traffic will fade away, the rapid traffic will decrease, and the traffic will become more precise, with hot products as the main body and new products with high click-through conversion rates as secondary products, and orders will be placed in a continuous cycle, which can ensure the core part of the entire GMV and continuously test new products.

At the end, we again use products with high transaction volume and low average order value, or even welfare products to end the show.

The first 30 minutes are the popularity model; the 30-90 minute transaction model is to sprint GPM; the 90-150 minute model is the hot-selling model; the 150-180 minute tail model is to continue to increase traffic and create system memory points

To sum up, in terms of user flow, follow the five-minute script rhythm to capture users, call users, and convert users; in terms of data flow, focus on popularity, density, and GPM; in terms of product flow, hold back orders at low prices when necessary, and promote products when necessary.

If each live broadcast is a theater, then multiple live broadcasts are a round.

The algorithm allocates the next broadcast’s traffic based on the traffic of a single broadcast, but only by weighting the traffic of multiple broadcasts can a continuous flow of traffic be obtained. Behind this is the operator’s ability to control each live broadcast.

As a qualified trader, you should treat every live broadcast as a brand new cold start.

The ceiling effect of a single account

When an account evolves to a certain level, you will find that the data is stagnant. Some people may suspect that it is the ceiling of the track, while others may suspect that it is the ceiling of traffic.

In fact, the team's ceiling may be the biggest obstacle to limiting revenue scale.

The capabilities of each team are limited, and the data level displayed by each account basically represents the team's operating level.

In my live broadcast room, once I encounter a growth bottleneck, I will solve the problem from three dimensions.

First, the team dimension. Check the data dimensions of the current traffic structure, including interaction data, traffic data, transaction data, and funnel data, to see which values ​​can be further optimized. To what extent is it optimized? Optimize to the point where the team can’t do better. At this stage, I would think that the team’s ability to pull a single flow has reached its ceiling.

Second, the channel dimension. When it is impossible to increase traffic in existing channels, we need to consider adding other traffic channels. For example, when the live broadcast room of natural flow encounters a growth bottleneck, we will start to increase the intervention of short videos. For the live broadcast room of paid flow, we will give priority to whether we can improve the input-output ratio through optimization. If that doesn’t work, we will consider the intervention of natural flow.

Third, the account dimension. If a single live broadcast room can no longer be optimized, we will consider adding a new live broadcast room and re-duplicate the live broadcast room based on our experience with the same category. In past operations, we have often found that most of the newly added live broadcast rooms can further increase revenue output and market share through matrixing.

Future models of natural flow gameplay

As the lowest-cost way to create an account, natural flow is still the choice of most small and medium-sized teams. We can also see a large number of live broadcast rooms with thousands or even tens of thousands of online users every day.

But in the near future, we have to accept the reality that there will be fewer and fewer natural live broadcast rooms with more than a thousand people online, and hundreds or even dozens of people online will become the norm.

In Douyin's traffic ecosystem, the growth of transaction users is increasingly lagging behind the growth of merchant live broadcasts, because from the perspective of traffic supply, the relationship between the two is not point-to-point, but many-to-one.

The platform's labeling of users has also become more complete. In the early days of live streaming sales, even if the price was low and the orders were held back, the system's push to the crowd was still unclear. But at present, the algorithm's classification of live broadcast room traffic labels has become very refined. This is why it has become increasingly difficult to transfer funds at low prices from last year to this year.

This has given people a lot of hints. The platform ecology is changing, and users are also changing. As merchants, we cannot always use last year’s experience to look into the future, and we cannot use existing experience to gamble on the future.

In my last live broadcast, I put forward a concept called "The key to gameplay design is conversion path, not traffic path."

A live broadcast room that only focuses on traffic will most likely have problems in the future, no matter how large the existing audience is. On the contrary, from the perspective of gameplay testing, we should think about how to design a live broadcast room model that has natural flow, low online traffic, but accurate labels and high transactions.

In the era of traffic, some people are lonely, while others become popular. This is the rule.

Every team that breaks through the circle in a new stage must have made some predictions before the new stage arrives and made preparations at the execution level. However, people are often prone to relying on past experience. For example, many people who sold goods on Doujia later failed to transform into live streaming. Similarly, many people who did Taobao e-commerce also failed halfway when they transformed into live streaming.

So, can you guarantee that those who are successful now will continue to make waves in the next stage? I think this needs to be questioned.

I'm just giving you a heads up because I'm already making this change.

Practical operations solve the fundamentals, and circles break the cognitive gap. If you lack cognition in the live broadcast field and have no idea about the practical operations, you might as well join my Knowledge Planet.

As the largest practical community in the live streaming field, you can join by swiping down to unlock the ultimate secret of selling products through live streaming.