My question is about the algorithms for deducing the "related artists". I always thought that related artists (to any artist) are chosen according to the listening habits of the users. So if -say- 1000 users listen to Artist A and Artist B, both artistst are related.
But... now compare my stats in Spotify and on Fans Insights. There is NO single name repetition. How do you explain this? (Picture a and b)
Besides, anecdotically, the "Who-else-do-they-listen-to" is quite static, didn't change since months. The related artists fluctuate quite a lot (at least in my case). overall, it seems to be more accurate.
Let me know if you have an answer to this. Even if it is not an easy answer. Thanks