Source: Youmi Youshu e-commerce classroom
01. In the final analysis, the selection is "common sense of life"
When merchants select products, the most concerned problem is how to determine the trend of a certain product.
When we select products, we most often use the "selection calendar" function. This function provides us with historical data of the best-selling categories and products in the past two years, and we can trace the best-selling categories and products according to different time periods to assist in product selection.
Taking the data during the Double Day period in 2024 (mid-December ~ early January) as an example, why is the category with the highest sales volume being "snack food"? Because the New Year is about to come to the Spring Festival, snack food is a must for people to reunite for the New Year and visit relatives and friends for the New Year.
There are meters and a number of "selection calendars" - 2024 Double Day period data
In the period before the Spring Festival (mid-January to early February), the sales of "daily department stores", especially home cleaning, kitchen supplies and other sub-categories, are the highest, because the Spring Festival is approaching, Chinese have the custom of "cleaning", to say goodbye to the old and welcome the new, to welcome the new year, for household cleaning supplies, storage, kitchen utensils, etc. are just needed in this period. In addition, for many female consumers, they always have to dress up beautifully during the New Year, and they will also have a greater demand for "beauty and skin care" and "clothing, shoes and bags".
There are meters and a number of "selection calendars" - data for the period before the Spring Festival in 2024
Offline business is still inseparable from each of us's food, clothing, housing and transportation. What kind of time node, what kind of life scene needs. For example, on the eve of the Spring Festival, there is a need for Spring Festival migration, cleaning, and preparation of New Year's goods; When resuming work after the Spring Festival, many migrant workers face the need to change jobs and move; These seasonal needs are "common sense".
Therefore, the selection of products should be based on different time nodes and the needs of people's life scenarios to choose suitable products. Considering the common sense of seasonal life and human needs, combined with the "product selection calendar" function, it can more accurately identify the trend of popular models, select the right products, and achieve explosive volume.
02. Use the "Trend/Potential List" to verify the direction of product selection
Generally speaking, the selection and preparation need to be 1~3 months in advance, and in March, you need to start preparing the goods in 4~5 months, for example, for home textiles, you need to consider preparing summer cool quilts and other summer products from March.
When we check the best-selling categories last year or the year before from the "Product Selection Calendar" and select a category direction, how to judge whether the category will explode in the same period this year?
Under normal circumstances, if the selected product has been sold out last year or in the same time period for two consecutive years, then if the customer unit price is appropriate, there is a high probability that it can still be sold out in the same period this year, but the quantity is small.
In a more secure way, we can use the "trend list" and "potential list" to help verify whether the direction of product selection is accurate.
There are meters and a number of "potential lists"
If the product appears in the recent "trend list" or "potential list", even if the current sales are not so high, there may be a trend of explosion in the future.
But on the contrary, if a category or product appeared in the "trend list" or "potential list" in March this year, but there was no explosion or even no explosion in April ~ May of last year or the year before that, the probability of the product in the next two months of this year is uncertain. Because, this product has not been educated on the Douyin e-commerce platform in the past, if this product is blindly promoted, the cost of educating users is relatively high, and the possibility of success is difficult to say.
In addition, we also do not recommend merchants to refer to the "Hot Product List" to select products. Since "explosive products" have usually run out of a certain volume, if merchants enter the game with the same products, they may not be able to compete for the market share that has been occupied. We also encourage merchants to try some products that have not yet fully exploded, but have the trend and potential to explode.
The key to product selection, on the one hand, is to choose the right time to enter the game, and on the other hand, to choose how to enter the game when the product has shown an explosive trend. The most accurate selection is always inseparable from "common sense of life".
03. Material reference focus in "look at the beginning" and "find hot spots"
Product selection is the first step, and materials are the key to explosion. How to find high-quality materials and learn from them and create your own?
First, first look at the materials of your own category track, and then learn from them across categories. Starting from the needs of our own products, crowds and scenes, for example, the products we want to play want to shoot materials through the style of "street photography", then we can look at the materials from the categories that are strongly related to our own products + cross-categories such as clothing categories that often use the "street photography" style.
Second, instead of filtering materials by data performance, pay more attention to the beginning copy and picture of the material. Unlike many businesses that use data indicators such as "material quality index" to screen high-quality materials, we are not limited to its "quality" when referring to materials, considering the common appreciation of elegance and vulgarity, different groups of people need different materials, and each material also has its highlights and shortcomings. However, it will focus on the opening screen and the first 3S lines of the video, and you can also find high-quality materials for the top 3S lines that have been popular recently.
There are meters and a number of "golden 3s lines"
Third, material creation should be good at learning from hot spots. On the one hand, through the "new product list" to track the recent hot new products, their materials may have new ways to play for reference; On the other hand, drawing on recent social hot spots, such as Jia Ling's "Hot and Hot" some time ago, brought up the hot topics of weight loss and fitness, such as the first 3S lines of the original material of beauty products, "Which girl doesn't like nude makeup and no makeup", combined with the hot topic of weight loss, it can be modified to "Who said that fleshy girls are not good-looking, women actually rely on dressing up", and implant with hot topics, which can enhance the user's sense of substitution.
Fourth, use AI tools to assist in video script creation. For businesses that don't have the ability to write material scripts, it is better to use AI generation tools to rack their brains and think out of thin air. You can extract every sentence of the script copy of the high-quality materials you find, and enter them sentence by sentence through the "AI-generated script" function, so that the AI can intelligently generate more choices, and then replace and modify them with scripts suitable for your own category scenes to shoot materials.
The material reference focuses on watching more, thinking more, and analyzing more, accumulating high-quality pictures and copywriting, forming a library of various video elements, and then borrowing them from your own product materials according to different scenarios.
04. The material is disassembled in seconds, and the material is efficiently analyzed and optimized
When a video material is released, it is particularly important to analyze and review the post-cast material. The "second-level disassembly of materials" function points out the direction of material optimization for us.
By accessing the Qianchuan account, you can present the second-level dismantling data and trend chart of each material in the account, and support one-click extraction of the "highlight clip" of the material.
So, how do we analyze and judge the optimization direction of a video material based on the second-level dismantling data of a video material?
The data will tell you what a video clip really looks like. Generally speaking, the qualified click-through rate of a video material is more than 2.3%, and the click-through rate of a stable content is 3%. The click peak of a video material is the place where the copywriting and picture of the material are relatively high-quality and attract users. The larger downward trend is the place where the user loss is large, and it is also the part of the material that needs to be optimized.
Take the second-level dismantling data of a post-cast material as an example: the peak click of the video appears in the 3rd s, and the second peak of clicks appears in the 18s, and the copywriting and images before and after these two points in time are relatively high-quality; And after the first 3s of the video, a relatively large landslide began to appear, obviously the copywriting or picture at the beginning of this connection is problematic, and there is more loss.
For the follow-up optimization of the video, you can consider moving the high-quality copywriting and picture at the 12~18s time point forward to improve the completion rate of the first 3s + grass planting link data, then the closing data in the back is generally not too bad. However, if the closing data falls relatively quickly, you also need to consider optimizing the closing mechanism copy and picture to keep it as stable as possible.
There are several cases of "second-level dismantling of materials".
"Second-level dismantling of materials" is especially practical for new businesses that have not been in the game for a long time, lack of experience, and do not know how to optimize materials, and it is also a powerful tool for testing new materials and using second-level data dismantling to verify the performance of every picture and copywriting effect of new materials.
Its core value lies in the fact that in the past, if a video material did not have an explosive volume, the effect would have to be redone, but now with the "second-level disassembly" data analysis, we can optimize the problematic clips according to the data performance of each second, and the data performance can be efficiently reused, and the highlight frames can be extracted with one click. This reduces the cost of material optimization and makes material optimization more scientific.
There is no shortcut to material analysis, and every business must be good at looking at data to analyze the effect of materials.