Business Overview

Netflix, Inc. (NASDAQ: NFLX)

9 September 2026

Evidence base: Netflix SEC filings FY2016–FY2025 (10-Ks), the Q2 2026 Form 10-Q and quarterly

shareholder letters through 16 July 2026, and the December 2025 – January 2026 merger 8-Ks, held in the

research folder; supplemented by independent industry sources — Nielsen, Ofcom, IAB, eMarketer, the WGA

and SAG-AFTRA 2026 agreements, and the filings of Disney, Warner Bros. Discovery, Paramount, Comcast, Alphabet, Amazon and Roku.

This is an investor-grade business overview. It is not a valuation and not a recommendation.

1. Executive Snapshot

2. What the Company Does

The customer problem Netflix solves is choice under a fixed budget of attention. A household wants a large, constantly refreshed supply of professionally produced video, available on any screen, without a contract, an installation visit, or a schedule. Netflix sells that as a monthly subscription in more than a hundred countries, at prices ranging from the US dollar equivalent of $1 to $37 per month depending on market and plan (FY2025 10-K). It describes its own strategy as growing “globally within the parameters of our operating margin target”.

The unit of economics is one paying membership-month. Every part of the model — pricing, content amortisation, advertising, the extra-member fee — resolves into what that unit earns and what it costs. Netflix stopped disclosing the number of those units during 2025, stating that it had “discontinued the reporting of membership numbers, including average paying memberships and average monthly revenue per paying membership, focusing instead on revenue and operating margin” (FY2025 10-K). The last figure it published was 325m paid memberships, crossed during Q4 2025 (Q4 2025 shareholder letter, 20 January 2026), and the last average revenue per membership it published was $11.70 per month for FY2024.

Tracing one unit

Follow a single membership-month through FY2025. It brings in roughly $11.59 of revenue — collected in advance, which is why deferred revenue sits on the balance sheet at $1.8bn and why working capital is a source of cash rather than a use. Against that, cost of revenues absorbs about $5.97, of which content amortisation alone is $4.21 and the remainder is streaming delivery, payment processing, studio operations and, in FY2025, a $619m Brazilian non-income tax charge. Marketing takes about $0.85, technology and development $0.87 and general and administrative $0.48, leaving about $3.42 of operating profit.

Per-unit figures are derived by dividing FY2025 income statement lines by the 325m year-end membership base and twelve months. They are illustrative: the average base during the year was lower than the year-end base, so true per-member revenue and profit were somewhat higher. Netflix no longer publishes the denominator.

The number that does not appear in that trace is the cash. Netflix paid out $17.1bn for content in FY2025 while charging $16.4bn to the income statement — about $4.38 of cash against $4.21 of expense per member-month. The gap is small today, but it is the mechanism that separates reported margin from cash generation, and management guides it wider in 2026 at a cash-to-amortisation ratio of roughly 1.1x.

What it sells, and what it does not

Netflix reports as one operating segment and earns revenue from three sources of materially different size. Membership fees are the business. Advertising, launched in late 2022, produced over $1.5bn in 2025 — more than 2.5 times 2024 — and is guided to roughly $3bn in 2026, which would be about 6% of revenue. Extra-member sub accounts, priced at $2 to $9 a month, monetise the household-sharing behaviour Netflix began enforcing against in 2023; they are deliberately excluded from the paid-membership count, which is one reason the historic membership series understates paying relationships from 2022 onward.

Games, live events and video podcasts are not revenue lines. Netflix discloses no games revenue, no games spend and no live-rights commitments, and management is explicit that live programming will be “just over 5% of our content spend but only ~1% of view hours” in 2026 (Q2 2026 shareholder letter). They exist to raise retention and acquisition on the subscription unit, not to be sold. Live events “accounted for six of the top 10 new member sign-up days over the last five years,” which is the case for them stated in the company's own terms.

The one product line Netflix has fully exited is DVD-by-mail, which stopped shipping on 29 September 2023. Its economics were the opposite of streaming: 58% contribution margin in its final reported year (FY2018) on a base that shrank every year, because each incremental disc carried real postage and handling. The exit is a useful contrast — Netflix gave up a high-margin business precisely because its costs were variable and its unit count was falling, while the streaming model rewards exactly the reverse.

3. Industry, Competitive Position and Moat

Streaming sells three things through one profit and loss account: subscription access to a library, attention resold to advertisers, and wholesale rights licensed to others. Value in the chain is created at four stages — production, studio and rights ownership, the consumer-facing platform, and the device or operating system that carries it — and it is captured at two. The evidence is in competitors' own reporting for the June 2026 quarter: Warner Bros. Discovery's Studios segment earned $96m of adjusted EBITDA on $2,328m of revenue, a 4.1% margin, and Paramount's Studios earned $36m on $1,314m, 2.7%. Roku's platform business, which owns no content at all, earned a 53.0% gross margin.

The aggregator layer is where the profit sits, and it is extraordinarily concentrated. Across the five platforms that disclose streaming profitability, the June 2026 quarter produced roughly $6.0bn of streaming operating profit in total — Netflix $4.19bn, Disney's Entertainment SVOD $712m, WBD Streaming $512m, Paramount's direct-to-consumer $366m and Peacock $189m. Netflix earned about 70% of it. The comparison flatters Netflix slightly, because three of those four rivals report EBITDA rather than operating income, but the order of magnitude is not in dispute.

What actually protects Netflix

The moat is not the library, and the last twelve months proved it. Warner Bros. Discovery owns HBO, DC and Harry Potter, and could not clear a 16.6% streaming EBITDA margin; it was auctioned and sold. Nor is it technology, which Amazon, Roku and Samsung all sell as a commodity. What protects Netflix is the amortisation base: a single global content pool spread across the largest paying audience in subscription video, with more than a third of all viewing coming from non-English titles (Q2 2026 shareholder letter). A title paid for once is monetised in every market simultaneously.

The consequence is visible in relative spending. Netflix spent $17.1bn on content in FY2025 and guides roughly 10% growth in 2026. Amazon disclosed $22.4bn of video and music content costs for 2025, Disney has guided to about $24bn for FY2026 split roughly evenly between sport and entertainment, and industry tallies put NBCUniversal near $37bn and YouTube near $32bn. Netflix spends less than any of them and earns more than all of them combined on the streaming line. That is the whole competitive case, and it is arithmetic rather than assertion.

Two barriers genuinely bind for a new entrant and one does not. The absolute cash outlay binds: roughly $20bn a year of content spend cannot be assembled by a subscale player, and the WGA's 2026 agreement prices a three-year one-hour residual at $89,370 for Netflix, Amazon and Disney+ against $36,478 for HBO Max and Peacock — scale is contractually taxed, not subsidised. Distribution gatekeeping binds: Roku alone reaches more than half of US broadband households, and Fox agreed in June 2026 to buy it for about $22bn, which is a content owner paying to own the toll booth. Brand and back-catalogue do not bind, as WBD demonstrated.

Testing the company's account of itself

The industry rewards two kinds of company: the single platform with global scale and one amortisation base, and the toll collector that never buys a script. Netflix is unambiguously the first — it earns most of the sector's profit, spends less than its rivals to do so, and in February 2026 declined to raise a signed $82.7bn bid for Warner Bros. rather than overpay. It is not the second. It owns no operating system, no device layer and no advertising infrastructure of the scale that Amazon and Alphabet operate, and the largest single consumer of American television time is a platform that pays for content out of a revenue share rather than a balance sheet.

4. Growth Engine

Netflix's revenue grew 16% in FY2025 to $45.2bn and 15% in the first half of 2026. Management attributes it, in the same order each period, to “growth in memberships, price increases, and increased advertising revenue, partially offset by unfavorable changes in foreign exchange rates” (FY2025 10-K). Netflix does not publish a quantified split among those three, and since it also stopped publishing memberships and average revenue per membership, the volume-versus-price decomposition can no longer be reconstructed from the filings. That is the single largest analytical gap in this business today.

What can be decomposed is currency, and it matters more than it looks. Fifty-six percent of FY2025 revenue was denominated in currencies other than the US dollar against only 31% of operating expenses — a structural mismatch that makes reported growth swing on rates. In FY2025 currency was a headwind worth $271m; in the first half of 2026 it flipped to a $715m tailwind, which is the entire difference between 15% reported growth and 13% on a constant-currency basis.

Reported growth against constant currency

Source: FY2025 10-K MD&A and Q2 2026 10-Q MD&A. Constant currency excludes the effect of rate movements and of hedging gains and losses realised in revenue. LATAM's FY2025 divergence — 11% reported against 23% constant currency — is the Argentine peso.

Acquisitions are immaterial to growth, which makes the reported-versus-organic question unusually simple here. Netflix bought nothing in FY2023 or FY2024, spent $17m in FY2025 and $587m in March 2026 on InterPositive, a generative-AI production toolmaker. All of the growth above is organic. The reported-versus-underlying distinction that does matter sits in earnings rather than revenue, and it is large: first-half 2026 net income of $8.68bn and diluted EPS of $2.03 include a $2.8bn pre-tax termination fee received when Warner Bros. Discovery walked to Paramount Skydance, recorded below the operating line in the first quarter.

The growth levers, ranked

• Price. Structural. The most reliable lever and the one with the clearest evidence. Netflix raised US prices in January 2025 and again in March 2026, and raised prices in Mexico and Spain in the first half of 2026, with management reporting “the impact consistent with prior price changes and our expectations.” Every major US service raised price by 8% to 18% in the twelve months to September 2026, so the pricing environment is industry-wide rather than Netflix-specific — which makes it more durable, not less.
• Membership growth outside UCAN. Structural, decaying. EMEA, LATAM and APAC together grew 16% in FY2025 against UCAN's 15%, and each passed a quarterly revenue milestone in Q2 2026. But UK household SVOD penetration reached only 70% in 2026, up two points in four years (Ofcom, Media Nations 2026), which is what saturation looks like in a mature market. Unit growth increasingly has to come from lower-ARPU geographies.
• Advertising. Management-driven. Revenue rose more than 2.5 times in 2025 to over $1.5bn and is guided to roughly double again to about $3bn in 2026, on an ad tier Netflix says reaches more than 250m monthly active users. The mechanism is real — the ad tier converts free-tier-priced demand into paid units and monetises them twice. The constraint is price: connected-TV inventory is being flooded simultaneously by every platform, and Disney's most recent print shows rates falling 4% while impressions rise 8%.
• Extra-member accounts and plan mix. Management-driven. Priced at $2 to $9 a month and excluded from the membership count, these convert previously unpaid household-sharing into revenue without acquiring a new customer. Netflix has never quantified the effect, and no competitor does either, although HBO Max and Disney+ both copied the mechanic.
• Foreign exchange. Cyclical, and currently a tailwind. Worth about two percentage points of reported first-half 2026 growth. It reverses.

5. Margin, Cash and Capital Allocation

The margin mechanism is that content cost is fixed and the audience is not. A title's cost is capitalised and amortised over the shorter of its availability window, its estimated period of use, or ten years, on an accelerated basis, with over 90% of any asset expected to be amortised within four years (FY2025 10-K). Once that schedule is set it does not move with subscriber count. So every additional member-month, every price increase and every advertising dollar drops against a charge that was already fixed, and operating margin rose from 20.6% in FY2023 to 26.7% in FY2024 to 29.5% in FY2025, with 31.5% guided for 2026.

The same mechanism runs in reverse and Netflix says so plainly in its own risk factors: “given, in particular, that our content costs are largely fixed in nature, we may not be able to adjust our expenditures or increase our revenues… commensurate with the lowered growth rate.” It goes further on liquidity — payment terms on content commitments “are not tied to member usage or the size of our membership base,” and the company “may be unable to react to any reduction in our cash flows from operations… by reducing our streaming content obligations in the near-term.” Total content obligations stood at $25.1bn at 30 June 2026, of which $19.6bn is off balance sheet, with $11.9bn falling due within a year.

The financial spine

Sources: FY2020, FY2022, FY2024 and FY2025 10-Ks and the Q4 2025 shareholder letter. Comparability warnings: (i) a ten-for-one forward stock split took effect on 14 November 2025 and all per-share and share-count figures above are restated to that basis; (ii) the free cash flow definition changed in FY2020, and the FY2024 and FY2025 figures are company-reported non-GAAP from the shareholder letters because the 10-K stopped reconciling the measure after FY2023; (iii) FY2019 predates both the DVD wind-down and the advertising tier; (iv) membership and average-revenue metrics were discontinued during 2025, so no unit column can be carried across this table.

Cash conversion is now better than earnings, which was not true for most of the company's history. FY2019 burned $3.3bn of free cash flow; FY2025 generated $9.5bn. The turn came from two places. Content cash spend flattened — it is up only 1.5% in total across the six years from FY2019 to FY2025 while revenue more than doubled — and from FY2023 the amortisation charge began running ahead of cash payments, which converts a working-capital drain into a source. Capital intensity outside content remains trivial: property and equipment purchases were $688m in FY2025, 1.5% of revenue.

Where the cash went

Across FY2023 to FY2025 Netflix deployed cash in a consistent order. Content came first at $45.9bn of cash payments. Share repurchases came second at $21.4bn, rising each year to $9.1bn in FY2025 and continuing at $6.0bn in the first half of 2026, including a record $4.7bn quarter in Q2 2026 executed at declining average prices from $97.47 in April to $79.08 in June. Property and equipment took $1.5bn, net debt repayment $0.4bn and acquisitions $17m. The company has never paid a dividend and states that it does not anticipate one.

That ranking describes a management team that treats content as capital expenditure and treats the equity as the residual claimant on whatever is left. The buyback is not a token: diluted share count has fallen from 4,495m in FY2023 to 4,261m in the second quarter of 2026, and the pace accelerated to a 2.0% year-on-year reduction in that quarter. The Board authorised a further $25bn in April 2026 on top of $6.8bn remaining, leaving $27.1bn of capacity against net debt of only $5.2bn.

The Warner Bros. Discovery transaction, which post-dates the annual report

On 4 December 2025 Netflix signed a merger agreement to acquire Warner Bros. Discovery's streaming and studios businesses — HBO, HBO Max and the film and television studios — for $27.75 per share, an equity value of about $72.0bn and an enterprise value of about $82.7bn, with WBD's linear networks to be spun off first. It amended the deal to all cash on 19 January 2026 and assembled $67.2bn of committed facilities: a $42.2bn bridge, a $20bn delayed-draw term loan and a $5bn revolver. Receipt of financing was not a condition to closing, and Netflix agreed a $5.8bn reverse termination fee payable if antitrust approval failed.

On 27 February 2026 WBD terminated the agreement to accept a superior all-cash offer of $31.00 per share from Paramount Skydance, and paid Netflix a $2.8bn termination fee. Netflix declined to raise. The financial consequences are specific and all sit outside the operating line: the fee was booked in interest and other income in the first quarter of 2026; about $85m of financing costs were written off through interest expense; transaction costs sit inside a $107m first-half increase in third-party general and administrative expense; and none of the $67.2bn was ever drawn. Management's own framing, in April 2026, was that “Warner Bros. would have been a nice accelerant for our strategy, but only at the right price.”

For anyone reading 2026 numbers, the practical effect is that the first half is not clean and the second quarter is. Operating income and operating margin are unaffected in both periods, but first-half net income of $8.68bn and the raised full-year free cash flow guidance of about $12.5bn both contain the fee — management stated explicitly that the guide rose from about $11bn “due primarily to the after-tax impact of the Warner Bros.-related termination fee.” The underlying free cash flow guidance is roughly $11bn.

6. Cyclicality, Constraints and Position in the Cycle

Netflix is among the least cyclical businesses in media, and the reason is the contract rather than the content. Revenue is a monthly recurring fee collected in advance, cancellable at will but rarely cancelled, and it grew through both the 2020 shutdown and the 2022 consumer squeeze. The genuinely cyclical exposure is advertising, which management guides to roughly $3bn or about 6% of 2026 revenue and which Netflix itself lists as subject to “seasonal, cyclical or other shifts in advertising spend, including the impact of macroeconomic conditions.” Geographic exposure is diversified but dollar-sensitive: UCAN 44.2% of FY2025 revenue, EMEA 32.1%, LATAM 11.9% and APAC 11.8%.

Where the business sits right now

On margin, at a record and still climbing. FY2025's 29.5% operating margin is the highest in the company's history, against 13% in FY2019 and a 2022 trough of 18%; 31.5% is guided for 2026. On cash, also at a record: $9.5bn of free cash flow against a worst year of negative $3.3bn in FY2019. On growth, past the peak and decelerating clearly — 17.6% year-on-year in Q4 2025, 16.2% in Q1 2026, 13.4% in Q2 2026 and 11.7% guided for Q3 2026, with constant-currency Q3 growth of 11%. Record profitability and decelerating growth are arriving together, which is the ordinary shape of a business converting from expansion to harvest.

Three current conditions are helping and will not repeat. Currency turned from a $271m headwind in FY2025 to a $715m first-half tailwind. The $2.8bn Warner Bros. fee lifted 2026 free cash flow guidance by about $1.5bn after tax. And content amortisation is guided to grow only about 10% in 2026 while cash content spend runs at roughly 1.1x that charge — the reported margin is temporarily flattered relative to cash by a widening gap that must eventually close. Two conditions cut the other way and are also temporary: Brazilian non-income tax assessments cost $619m in FY2025 operating expense and $729m of first-half 2026 cash, which management says it does not expect to recur materially.

Durable against borrowed

What to monitor, and where it is published

7. Risks, Unknowns and Questions for Deeper Work

The risks below are those that cyclicality does not capture. They are ordered by how much they compound.

• Disclosure has been withdrawn faster than the business has simplified. Memberships and average revenue per membership went in 2025; the semi-annual view-hours report moves to annual from 2027. An investor can no longer separate volume from price, cannot compute content cost per subscriber, and cannot verify churn. The consequence is not that the business is worse — it is that the tools to detect deterioration early have been removed, so a slowdown will be visible only in the revenue line, after the fact.
• Content amortisation is an estimate that management sets, and it is the auditor's critical audit matter. Netflix amortises “based upon various factors including historical and estimated viewing patterns,” and its own risk factor states that if actual patterns differ, “it could result in greater in-period expenses, which could cause us to miss our earnings guidance.” No content impairment has ever been recognised. With $32.8bn of content assets against $13.3bn of annual operating income, a modest change in the amortisation curve is the difference between margin expansion and margin contraction, and nothing outside the company can test the estimate.
• The advertising build is a price bet, not a volume bet. The volume opportunity is real — US connected-TV upfront spend passed primetime linear for the first time in 2026. But every platform is adding inventory simultaneously, and the only company that breaks out rate against volume, Disney, reported streaming ad revenue up 3% on impressions up 8% and rates down 4%. If Netflix's ~$3bn 2026 target is met on volume while rates fall, the incremental margin on that revenue is materially worse than on subscription, and Netflix does not disclose enough to tell.
• Live sport is the one content category that cannot be smoothed. Sports rights are expensed largely as aired rather than amortised over ten years. Netflix has been adding them — an expanded NFL agreement through 2029-30, WWE Raw, MLB events, the World Baseball Classic. Comcast's Media segment operating expense rose 29.1% year on year in the June 2026 quarter on NBA and World Cup costs, which is what one heavy sports slate does to a P&L. At just over 5% of content spend the exposure is currently small; the risk is that competitive escalation makes it not small.
• Fixed content cost plus fixed obligations is the compounding failure mode. $25.1bn of content obligations, $11.9bn of it due within twelve months, with payment terms explicitly not tied to membership. Netflix says it “may be unable to react to any reduction in our cash flows from operations… by reducing our streaming content obligations in the near-term.” On its own this is survivable given $9.1bn of liquidity and $5.2bn of net debt. Combined with a growth stall and an advertising shortfall, it is the mechanism by which margin expansion reverses quickly.
• Regulatory cost is national, rising in Europe and reversing in North America. The EU's revised Audiovisual Media Services Directive requires European works to be at least 30% of on-demand catalogues and permits member states to levy providers targeting their territory. France's 2026 rule ring-fencing 20% of the SMAD contribution to animation, documentary and live performance is under challenge in court by Netflix, Amazon and Disney individually. Canada moved the other way — the government stated in July 2026 an intention to eliminate the base contribution requirement. These are direct margin items in specific markets, not headline risks.
• Concentration and uninsured exposures. Netflix runs “the vast majority of our computing on AWS” and states it “cannot easily switch”. It also states plainly that it does not carry insurance against service disruption or data breach. Device partner agreements run one to three years. None of these is likely; each is unhedged.

What the sources could not answer

Each of the following is a gap in the record rather than a failure of research, and each would need resolving before a thesis could be underwritten.

• The split of revenue growth among memberships, price and advertising — never quantified in any filing, and now unrecoverable because the membership and ARM denominators have been discontinued.
• Advertising revenue as a reported figure. It appears only in shareholder-letter commentary; the 10-K states that revenue other than membership fees “was not a material component,” which sits awkwardly against a ~$3bn 2026 target on ~$51bn of revenue.
• Churn, at any level of aggregation, in any year. Netflix has never disclosed it.
• Games and live-programming economics: no revenue, no spend, no commitments, no user metrics. The NFL and MLB agreements carry no disclosed dollar values.
• Maintenance against growth capital expenditure, and any split of content spend between library maintenance and new-title investment.
• The tax effect of the $2.8bn termination fee, which is not disclosed, so underlying first-half 2026 earnings cannot be computed precisely.
• Netflix's per-tier US prices before and after the March 2026 increase; the company discloses only a global range of $1 to $37.
• Whether the second-half 2026 margin implied by guidance — roughly 30.3% against 32.8% delivered in the first half — is entirely the content-amortisation phasing management describes, or contains something else.

8. Investor Takeaways

• What this business really is: a fixed-cost content library rented by the month to the largest paying audience in subscription video, where the marginal viewer costs nothing to serve and the cost of a title is the same whether ten million or three hundred million people watch it.
• The core economic engine: one paying membership-month generating roughly $11.59 of revenue and $3.42 of operating profit in FY2025, against content amortisation of $4.21 that was fixed before the member arrived.
• The main growth lever: price, reinforced by advertising. Household penetration is saturating in mature markets, and every major service raised price by 8% to 18% in the year to September 2026 without a demand break.
• What could break the story: a growth stall against obligations that cannot be reduced — $25.1bn of content commitments with payment terms unlinked to membership — or an advertising build that delivers volume into falling rates.
• What to monitor: constant-currency revenue growth against the decelerating trend, the cash-content-to-amortisation ratio, and whether advertising revenue ever becomes a reported line rather than a commentary figure.