In reality, many students will ask this question: Why is there no natural traffic in my live broadcast room?

In fact, before thinking about this question, what we should think about is: what kind of traffic should the system allocate to us? And why is such traffic allocated to us?

In the previous article, we have shared with you the Douyin e-commerce live broadcast recommendation model. Among them, the first step is very critical, which is to let the system identify: what kind of user is more suitable to enter our live broadcast room? And what kind of users are more likely to interact or consume in the live broadcast room?

To achieve this goal, the prerequisite is: our account and live broadcast room establish stable population labels. Only when stable labels are established can the system recommend the live broadcast room to the corresponding population based on the labels; and only after these people have interacted or consumed in the live broadcast room, the system will refer to the population portrait that generated the conversion behavior and recommend more similar users to our live broadcast room.

In other words, before starting Douyin e-commerce live broadcast operations, the first thing we need to think about is: what kind of crowd labels should be established for our account and live broadcast room? How can we establish stable and accurate population labels?

Here we need to mention a kind of thinking: group thinking.

The so-called crowd thinking refers to: based on the recommendation intelligence of similar groups on Douyin, quickly find the core groups under the account, and establish core group labels based on this, so that the system can identify our core groups more accurately, and thus complete diffusion and conversion more efficiently.

From the perspective of business circulation, people can be divided into two categories: fans and trading people .

The fan group refers to the group of people reached by the content posted by the account, including short videos and live broadcasts, that is, new fans converted through content and loyal fan groups; while the transaction group refers to the group of people who make purchases after seeing the products on Douyin.

Many students may say, aren’t fan groups and trading groups the same thing? Actually, it is not.

For example: When I see a luxury goods advertisement, I will most likely become a fan of the brand because of the style and tone of the video content, but will I definitely be a trading partner? Not necessarily, because its price may be beyond my affordability at the moment; but when I see an advertisement for a daily necessities product and the price is very cheap, and my family just needs it, I will most likely place an order immediately, so I am a trader of this product, but will I become its fan? Not necessarily. After all, I have no emotional connection with this product. When I see better products and content-seeking videos, I am likely to fall in love with someone else.

This case fully illustrates that the fan group and the trading group may overlap, but the fan group is not necessarily equal to the trading group.

After having a certain understanding of the fan group and the transaction group, let’s return to our topic - the core thinking that should be mastered when doing Douyin e-commerce: crowd thinking.

In the background of Doudian Compass, we can get a set of data called crowd portrait. Crowd portraits display the characteristics of the crowds flowing along the two paths of content and products. There are two types of population characteristics: the first is the basic user attributes, which are what we often call statistical attributes, such as user age, gender, region, etc.; the other is user population preferences, including purchase preferences and content preferences, which refers to what kind of products and types of content the users who intersect with you are more interested in.

For merchants, they should accurately locate the core consumer group under the account based on the basic attributes of the group; then, based on the group's preferences, reversely optimize our content so that our content can reach the core group more accurately and bring in a wider range of users.

For example, if we are running a knowledge-based paid account and our service target are English test takers facing the high school entrance exam, then our content (including short videos and live broadcasts) should be practical content that serves these students. Only in this way can the system better identify your target users and import more similar people into your account and live broadcast room.

Generally speaking, Douyin officially divides consumer groups into 8 categories, namely:

Small town youth and small town middle-aged and elderly people refer to people living in fourth-tier cities and below, aged 32 or younger;

Generation Z refers to young people under 24 years old living in third-tier cities and above;

Exquisite mothers refer to white-collar women aged 25 to 35 who are planning to have a baby or have already had a baby and live in third-tier cities or above.

New white-collar workers refer to white-collar workers, IT workers, and financial workers aged 25 to 35 living in third-tier and above cities;

Senior middle class refers to white-collar workers, IT workers, and financial workers aged 36 to 50 living in third-tier cities and above;

Urban blue-collar workers refer to people living in third-tier cities or above, aged 25 to 35, with medium to low spending power;

The last group is the urban silver-haired population, which refers to people over 50 years old living in third-tier cities and above.

What we need to remind everyone here is that this consumption capacity refers to the consumption capacity on Douyin, and does not represent the actual consumption capacity of the corresponding group. But we can still use this as a direction and reference to develop a core consumer profile for our account. For example, the knowledge-based paid products mentioned above should correspond to the officially defined “Generation Z” as their consumer group.

However, even for the same product, our content can target completely different groups of people.

For example, for the same knowledge-based paid product mentioned above, we can also target parents who are concerned about their children’s English performance. However, because the target audience is different, whether it is short videos or live broadcasts, the content provided by this account should be clearly different from the previous account.

To sum up, the so-called crowd thinking means that before we create an account, we must first clarify who the core consumer group is, and then customize the production of high-quality short videos and live broadcast content based on the needs of this group of people, and through operational means, continuously expand the proportion of this core consumer group in the entire crowd portrait . In this way, the system will continuously import accurate traffic to our live broadcast room, and the scale of our business can also be expanded.

02 How to formulate clear account positioning based on crowd thinking?

So, how can we develop a clear account positioning based on crowd thinking? We first need to understand a major premise, that is: we need to choose the account type.

Generally speaking, account types are usually divided into three categories: the first category is brand accounts, the second category is product accounts, and the third category is IP accounts.

The so-called brand account, as the name suggests, is an account that emphasizes the brand and is created for the brand. The purpose of building this account, in addition to product sales, is also to showcase the brand image and realize brand value sedimentation. This type of account has a strong commercial attribute, so the overall traffic will be more dependent on paid traffic, and the account page decoration will be more refined. Its counterpart is Tmall's flagship store page, such as Xiaomi's official flagship store, which will publish videos related to Lei Jun;

The second type is product-type accounts, which focus on the sale of products. Their characteristics are that the content and products are highly matched. Therefore, the accuracy of the covered population is relatively high, and the marginal cost of acquiring accurate users is low. For example, the store accounts under Xtep will publish product-related videos.

The third type is IP-type accounts. The core of IP-type accounts is to establish a personality that is recognized by many users. The core appeal is to bring higher monetization capabilities based on the accumulation of fans. Generally speaking, fan recognition and fan stickiness are its biggest advantages. For example, Lafeige from Jiuxian.com is a typical IP account.

Obviously, each type of account has its own advantages and disadvantages. We strongly recommend that you build a matrix account. For example, Jiuxian.com has established both brand accounts and product and personal accounts.

After understanding the classification of account types, let’s learn how to locate accounts.

In the previous article, we mentioned that core population positioning is based on two points. The first is population preference positioning, and the second is population basic attribute positioning.

Crowd preferences are divided into purchasing preferences and content preferences. Purchasing preferences can be understood as "product categories", while content preferences refer to "content categories". Among them, "product category" is the first-level category we choose for the product when we put the product on Doudian. Currently, Douyin provides 19 major business categories, including 121 first-level categories. The keywords corresponding to each first-level category can be understood as user purchasing preferences; and "content category", we can view it through the Star Map Expert Classification, which is divided into 24 categories in total.

A special reminder here: In the daily operation process, we need to pay special attention to the proportion of purchase tags and content tags in the Doudian Compass background under our account to ensure that the core categories and core content categories we operate always occupy the highest proportion, so as to ensure that the system matches us with sufficiently accurate traffic.

After talking about crowd preference positioning, let’s take a look at the basic attribute positioning of the crowd. The basic attributes of the crowd are also determined by both products and content. The difference is that the attributes behind the product, the price, and the personality and scenario behind the content will jointly determine the basic attributes of the user.

For example, the screenshots corresponding to the following three groups of videos all sell women's clothing for work, but the product in the first video is priced at 89 yuan. The live broadcast scene is a factory, and the anchor's personality is a factory employee. Therefore, his consumer group is concentrated in low-customer-order women in the workplace, which corresponds to the urban blue-collar workers among the eight major consumer groups defined by the official definition;

In the second video, the product is priced at 198 yuan. The scene is an offline store, and the anchor is a clerk in the offline store. Therefore, his consumer group is concentrated in the middle-income working women who are used to shopping offline before, that is, the new white-collar workers among the eight major consumer groups.

In the third video, the product is priced at more than 800 yuan, the host’s personality is a successful woman, and the live broadcast scene is a high-end residence. Therefore, the consumer group is concentrated in urban working women with stronger spending power, that is, the senior middle class among the eight major consumer groups.

These three cases tell us that the attributes and prices of products will affect the user portrait, and the scenes and characters of the content will also affect the user portrait. Therefore, in the process of building and optimizing accounts, we must pay attention to ensuring that the user portrait mapped by the product attributes and prices under the product is consistent with the user portrait mapped by the scenes and characters under the content.

For merchants, our inventory is fixed, and the product categories, attributes and prices are relatively clear. Therefore, how to create the character and scene positioning under the content is the key to operating Douyin.

03 How to set up a good account’s personality and scene positioning?

Generally speaking, the personality positioning of an account can be divided into three categories.

The first category is high potential energy, the second category is flat potential energy, and the third category is low potential energy.

Common high-potential personalities are experts, idols, and bosses. They usually use these personalities to build fan trust and bring in follow-up consumption . Average-potential personalities usually represent close friends and besties . The purpose is to reduce the sense of distance with users and ultimately lead to recommendation-based consumption. Low-potential personalities refer to sales and promoters . These personalities do not have a strong presence and are mostly based on product selling points. Their purpose is to create a better sales atmosphere to attract user consumption. From the current perspective, brands often use "low-potential" personalities for self-broadcasting.

For businesses, the threshold for establishing a low-potential personality is undoubtedly the lowest, but it will correspondingly increase the reliance on "quality" and the "script/rhetoric" of the live broadcast room; a persona with average potential will test the team's content creation capabilities; finally, a high-potential personality has better growth potential, but the requirements for "people" will be more stringent. At the same time, a certain period of time is required to accumulate fans for the account.

So, how can we establish a different personality?

The most direct way is of course to publish a video.

Relatively speaking, when creating a high-potential personality, the quality and requirements of content release will be higher, in order to gain recognition and trust from fans. The goal of video creation is to attract fans and build fan trust. For the creation of a medium-potential personality, the video content should be designed around the preferences of similar groups of people. For example, for a dressing account, you can share shopping/dressing concepts and your own life experiences. The goal of video creation is to stimulate user empathy so that short videos can get better feedback and exposure. For low-potential accounts, the focus of video creation will be on the product itself, including its functions, usage scenarios, and promotional policies. The core appeal of video creation should be to highlight the value of the product to stimulate user interest in the product.

In addition to character positioning, let’s take a look at scene positioning. The scene here mainly refers to the live broadcast room scene.

Generally speaking, we divide the scenes in the live broadcast room into three types: product type, personality type and companionship type.

Product-type, as the name implies, is mainly about displaying products in the live broadcast room. This type of scene is also the most common scene type for merchants to conduct Douyin e-commerce live broadcasts. It is more suitable for low-potential anchors. It should be noted that in the process of building a product-type live broadcast room scene, it is necessary to maximize the selling points of the product and lock in the core target consumer group ;

Personality scenes are more common in high-potential and medium-potential anchor personalities. When we build the live broadcast room scene, we should highlight the characteristics of the anchor's personality .

The last type is the companionship type. This type of e-commerce live broadcast room is relatively rare. The decisive factor for users to stay in the live broadcast room does not lie in the products and the host’s personality, but depends more on the theme of the live broadcast room and the corresponding gameplay .

To sum up, the live broadcast room is the key for consumers to complete transactions and conversions on Douyin, so scene building is very important, and the scene we build should also be strongly related to the character positioning and core sales products. The lower the potential energy of the character, the more it should highlight the product itself, and the higher the potential energy, the more it should be combined with the characteristics of the character to build the scene.

04

Finally, I will provide you with some tips on how to quickly create tags when your account is cold started.

First, in the process of account decoration, we need to add more keywords related to products and content in the personal description, so that the system can complete more records in a short time, so that the system can better identify your target users and bring you relatively accurate traffic;

Secondly, when publishing content and starting live broadcasts, you should also make relevant targeted settings. If we are doing e-commerce live broadcasts, we need to turn off local recommendations and friend recommendations to improve the accuracy of traffic;

Finally, we need to log in to the creator backend and the enterprise account backend to fill in the content category labels for the account. During the cold start phase, the labels we fill in ourselves will also serve as a reference for the system when it comes to identification.