Recently, I’ve been uploading dynamic heads from a different account and regularly checking the best sellers to evaluate their performance compared to others. However, the sorting for “past day” and “past week” across all categories does not accurately reflect the best-selling items. Instead, it seems to prioritize items based on the algorithm’s interpretation of user preferences, rather than actual sales data. This makes it difficult for me to see how well my item is actually performing and does not provide objective data.
Reproduction Steps:
- Go to the best sellers section for dynamic heads (or any other category), and select the “past day” or “past week” filter.
- Compare the rankings while logged in and logged out, or use a different account to view the rankings.
- Observe the differences in the sorting of the best-selling items based on these states.
For convenience, you can use this link to see the “past day” best-selling dynamic heads: Best Selling Dynamic Heads - Past Day. However, any other category and time filter can be used to test this issue.
Expected Result:
The best-selling rankings should be objective, solely based on actual sales data, and should not be influenced by any algorithmic personalization or user-specific data. The rankings should remain consistent regardless of whether I am logged in, logged out, or using a different account.
Actual Result:
The rankings for the best-selling items appear to be influenced by the algorithm’s interpretation of user preferences, not just sales data. This becomes apparent when comparing rankings while logged in versus logged out, or when viewing the list from different accounts. This behavior distorts the objective ranking system, making it difficult to track market trends accurately.
Screenshots:
Logged in:
Logged out:
(Note: The time difference between the screenshots is limited to just a few minutes, so the only independent variable here can be considered whether I am logged into my account or not.)