1. Data sources
The Crown Country chart aggregates signals from six independent music platforms. Each one is pulled daily through its official API (or RSS feed in Apple's case) and stored as a weekly snapshot in our database.
| Source | Signal | Weight |
|---|---|---|
| Spotify (via Songstats) | Monthly listeners, stream velocity | 35 % |
| Apple Music / iTunes | RSS chart position (genre 6, multi-country) | 25 % |
| YouTube Music | Search volume, official video views | 20 % |
| Deezer | Fan count, chart rank | 8 % |
| Shazam | Discovery rate (tags / day) | 7 % |
| iTunes Sales | Paid downloads | 5 % |
Weights are tuned to reflect each platform's relative reach in the country genre. Spotify carries the largest weight because it represents the majority of paid streaming for country music globally; iTunes sales are kept as a small but stable purchase signal.
2. Normalization (Z-score)
Raw numbers can't be compared directly — a Spotify monthly-listener count is in millions, a Shazam tag count is in thousands. We normalize each weekly signal against the country genre baseline using a z-score:
z = (value − mean_country_genre) / std_country_genre
This gives every signal a comparable distribution: 0 means "average country track this week", +1 means "one standard deviation above average", and so on. The same formula applies to all six platforms.
3. Raw Listening Index (RLI)
Once every platform is normalized, we combine them into one weekly score using a geometric mean — chosen because it punishes one-platform hits and rewards tracks that perform consistently everywhere:
RLI = (∏ (sigmoid(z_i) ^ w_i)) ^ (1 / ∑w_i) × 100
Where z_i is the z-score of platform i, w_i is its weight from the table above, and sigmoid() maps z-scores to a [0, 1] range so the product stays well-behaved. The result is a single 0–100 score that determines a track's position on the Crown Country Top 100.
4. Crown+ — Holt-Winters forecasting
The RLI captures where a track stands now. Crown+ captures where it's going. We fit a Holt-Winters damped-trend model on each track's last 12 weeks of RLI history, then project 4 and 8 weeks forward:
- level = current value (smoothed)
- trend = weekly velocity (smoothed, damping factor ϕ = 0.95 to avoid runaway forecasts)
- confidence = inverse of forecast residual variance, clamped to [0, 1]
Each track receives a label based on its forecast trajectory:
5. Local country charts
For each major country music market, we publish a separate chart drawn from that market's Apple Music RSS feed (genre 6 = Country). This gives a real-time snapshot of what people in that country are actually streaming, distinct from our global RLI.
Currently published local charts:
6. Refresh cadence
- Streaming signals
- Daily, around 04:00 UTC
- RLI recompute
- Weekly, Thursday 20:00 ET
- Crown+ forecast
- Weekly, right after RLI
- Local Apple charts
- Every 4 hours (cached at the CDN)
- Apple Awards
- Thursday after the RLI recompute
7. Editorial integrity
The Crown Country chart is computed exclusively from streaming and discovery data — there is no editorial weighting, no label-paid promotion, and no human override. Artists and labels cannot pay to be ranked.
Affiliate links to Apple Music and Amazon Music appear next to chart entries to help readers listen; they are clearly marked with rel="sponsored" per Google's guidelines and do not influence the ranking in any way.
8. Data access for journalists & researchers
Crown Country publishes a weekly CSV of the Top 100 (RLI scores, ranks, movements) free of charge for editorial use. Crown Country Pro subscribers receive the full dataset (200+ artists, 12-month history, country-level splits) plus Crown+ forecasts and pre-generated PDF briefs.