No single company does all of it. A Japanese adult video release passes through a maker that funds and owns it, a production side that shoots it, an agency that supplies and manages the performer, a review body that approves the master, and a retailer that sells it. Those are five separate businesses with five separate incentives, and most confusion about the industry comes from collapsing them into one.
The structure also explains why the catalogue is so hard to measure from outside — a point that matters directly for anyone trying to count titles, performers or output.
Who are the actual parties in a single release?
Six roles, at most, and often six different companies. Each holds a distinct decision right over the finished product.
| Role | Japanese term commonly used | What it decides | What it owns |
|---|---|---|---|
| Maker | メーカー | Whether a title is funded, what it costs, what code it gets | The master and the title's rights |
| Label | レーベル | Editorial identity, genre, art direction | Nothing separately — a brand of the maker |
| Production company | 制作会社 | Crew, location, shooting schedule | Usually nothing; work for hire |
| Talent agency | プロダクション / 事務所 | Which offers a performer takes, terms of appearance | The management contract |
| Review body | 審査団体 | Whether a master meets censorship standards | The approval that lets it ship |
| Retail platform | 配信・販売サイト | What gets listed, priced and promoted | The customer relationship |
Reading a product page, you generally see the label name most prominently, the maker somewhere in the metadata, and nothing at all about who actually shot it. The production company is the least visible party in a business whose output it physically creates.
What is the difference between a maker and a label?
A maker is a company; a label is a brand that company operates. One maker typically runs several labels, each aimed at a different genre or audience.
The distinction is easy to miss because labels are marketed as if they were independent studios — separate logos, separate art styles, separate release schedules. Structurally they are closer to imprints in book publishing. The capital, the contracts and the legal exposure sit with the maker.
This has a practical consequence for anyone cataloguing the industry: counting labels overstates the number of independent businesses considerably, and two "competing" labels are frequently the same company serving two segments.
We give no figure for how many makers are active, because none exists. Neither the trade associations nor the review bodies publish a member list or a member count, and no government statistic measures this population — the police registration series that gets quoted counts businesses licensed to transmit adult imagery, which is a different and much broader set than production companies. Any specific number you see for "how many JAV studios there are" is an estimate someone made, not a published count.
Checked 2026-08-03 against the trade and review bodies' own sites and Japanese government statistical publications.
Why does the product code tell you who owns the title?
Because the code is issued by the maker, and its letter prefix identifies the label that issued it. The code is the closest thing the industry has to a primary key.
A code splits into a prefix and a serial number. The prefix maps to a label, which maps to a maker; the number is that label's sequence. That is why the code, not the title, is what people search with — titles are marketing copy and change between regions and platforms, while the code is issued once and stays attached to the release everywhere it appears.
It is also why the code is a rights signal. If two listings carry the same code, they are the same underlying release regardless of which platform they appear on, what the thumbnail looks like, or what language the description is in.
What does an agency actually do for a performer?
It holds the management relationship: it fields offers from makers, negotiates terms, schedules work, and handles the promotional obligations that come with a release.
The agency sits between the performer and every maker, which is the single most important structural fact about how careers work here. A performer does not build a relationship with one studio the way a contract actor might in another industry. They build a relationship with an agency, and the agency's relationships with makers determine what work appears.
That intermediary position is also where the contractual questions of the last several years have concentrated — what an agency can commit a performer to, what notice and refusal rights exist, and how those are documented. Those rules have changed materially, and they are covered separately in consent and contracts.
What is the difference between an exclusive and a single-title performer?
An exclusive is contracted to one maker for a defined period and appears only in that maker's releases. A single-title performer is booked production by production and can appear across many makers in the same year.
The two tracks differ in almost every operational respect:
| Exclusive | Single-title | |
|---|---|---|
| Contracted to | One maker, for a term | The individual production |
| Release cadence | Regular, planned in advance | Irregular, demand-driven |
| Promotion | Maker-funded campaign, cover-billed | Minimal; often ensemble billing |
| Title count per year | Low, by design | Can be much higher |
| Career visibility | High — a named brand | Lower, and harder to track |
The asymmetry matters when reading any statistic about the industry. An exclusive with few releases may be far more prominent than a single-title performer with many, so title counts and visibility measure different things and correlate weakly.
Not everyone fits cleanly into either track. There is a middle arrangement where a performer is billed as a lead but is booked per production rather than held under an exclusive term, which is common enough that treating the market as a clean binary will misclassify a large share of it.
The industry's own vocabulary for these tiers is 単体 (tanpin, a release built around one billed performer), 企画 (kikaku, one built around a concept) and 企画単体 (kikaku tanpin, the middle case), with 専属 (senzoku) marking the exclusive contract separately. These are catalogue and trade terms rather than anything defined in a published standard; we set out how they are used in tanpin vs kikaku.
Where does the money actually sit?
With the maker and the retail platform. The maker carries production cost and owns the asset; the platform owns the customer, the pricing and the merchandising surface.
That leaves the production company on work-for-hire margins and the agency on a share of appearance fees. The economics push in a direction worth understanding: whoever owns the catalogue and whoever owns the storefront capture the recurring revenue, and everyone else is paid per job. Streaming intensified this, because a catalogue that keeps earning is worth far more than one that sold copies once — a shift traced in from VHS to streaming.
Why is this structure so hard to measure from outside?
Because no party publishes complete figures, and the layer most visible to outsiders — aggregator and streaming sites — reflects what was uploaded and promoted, not what was produced.
We run into this directly. Our index holds 241,792 faces, of which 2,333 are linked to a named performer across 106 sites (our index, 2026-08 snapshot). That looks like industry coverage until you look at where the images come from: 2,246 of those 2,333 representative images originate from a single site, and only a small minority of the 106 sites carry Japanese product codes at all.
So what an index like ours measures is exposure on aggregator platforms, not release volume by makers. Those two numbers can diverge sharply — a maker with a large catalogue that is poorly represented on aggregators will look small, and a title that circulates widely will look important regardless of its budget. Any claim about studio output built on aggregator data is measuring the wrong layer of the structure described above.