Turn on Douyin, and there are various delivery modes. Those who eat and broadcast, those who shout wheat, those who dance, those who talk about cross talk, all kinds of forms of expression can be used as a means of bringing goods.
But how to replay the actual effect? Once the perceptual mode of carrying goods returns to rational analysis, the problem will not be so simple.
The key metric for e-commerce is GMV. What factors affect GMV in live delivery? It usually mainly includes: total number of viewers, audience stay time, and total live broadcast time.
But here comes the problem: These statistical-level data cannot reflect process factors. Even, sometimes the real situation in the live broadcast is completely opposite to our understanding of the data.
For example, the following picture is a real data statistics chart in a recent popular live broadcast room:
The green curve in the figure represents the transaction amount, and the red curve represents the number of real-time online users.
According to general rational judgment, the transaction amount of the live broadcast room is positively correlated with the number of real-time online users. But from this picture, we found that from about ten o'clock, as the number of real-time online users dropped sharply, the transaction value did not decrease accordingly, and the transaction value even reached a new high at certain moments.
What is the reason behind this somewhat counterintuitive phenomenon? This is the so-called digitalization and refined operation.
On November 18, the consumer brand tea party jointly organized by "Whale" and 36 Krypton was held in Shanghai as scheduled. In this tea party, we will use "New Consumer Brands x Digitalization - How can new consumer brands maintain vitality?" " as the theme, to discuss the experience and trend of new consumer brands in digital practice.
The tea party brought together the product director of Weiwei, the general manager of Lin Qingxuan Digital Information Center, the founder of Daily Fungus, the founder of TNO, the founder of MYOMY and other cutting-edge technology companies and representatives of cutting-edge brands. Everyone shared their successful practices and pain points on the road to digitalization Another main topic at the meeting was the above-mentioned problem: how new consumer brands can initiate a real digital and refined operation reform in the live broadcast room.
01
The essence of data analysis is anomaly detection
In the new consumption era, if a brand wants to achieve sustained commercial success, it cannot simply rely on uncertain factors such as "traffic outlets" or extensive operating methods.
Brands need a more systematic and effective marketing operation methodology, as well as a set of handy digital tools, so that the brand's marketing operation strategy is truly targeted and effective.
At the meeting, representatives of various brands discussed topics such as "the first principle of retail growth", customer insights, and user experience decision-making insights.
So, to what extent can the theory be truly transformed into an "Aha Moment" to retain consumers in real business, and ultimately lead to transformation and long-term user value? In this regard, the founders and representatives of various brands shared their highlights and experiences in the process of digital transformation.
Among them, the guests present summed up some thought-worthy views and shared them with you:
First of all, data analysis cannot tell you what is right, it allows you to find abnormalities.
For example, in the case of the above-mentioned live broadcast room, the data tells you what is abnormal and where you need to focus on and analyze. The essence of data analysis is anomaly detection.
Second, it is difficult to gain insight into something deeper from statistical-level data, and it can even be deceiving. You need procedural metrics.
Third, live broadcast analysis, there must be no game missing.
If you ask the boss: Which live broadcast is the most successful in your history? His answer is mostly: the next game.
Brands must learn from each live broadcast to optimize the next live broadcast. This is a truly data-driven benign process. Your user base is constantly changing, and you need to iterate continuously and quickly to keep up with or even lead the trend.
02
What kind of review is called in-depth review?
Since statistical data is not enough, what kind of replay can allow brands to truly grasp the key variables in the live broadcast room?
At the meeting, Liu Jia, head of Weiwei products, shared his experience and views.
Liu Jia mentioned that in order to do a good job of review, we must first do a good job of attribution.
Just like a person loses weight, the first thing to know is that the key to losing weight is exercise and diet. Secondly, further dismantling, which exercises are suitable for weight loss and applicable to me, how many calories are in various foods and drinks, and based on my weight loss goal, what level should I control my diet every day.
This seemingly ordinary thinking process actually includes the two most important factors: attribution and modeling.
The same is true for live replays. Just like in the above case, why did the turnover of the live broadcast room not decrease after the number of real-time online users dropped significantly? It may be that the anchor said a certain stalk during the interaction, it may be that a certain word moved potential consumers, it may be that the anchor sent a lucky bag, or even a cat appeared in the live broadcast room during that time.
The above factors are the attribution elements in the replay. Based on these modeling, by doing AB test and other means, you can make your live replay truly go from the result index to the process optimization.
03
A wonderful replay, what exactly did it bring?
As we mentioned above, a breakthrough in the digitalization of live broadcast rooms is to use attribution and modeling to truly move from result indicators to process optimization.
In actual combat, to realize this process means that you must also digitize the process .
During the tea party, the participants were divided into groups and used the "six thinking hats" method to conduct a simulation analysis.
First of all, everyone watched a video with goods in a live broadcast room. Then, each group gave their own evaluation of the performance of the anchor in the video.
According to the prearrangement, the red hat group gives emotional and intuitive feelings without much thinking; the yellow hat group gives positive judgments, including the advantages and value of the anchor; Look for questions in the live room.
Through the above-mentioned simulation discussion, everyone found that no matter it is an individual or a certain characteristic organization, it is easily affected by subjectivity and stereotyped prejudice , which makes the review analysis unable to be truly objective and true, let alone ensure that the review given by oneself Whether the disk results can really bring positive value to subsequent operations.
This is also the value of "Whale Cast" . By visually viewing the indicator data needed by the brand in the live broadcast room Kanban, deconstructing the live content and data, and through highlight content tracing and semantic label deconstruction , to help brands and anchors quickly locate and discover this The success factors and abnormal problems of the live broadcast, and finally empower the brand live broadcast plan and content structure optimization precipitation.
At the meeting, Liu Jia also specifically pointed out that although a wonderful live broadcast replay brings accumulated experience and can be reused efficiently by other live broadcast rooms of the brand, reuse does not mean copying. The purpose of replay is not to make all live broadcasts Same thing.
On the contrary, the purpose of the replay is to help the live broadcast studio find its own distinctive successes and consciously magnify their advantages. Just like the precise recommendation of algorithms on many platforms, it is a continuous positive promotion.
epilogue
The premise of consumer value operation is precise insight into consumers. This not only requires the brand's super-high sensitivity in sensibility, but also inseparable from the rational level, making objects recognizable, accessible, and analyzable.
As a brand of AI Copilot , Weiwei provides brands with data and artificial intelligence-driven actionable insights (Actionable Insights) and Operation Automation helps brand marketing operations reduce costs and increase efficiency, and create a deeper connection with hundreds of millions of consumers, thereby promoting the digitalization process of global brands on the consumer side.