當一個擁有較少粉絲數的直播賬號開播時，很多人會覺得會沒人看，但現實是在線觀看人數很多。

When a live streaming account with fewer followers starts broadcasting, many people may think that no one will watch, but the reality is that there are many online viewers.

出現這個現象的原因就是大家混淆了直播權重和短視頻賬號權重，實際上二者為獨立運行的兩個系統，互不影響。

The reason for this phenomenon is that people confuse the weight of live streaming and short video accounts, but in fact, the two are independent systems that do not affect each other.

1、抖音直播間權重

1. Weight of Tiktok studio

這些數值綜合起來就產生一個綜合數值，也就是我們的直播間權重。而這個綜合數值作為電商直播運營人員大家都是無法看到的，如果這個數值的具體情況被公布，作為電商直播運營人員完全可以靠技術手段把任何產品都打造成爆款直播間，這對于平臺的良性運營是不利的，電商平臺只有把流量的價值更高的轉化才能體現出平臺的高價值，如果把平臺的計算數值的方式公布就會導致流量被手動操控在不良產品和直播間手中，而不是以產品優劣為導向分配流量，就像是考試前就得知了正確答案，這自然能獲得好的成績，但這對于其他考生是不公平的，自然這個考試的結果也就沒什么意義了，而這個權重數值計算方式也就相當于是考試答察。

These values are combined to produce a composite value, which is our live broadcast room weight. However, as e-commerce live streaming operators, everyone cannot see this comprehensive value. If the specific situation of this value is announced, as e-commerce live streaming operators, they can rely on technical means to turn any product into a popular live streaming room. This is not conducive to the healthy operation of the platform. E-commerce platforms can only reflect their high value by converting the value of traffic to a higher value. If the calculation value of the platform is announced, it will lead to traffic being manually controlled in the hands of bad products and live streaming rooms, rather than being allocated traffic based on product quality. It is like knowing the correct answer before an exam, which can naturally achieve good results. However, this is unfair to other candidates, and the result of this exam is meaningless. The, And the calculation method of this weight value is also equivalent to exam answering.

2、相似標簽和賽馬機制

2. Similar tags and horse racing mechanisms

相似標簽主要包括用戶基礎信息（內容偏好、購買偏好）、用戶行為數據（人氣數據和成交數據）和直播間標簽（淺層標簽是人氣模型，深層標簽是成交模型）。

Similar tags mainly include user basic information (content preferences, purchase preferences), user behavior data (popularity data and transaction data), and live room tags (shallow tags are popularity models, deep tags are transaction models).

賽馬機制主要是看人氣數據和成交數據，這兩個數據代表直播間權重。

The horse racing mechanism mainly relies on popularity data and transaction data, which represent the weight of the live broadcast room.

這時的賽馬機制是這樣的：初始流量→賽馬排名→優先推流。開播后的初始流量是由上一場直播的數據決定的，根據上周的直播數據和上一場直播的數據，決定了初始流量是多少的，這樣基本是開播15分鐘之內，至于第2波、3波、4波流量主要是根據人氣數據和成交數據決定的，數據好就會繼續推流，數據表現不好就不會推流。

The horse racing mechanism at this time is as follows: initial traffic → horse racing ranking → priority promotion of traffic. The initial traffic after broadcasting is determined by the data from the previous live broadcast. Based on last week's live broadcast data and the data from the previous live broadcast, the initial traffic is determined, which is basically within 15 minutes of broadcasting. As for the second, third, and fourth waves of traffic, they are mainly determined by popularity data and transaction data. If the data is good, the traffic will continue to be pushed, and if the data performance is poor, the traffic will not be pushed.

三、直播間自然流推流的決定因素1、直播間流速的背后邏輯。

3、 The determining factor of natural streaming in live streaming rooms is 1. The logic behind the flow rate in live streaming rooms.

停留時長定義內容質量:停留時長是決定直播間推流的一個重要因素(5分鐘推流一次)但停留時長沒有一個標準值，比如你新號開播的時候，同時間段跟你一起開播的競爭對手，他的停留時長只有19秒，而你有20秒，那么系統一樣會給你推流。如果你是老號，粉絲沉淀到了一定程度后，你得區分開粉絲的停留時長跟新用戶停留時長，這個數據在電商羅盤里可以看到。

Definition of dwell time and content quality: dwell time is an important factor in determining the streaming quality of a live broadcast room (once every 5 minutes), but there is no standard value for dwell time. For example, when your new account starts broadcasting, if a competitor who broadcasts with you during the same time period only stays for 19 seconds, while you have 20 seconds, the system will still stream you. If you are an old account and your fans have accumulated to a certain extent, you need to distinguish between the length of time your fans stay and the length of time new users stay. This data can be seen in the e-commerce compass.

成交密度決定內容效率:在同等的流量和曝光情況下，你的訪客效率最低，為平臺產生的利益最小，那么系統就會把流量給到更優質的直播間，所以成交密度也會影響你的推流速度?？亓魉俚幕具壿嬀褪前涯銌挝粫r間內的轉化單量跟單位時間內的轉化金額做好如果能把這兩點做好，你的流速就相對可

Transaction density determines content efficiency: under the same traffic and exposure conditions, your visitor efficiency is the lowest and the benefits generated for the platform are the smallest. Therefore, the system will allocate traffic to higher quality live streaming rooms, and transaction density will also affect your streaming speed. The basic logic of controlling flow rate is to balance the number of conversion orders per unit time with the amount of conversion per unit time. If these two points can be done well, your flow rate will be relatively stable