First, because there are occasionally products that do not belong to the keyword under the keyword search results, the operator needs to use the COUNTO function to calculate the number of times the product appears on the keyword search page in the past 14 days, and then exclude links that appear 4 times or less. This will not cause misjudgment in the later data analysis due to retaining too much data, and can retain as many potential new products as possible.
Secondly, according to operational experience, if the ranking of the apparel category is within 100,000, it can be guaranteed that the link will be ordered every day. The value of ranking 100,000 belongs to the range of potential new products. At the same time, if the ranking of the product category is already within 5,000, and the daily sales volume is about 100 orders or more, it is generally difficult to beat the competitor. Use the MINO function to analyze the best link ranking in the past 14 days, and retain the links ranked between 5000 and 100,000 through digital filtering.
Digitally filter the data in this column, select the “Between” option, and set the filter condition to greater than or equal to 5,000 and less than or equal to 100,000 in the pop-up dialog box. For products in other categories, the value can be adjusted according to actual conditions.
At the same time, even if the life cycle of the category is in the rising period, the ranking of each product will fluctuate up and down. Among them, the ranking of potential products may decline within a few days, but the ranking will generally continue to rise within 14 days. At the same time, the fluctuations in the early stage of the link are often more obvious. After one order of clothing products is put on the shelves, the ranking can rise to about 300,000. If the sales volume continues to grow, the ranking will rise at a very fast rate. Using the MIN function, calculate the lowest ranking value in the first cycle minus the lowest ranking value in the second cycle to find out how many places the link ranking has risen in the two cycles. Fill in other cells downward to finally get a table.
According to this table, it can be seen that the link ranking fluctuates with different amplitudes. According to operational experience, the ranking trend of hot products often rises exponentially, while the ranking changes of ordinary styles are more gentle. Therefore, in the early stage of new products, the ranking changes of hot products are often more drastic than those of ordinary styles. At the same time, due to the unstable sales volume in the early stage of the link, the ranking is likely to fluctuate from millions to tens of thousands, so the operator can filter the ranking difference to find products with higher ranking fluctuations.
Digitally filter the data in this column, select the “greater than” option, and set the filter condition to be greater than or equal to 50,000. Similarly, for products in other categories, the value needs to be adjusted according to the actual situation.
Finally, due to the influence of the life cycle, many old links will rise in ranking again when the next sales season arrives. However, it is difficult for these links to be used as a breakthrough to create explosive products with historical accumulated comments. At this time, the listing time can be sorted in descending order to find the most recently listed links.
After these 4 steps, about 100 potential links can be screened out from more than 10,000 links. Operators can also conduct secondary sorting and find products that are more in line with the operating conditions for development through specific comparisons on the product details page. In addition to the three types of data such as the number of crawls, the highest historical ranking, and the rising ranking, the product evaluation and production difficulty can also be comprehensively compared to finally determine the product selection.
It should be noted that since this method belongs to the in-site product selection, for special categories such as toys, it is also necessary to check whether the product itself has infringement risks, otherwise it will face the risk of being complained and closed later.