Two scoring changes shipped this week. The first changes where the originality multiplier is applied so the penalty for licensed/aggregator content stops cascading through unrelated criteria. The second adds a content-model classifier so photo-archive sites stop being scored as if they were video-only.
v4.8, Originality moved to the final score
Every site is classified into one of five originality tiers based on how much of its catalog is its own production:
| Tier | Multiplier | Examples |
|---|---|---|
| Exclusive | 1.00 | Brazzers, BangBros core, Czech VR |
| Mostly exclusive | 0.90 | Adult Time (Pure Taboo, Burning Angel, Girlsway) |
| Mixed | 0.70 | FapHouse Originals + marketplace, real-vr |
| Mostly licensed | 0.50 | POVR, VRPorn, mr-skin, xnxx-gold |
| Aggregator | 0.30 | VOD storefronts without own production |
Previously the multiplier was applied directly to video_count before any criterion was evaluated. That created a cascade: a 0.5 multiplier shrunk the scene count, which dragged down content volume tier, which then propagated into bonus content (formula depends on scene count) and value-for-money (price-per-scene). A licensed-heavy site could end up two tiers below where its actual quality placed it.
The fix is straightforward: apply the multiplier once, at the end, to the composite of content volume, bonus content, and value-for-money. The multiplier still expresses the same thing, aggregators get less credit for catalog they didn’t produce, but it stops leaking into unrelated places.

Reclassification of large catalogs
Alongside the formula change, a few large catalogs moved between tiers. The headline ones:
- Adult Time moved from mixed (0.70) to mostly exclusive (0.90). The bulk of the AT catalog is produced by their own studios, Pure Taboo, Burning Angel, Girlsway, Adult Time Originals. The licensed portion is a minority slice, not the spine of the offering.
- POVR, VRPorn, and Mr. Skin moved to mostly licensed (0.50). These are VOD storefronts and curated clip catalogs, not studios.
- Brazzers and BangBros stay at exclusive (1.00), their catalogs are 100% own production.
We also fixed a long-standing data issue with bonus_sites_count: Brazzers had 3 in the database when the actual count is 26; BangBros had 13 when it’s 55. That was suppressing their bonus-content scores for years.
v4.9, Photo-archive sites get a proper classifier
Adult sites come in two broad shapes, video-primary (a paysite or VOD platform) and photo-primary (an archive of photo sets, sometimes with bonus videos). Until v4.9 our scoring used a 10% threshold to detect photo-primary sites: if more than 10% of the catalog was video, the site was classified as video-primary and scored against video tiers.
The threshold was too generous. A photo archive with a small bonus video section (5,000 photo sets, 200 videos) would tip into video-primary and then take a hit on content volume because 200 videos is “low” on the video scale, even though the actual catalog is huge by photo standards.
The new classifier:
- Photo share below 20% → video-primary
- Or video count is at least 5x photo count → video-primary
- Otherwise → photo-primary, scored against PHOTO_TIERS
The classification is stored as offer_facts.content_model and can be overridden manually for edge cases (sites that change focus, unusual archives).

What this means in the rankings
Several legendary catalogs moved up in our top sites list and within their category pages. VOD storefronts and clip aggregators dropped a tier, not because they got worse, but because the scoring model now reflects who actually produced the content. Photo-heavy sites are no longer being measured against the wrong yardstick.
For the full breakdown of how the seven scoring criteria are weighted, see How We Score.