The new gatekeepers: searching for bias in Spotify's curated playlists
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
On streaming platforms, playlists have become the backbone of listening. A high ranking in a top playlist is often the difference between success and failure for a new song. McKenzie et al. (2021) find that songs that feature other artists perform better than songs that don’t feature other artists. At the same time Waldfogel et al. (2021) provide evidence Spotify’s playlist curators bias certain songs in their ranking decisions. In light of these two distinct findings, we ask if it’s possible that the ‘feature’ effect observed by Mckenzie et al. is driven by a bias in playlist rankings? We ask a similar question of the success of major label and local songs. To answer our research questions we focus on the popular New Music Friday playlists, which rank new music the day of release. We first conduct an ‘outcomes-based’ test to identify bias, finding that feature and major label songs receive lower ex-ante playlist ranks than their ultimate streaming outcome warrants across the US, Canadian, UK and Australian playlists. We also find curators of the non-US playlists rank local songs more than their streaming outcome warrants. Second, we build a weekly streaming model of top 200 chart songs for the same countries. Using artist-fixed effects, we provide evidence that the feature artist effect observed by McKenzie et al. exists beyond the US, and that its strong enough to offset the unfavourable playlist rankings. Our results show Spotify’s curators do not always rank songs strictly in terms of appeal, and speculate that they may be motivated to ‘level the playing field’ between certain groups of artist and support local content.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it