The Library Drives the Hours, Licensing Drives the Sign-Ups — Where Netflix's Viewing and Revenue Come Apart

Two-thirds of Netflix's 97 billion hours came from old titles

Netflix members watched more than 97 billion hours in the first half of 2026, up 2% year over year — faster than the 1.5% growth in the same period of 2025, in a half-year that contained both the Winter Olympics and the FIFA World Cup. But close to two-thirds of those hours came from titles that were not new.

Parrot Analytics, in analysis published July 27, puts the library share of Netflix viewing at 63% in Q2 2026, up 12 percentage points from 51% in Q1 2023, while season debuts fell from 12% to 8% and returning seasons from 22% to 18%. Subscriber data the same firm published through TheWrap on July 31 points the same way: licensed series have accounted for more than half of all series-driven global sign-ups in every quarter since Q1 2025, clearing 53% in Q1 2026. On both viewing and acquisition, the axis holding Netflix up has moved from new releases to the catalog.

The shift coincides with a business model moving from a subscriber-count contest to a revenue-structure contest. Netflix stopped reporting quarterly subscriber numbers in 2025, and in its Q2 2026 results said it will publish the semi-annual "What We Watched" report only once a year from 2027. The frequency with which the company shows the market its aggregate numbers has itself been cut.

Co-CEO Greg Peters told the earnings call there is no linear relationship between view hours and revenue or profit, because not all hours are created equal. An hour that produces a sign-up, an hour that prevents a cancellation and an hour an advertiser pays a premium for are worth entirely different amounts. With the scale-building phase over, what matters is not the volume of viewing time but its unit price. Investment priorities follow: the formats with the highest value per hour — the ad tier, live events and games.

Live programming makes the point most sharply. Live will take just over 5% of the 2026 content budget while generating about 1% of view hours. On aggregate measures alone it is a losing proposition. Yet live events produced six of the ten biggest single days for new sign-ups over the past five years. Value exists that viewing hours do not describe.

Advertising works the same way. Ad revenue is set to double from $1.5 billion in 2025 to $3 billion in 2026, and ad pricing depends on who is watching what and when, not on aggregate hours. Netflix has already entered the range where the total-hours metric stops explaining the business.

The bill for that change passes straight through to content suppliers. "Our title was watched for X hours" no longer defends a price. Without evidence of which subscribers those hours brought in and which ones they kept, even large viewing numbers are hard to convert into value at the negotiating table.

The finding that licensed titles produce more than half of all series-driven sign-ups is both an opportunity and a condition for suppliers. The basis of leverage is moving from the buzz around a single new title to the depth of a multi-season library and its coverage of specific markets. Korean broadcasters and producers face the same requirement: restructure negotiations around multi-season packages and demonstrated market-by-market acquisition and retention contribution rather than one-off supply deals.

The World Cup pushed down viewing and sign-ups at the same time

Parrot Analytics' weekly-adjusted data shows sharp drops across football-focused markets after the June 11 kickoff. June-versus-May declines ran to -50% in Uruguay, -49% in Argentina, Croatia and Norway, and -47% in Paraguay, Belgium, Germany and Portugal. Year-over-year comparisons were steeper still in places: -69% in Paraguay, -51% in Argentina, -47% in Türkiye.

The decline was not confined to existing members reallocating their time. Parrot Analytics' net-additions model showed expectations for new-member additions weakening late in the quarter, particularly across Latin America and EMEA. The tournament did not merely redirect viewing; it reduced the urgency for prospective subscribers to sign up. Netflix's own disclosure that U.S./Canada view hours declined in the quarter sits in the same picture.

South Korea ranks among the top markets by weekly average hours, behind the U.S., Brazil and Mexico, and the chart shows its June figure falling roughly 40-plus percent from May — a decline in line with the global pattern.

June 2026 engagement change in markets with a World Cup team. Left: weekly average hours in top markets (June 2025, May 2026, June 2026). Right: markets ranked by decline. Source: Parrot Analytics.

The library generates 25 billion hours a quarter

A few definitions govern how Netflix viewing is split. A season debut is the first season of a new series arriving for the first time; a returning season is a new season of an existing series. Pay 1 is the window where a film lands on streaming after its theatrical run ends — the first paid streaming slot after release, and the most expensive. Pay 2 and 3 are the second and third rounds after that window closes: cheaper, and able to move across multiple platforms. Library covers everything old enough that it is no longer treated as new. Originals and acquisitions alike end up here with time.

Library titles now attract more than 25 billion hours of engagement per quarter, close to two-thirds of total global hours. Category share changes from Q1 2023 to Q2 2026 run +11.8pp for library and +0.8pp for pay 2&3, against -4.6pp for season debuts, -4.5pp for returning seasons and -3.5pp for pay 1. Viewing that new releases used to generate has moved to older titles. Engagement with both Netflix-owned library titles and acquired licensed library titles grew by roughly 50% over the period.

The same dataset carries a countervailing signal. Individual library titles are losing engagement faster than comparable titles did in earlier years. As the period a single title holds up — its half-life — shortens, sustaining aggregate library viewing requires continually adding more titles. The total looks large not because each title is strong but because the shelf keeps being restocked. Parrot Analytics reads Netflix's growing use of licensed local titles, which can cover particular markets efficiently, as the offset.

Category share and total volume of Netflix hours, Q1 2023 – Q2 2026. The library share rose from 51% to 63%. Source: Parrot Analytics.

A deep library reduces dependence on an unbroken succession of global blockbusters and supports retention. But if per-title contribution keeps falling, holding the total requires rising licensing spend, greater catalog depth and continued investment in discovery. The moment costs grow faster than hours, the strategy inverts. The measure that matters is not how many hours the library produces, but how much incremental retention value those hours create relative to the cost of acquiring, producing and promoting the content.

For Korean suppliers, the relevant reading is leverage. If acquired titles produce more than half of new sign-ups, the strength of Korean broadcasters and producers lies less in the rights to a single new title than in the depth of a multi-season library. Leading with multi-season packages and market-coverage value beats a one-off supply deal.

40% of the catalog, 50% of the sign-ups

The asset class carrying the viewing load moves the same way on acquisition. Licensed series account for roughly 40% of the Netflix global catalog by title count — and more than 50% of series-driven sign-ups come from them. The yield per title is disproportionate. Christofer Hamilton, industry insights manager at Parrot Analytics, finds that multi-season dramas and repeat-viewed comedy libraries convert casual viewers into paying subscribers at a higher rate per title than the average original production.

The trajectory of that share shows a structure that broke and then reassembled. The licensed share fell from 73% in Q1 2020 to 44.7% in Q4 2022, the peak of the originals spending cycle. It climbed back above 50% within five quarters and has not dropped below that line since the start of 2025. The center of gravity that swung toward originals has swung back.

Title-level results show how the two asset classes divide the work. The single biggest acquisition driver in Q1 2026 was an original: "Bridgerton," which released its fourth season in two parts, pulled in more than 360,000 new sign-ups during the quarter. Behind it, two licensed properties of entirely different character finished within striking distance of each other — "One Piece," the original anime rather than the live-action remake, at 178,000, edging out "WWE Monday Night Raw" at 173,000.

"One Piece" draws its strength from a back catalog that does not exhaust itself after sign-up, reinforced by a halo effect that sends fans of the live-action version back into the source anime. Netflix has said anime reaches more than half of its global audience. "Raw" contributes past the point of acquisition: weekly appointment viewing, meshed with the archival WWE library, converts into a retention instrument. Parrot Analytics estimates that in 2025, the first year of the 10-year partnership, WWE content on Netflix prevented roughly 1.25 million subscribers globally from churning each quarter. Netflix secured "Raw" in January 2024 at $500 million per year across a decade, with programming starting January 2025.

Share of Netflix global subscriber acquisitions attributed to licensed vs. original series, Q1 2020 – Q1 2026. Source: Parrot Analytics DEMAND360, Content Valuation iteration 07-2026.

Programming plans point the same way. Netflix has signaled it will expand licensed programming while spending a $20 billion content budget in 2026 — an estimate 10% above last year. It picked up roughly 20 titles from Paramount Skydance, including "Matlock" and "The King of Queens," and widened its pay-1 arrangement with Sony Pictures Entertainment into a global deal.

On July 31 it closed a five-year, $500 million co-exclusive agreement with AMC Global Media covering the "Walking Dead" universe — seven series, 371 episodes — streaming simultaneously on Netflix and AMC+ from 2027. Netflix's sole U.S. streaming position, held since 2011, ends there. In exchange, the rights widen across the U.K., Italy, Australia and New Zealand. A company that structures almost all of its content deals for exclusivity gave up that exclusivity to keep the library — a transaction that shows how high securing library rights now ranks.

Non-English at 30%, animation at 18% — and the audience mix moved too

The content mix shows the same movement. Animation rose from about 14% of Netflix viewing in early 2023 to 18% in Q2 2026. Non-English programming grew from 27% to 30% over the same period. Netflix's own disclosure for the first half of 2026 reports non-English content driving more than a third of all viewing, with titles from Korea, Japan, Spain and India performing.

The audience-segment data is more specific. Among titles with a measurable skew, the share of engagement from titles skewing to younger males grew fastest, from 21% to 26%. Younger females fell 6 percentage points, from 43% to 37%. Older females rose from 14% to 17%; older males slipped from 22% to 20%. The pattern reads as a function of animation's unusually high index with younger male audiences. Netflix has said more than half its members watch anime.

Share of Netflix hours by language, animation and audience segment, Q1 2023 – Q2 2026. Source: Parrot Analytics.

The two categories do different work. Non-English programming raises the service's fit in local markets while leaving room for a global breakout. Animation produces durable, repeatable viewing and holds younger audiences. Neither is valued on total hours alone. Audience expansion and retention, franchise potential and acquisition cost all have to be in the calculation before a number comes out.

No obvious "Squid Game" in the second half

Parrot Analytics' analysis of Netflix's announced returning seasons finds the slate itself competitive, but no returning title with the anticipated global impact of *Squid Game*, *Wednesday* or *Stranger Things*. The issue is not title count or quality — it is that the event window in which one title dominates the global conversation is not on this year's second-half calendar.

The gap shows in the numbers. On first-90-day views, *Squid Game* cleared roughly 170 million in 2H2024; in 2H2025, *Stranger Things* reached about 360 million and *Wednesday* about 137 million. For 2H2026, where each season is proxied by its prior installment, the top three — *Lupin* S4, *Emily in Paris* S6 and *Avatar: The Last Airbender* S2 — all cluster below 70 million. The distinguishing feature is not that the top of the slate is low but that there is almost no spread within it. The single title that carried each of the past two half-years is gone, replaced by a row of comparable weights.

Franchise cycles compounded it. *Stranger Things* ended with its fifth season on December 31, 2025, and *Squid Game* closed with season three in 2025. The year-end sign-up surge those two produced is absent this year. For Netflix this is the first second half with two tentpoles expiring at once, and the analysis finds no successor franchise yet visible in the data.

First-90-day views per season by half-year slate. The 2026 second-half slate is proxied by prior installments. Source: Parrot Analytics.

The analysis covers returning seasons only. The potential of new series and films is not in the calculation. Netflix has outlined a second-half offering spanning scripted series, films, international productions and live programming, and unexpected hits have come from season debuts more than once in recent years. The estimates also carry the limitation of being prior-season results transposed forward.

What matters is substitutability: whether the acquisition lift and cultural momentum a single tentpole produces visibly can be replaced by the sum of several titles. The rise of the library and licensing shown in the earlier data is itself Netflix's answer — a deep catalog, live programming and repeat-viewed licensed series holding up baseline viewing and retention, damping the amplitude of a tentpole gap. Whether that design works will show in how quickly subscriber growth recovers once the World Cup effect clears. The second half is the first test of whether Netflix has built a structure that holds without a hit.

What this means for Korean rights holders

The negotiating position of back catalogs has changed. The asset class generating 63% of all viewing and more than half of all sign-ups includes the multi-season libraries held by Korean broadcasters and production companies. Until now, negotiations with global platforms have centered on new titles, with catalog rights attached as a secondary term. If shortening title half-lives force platforms to keep expanding volume, catalog packages become an independent line item rather than a dependent one. What determines price is whether a supplier can produce quarterly viewing- and retention-contribution figures at the negotiating table.

Demand for local licensing is rising. Parrot Analytics reads Netflix's expansion of locally licensed titles as its offset against shorter content half-lives. Korea is simultaneously a top-tier market by weekly hours and a supplier of non-English content. Mid-scale titles with limited global breakout potential can still clear the licensing bar on in-market retention contribution alone. Ampere Analysis reports Korean content accounts for 8–9% of total Netflix viewing hours, second only to U.S. content.

Co-exclusivity is now a live deal structure. AMC retained full IP ownership and its own platform rights while collecting $500 million. For Korean companies running their own OTT services while supplying global platforms, that is a workable template: rights can be split by territory, term and platform rather than transferred wholesale.

Animation and younger male audiences are the weak link in the Korean slate. While animation climbed to 18% of viewing and the younger-male segment grew fastest, Korean titles performing on Netflix have concentrated in live-action romance and thriller. Webtoon-to-animation adaptations, action and fantasy genres, and game-IP tie-ins are where the data shows an opening. *KPop Demon Hunters* setting an animated-film viewing record stands as evidence that the animation route for Korean IP works.

Pairing live with archive is what raises the price. The lesson of the WWE deal is not the weekly broadcast itself but the path it opens into the archival library. For rights holders with K-pop concert footage, live variety formats or sports events, bundling archive packages into long-term agreements — rather than selling one-off broadcast rights — is the route. The same arithmetic applies to domestic OTT operators: major events depress viewing and sign-ups at the same time, and the library is what cushions churn during those windows.

The space left by "Squid Game" is open. The absence of a global tentpole in the second half of 2026 is an entry window for Korean producers. In a half-year where Netflix must compete on breadth rather than a single event title, slate slots and marketing allocation for mid-scale productions expand. The same logic sharpens the performance bar: negotiation materials now need acquisition and churn-prevention contribution alongside 90-day view counts.

Sources

• Parrot Analytics, "Netflix is Growing Engagement But What Generates Economic Value," July 27, 2026 (source data for the library, World Cup, language/animation and second-half slate charts)

https://www.parrotanalytics.com/insights/netflix-is-growing-engagement-but-what-generates-economic-value/

• Christofer Hamilton, "Licensed Shows Now Drive Over Half of Netflix's New Subscribers | Chart," TheWrap (WrapPRO), July 31, 2026 (source data for the licensed vs. originals acquisition chart)
• Netflix, Q2 2026 Shareholder Letter, July 16, 2026 — 97 billion hours in H1 2026, up 2%; non-English content above one-third of viewing; "What We Watched" moving to annual publication
• Netflix, "What We Watched: The First Half of 2026," about.netflix.com, July 16, 2026
• CNBC, "Netflix (NFLX) earnings Q2 2026," July 16, 2026 — Greg Peters on hours not being created equal
• Variety, "Netflix Q2 Earnings Results In-Line With Expectations, Stock Drops on Lower Q3 Revenue Outlook," July 16, 2026 — live programming at 5% of spend and 1% of hours; six of the top ten sign-up days; $3 billion ad revenue target
• The Hollywood Reporter, "Netflix Q2 2026 Earnings Review: Stock Analyst Reactions," July 17, 2026 — U.S./Canada view-hour decline
• Variety, "Netflix Tops 325 Million Subscribers, Plans to Boost Content Spending 10% to $20 Billion in 2026," January 20, 2026 — Paramount Skydance titles; Sony pay-1 expansion
• Variety, "'The Walking Dead' Franchise to Stream on Both Netflix and AMC+ Beginning in 2027 Under New $500 Million Deal," July 30, 2026
• Variety, "Netflix, WWE Strike Deal to Move 'Monday Night Raw' to Streamer Beginning in 2025 for $500 Million per Year," January 23, 2024
• Deadline, "Netflix 2026 TV Slate: Bridgerton, Beef, Hunting Wives, More," January 8, 2026 — 2026 lineup after *Stranger Things* and *Squid Game* concluded
• Ampere Analysis via Variety, "'Squid Game 2' Leads Korean Content Dominance on Netflix Global Charts" — Korean content share of Netflix viewing