On Friday, Netflix fell as much as 12.5% before closing down 7.3%, its biggest intraday drop since April 2022, per Bloomberg.
April 2022. Long-time readers know that date. Netflix had just lost subscribers for the first time in a decade, the stock collapsed 25.8% overnight, and the world was certain streaming was broken. That’s when I started covering the company in this newsletter.
Almost exactly a year ago, I wrote a report titled Netflix: Priced for Perfection. The shares had breached $1,300 (about $130 after last November’s ten-for-one split). I wrote that “the pendulum has swung completely,” that the stock wore a “golden halo” of the kind “usually reserved for defensive stalwarts like Costco,” and that if we were to invest in Netflix then, the expected returns were “not something to be excited about.”
One year later, the halo is gone. The stock has given back more than 45% from its peak.
At the end of 2021, Netflix earned $6.2 billion in operating profit, and the stock closed the year at a split-adjusted $60.24. Today, operating profit over the last twelve months stands at $14.4 billion, up 132%. The stock closed Wednesday at $68.43, up 14%.
The business more than doubled its profits over four and a half years. The stock has barely moved. Everything in between, the 2022 collapse, last year’s euphoria, Friday’s plunge, were just Mr. Market feeling very differently about Netflix’ business.
Frankly, when I reviewed the most recent quarter, the results were fine. The problem was Mr. Market expecting Netflix to grow like a young’un when it’s about to hit the big 3-0.
(Ok, technically it turns 29 next month, but 29 doesn’t have quite the punch of 30.)
The Quarter in Ninety Seconds
Q2 itself was fine. Revenue of $12.56 billion grew 13.4% (12% excluding currency), and EPS of $0.80 came in slightly above forecast. Operating margin was 33.4%, down from 34.1% a year ago but a touch ahead of plan on timing of spend.
Investors were probably disappointed with the guidance. Netflix guided Q3 revenue to $12.86 billion, up 11.7%, a notable deceleration from recent quarters, with EPS of $0.82 against the street’s $0.84. The full-year range narrowed to $51.0 to $51.4 billion, midpoint unchanged, with the 31.5% margin and roughly $12.5 billion of free cash flow reaffirmed.
Long story short: the year is intact. But there’ll be a bit of a speedbump next quarter.
CFO Spencer Neumann explanation on Q3’s deceleration: “We don’t manage the business on a quarter-to-quarter basis.” He also noted that last year’s growth was more back-half weighted, which flatters this year’s first half and dings the second.
What’s Actually Driving Growth
Viewing hours grew 2% in the first half, to 97 billion. Revenue grew 13%. So where’s the growth coming from? Price and ads, not eyeballs.
It’s great that Netflix can demonstrate this kind of pricing power, and it’s well placed to keep growing the ad tier. But the raw material for both, especially ads, is member growth, which they’ve stopped reporting. Hours up just 2% with membership growth (per what management said), just means that engagement has fallen.
Source: Fiscal AI (get a 15% discount using this link)
The Ad Engine
Netflix didn’t formally disclose ad revenue in its early years, but piecing together management’s disclosures, the trajectory looks like this:
2024: ~$600 million (implied from the disclosure that 2025 grew “more than 2.5x”)
2025: over $1.5 billion (Q4 2025 shareholder letter)
2026 target: ~$3 billion (reaffirmed in the Q2 letter)
A business on track to 5x in two years. But how it’s growing matters as much as how fast.
The growth isn’t coming from charging advertisers more. Management has been explicit about this. Asked on the Q4 2025 call whether Netflix could deliver another rough doubling of ad revenue, Greg Peters said:
“…we see all of that adding up to the same kind of performance we saw the last year, which means we can grow revenue targeting doubling that revenue growth by improving fill rate and growing inventory with similar CPMs.”
Two levers, and price isn’t one of them. Fill rate is the percentage of available ad slots that actually get sold. And “growing inventory” doesn’t mean showing you more ads per hour.
Peters defined it himself in an interview with Stratechery in January: “One driver is just inventory growth. We’re getting more subscribers on the ads plan, that means there’s more inventory to sell.” More members watching, not more ads per member.
Think of a stadium. They aren’t raising ticket prices or squeezing extra seats into your row. They’re filling the empty ones and building more stands.
Why not just raise prices? Because in advertising, Netflix can’t. Netflix entered the ad market in late 2022 with CPMs (the price per thousand views) reportedly as high as $50 to $60, among the most expensive in streaming, and spent 2023 walking them down as buyers pushed back.
Then Amazon entered the market. In early 2024, Amazon converted its entire Prime Video base to ad-supported viewing, with ad-free as a $2.99 opt-out, flooding the market with inventory overnight. By June 2024, the Wall Street Journal reported Netflix was asking some brands for roughly $29 to $35 per thousand viewers, down from $39 to $45 the previous summer, with buyers saying Amazon was driving down prices for everyone.
To appreciate the difference in scale of their ad tiers, at the time, Amazon said Prime Video’s ad tier averaged 115 million monthly viewers in the US alone. Netflix told advertisers its ad tier reached 40 million monthly actives globally, up from 23 million four months earlier.
This is the structural difference between Netflix’s two businesses. On subscriptions, demand is relatively price inelastic: as long as the value delivered exceeds the price paid, Netflix can keep nudging prices up without meaningful churn. And the value gap is still wide. Peters pointed out on the Q2 call that US subscribers pay the least per hour of viewing among comparable services, in some cases half what a competitor charges per hour. That’s why the first-half price hikes in the US, Mexico and Spain went through with the impact “consistent with prior price changes and our expectations,” in the letter’s words. Netflix prices behind the value it delivers, and the room to raise hasn’t run out.
Advertising doesn’t work like that. The customer is the advertiser, and advertisers weigh returns on every dollar of spend. If Netflix quotes $50, they buy Prime Video instead, all else constant.
So Netflix’s ad growth plan runs through volume. The ad tech build-out goal: make Netflix easier to buy, for more advertisers, with less human effort. From the Q1 2026 call:
“…we’ve added more and more DSPs, which, of course, are more ways to buy. And we’re seeing through that pretty significant growth in programmatic, which is on its way to becoming more than 50% of our nonlive ads business… Our advertiser base grew over 70% year-over-year in 2025 to be more than 4,000 advertisers… Today, we’re still currently concentrating in those top advertising accounts, the largest buyers, which are serviced primarily by the Netflix sales team…”
A sales team can only hand-serve so many accounts. Software can serve them all. That’s how you grow 4,000 advertisers into a much bigger number, and this quarter the expansion took its next step. From the Q2 letter:
“We’re also automating more of the workflow around how advertisers transact with us by extending programmatic access to Pause Ads and live inventory this summer. This reduces the manual effort that has historically limited access for smaller buyers, opening Netflix to a broader range of advertisers over time.”
Live inventory joins the automated pipes, and the door opens to more buyers. More advertisers competing for the same slots means higher fill rates, which is exactly the lever management says drives revenue. It also helps that Netflix heads into the US upfront with its strongest hand yet. The letter says negotiations are in advanced stages with commitments expected to close in the next few weeks, on the back of a live slate spanning the Women’s World Cup, an expanded NFL slate, WWE and MLB.
Meanwhile, the ad tier still earns less per member than the ad-free plans. I’m a glass half full kinda guy, so I read that gap as stored upside: revenue Netflix can capture from members it already has, just by getting better at selling ads.
Peters on the Q2 call: “…there’s still a gap between ad tier ARM and then ARM for our standard without ads tier. But that gap is narrowing. And I think of that gap is essentially near-term underrealized revenue growth… We’re adding features. We’re adding more ads products. We’re adding more measurement. We’re making it easier for us — for folks to transact with us. Those all drive demand. They drive competitiveness. That yields increased fill rates. It pushes ads arm higher.”
And parity carries a second-order effect I find more interesting than the gap itself. Today, when Netflix raises prices on the ad-free tiers, some members downgrade to the ads plan and Netflix earns less on them. At parity, that downgrade stops costing anything, and price hikes on the premium tiers become close to risk-free.
But the most interesting thing management said about ads this quarter wasn’t about ads at all:
“A free offering could make sense in some markets… having an effective scaled ads business in any candidate country for such an offering is clearly an important enabling factor to make those economics work… free is something that we’re going to continue to consider, but we have no near-term plans to launch something.”
A free Netflix. Hastings spent a decade saying Netflix would never run ads, then reversed course. No sports became the NFL on Christmas Day. Builders, not buyers became an $82.7 billion bid for Warner Bros. This management has no sacred cows, only economics. Today they’re openly entertaining a free tier. That’s how serious they are about ads as a future growth driver.
Sometimes the Silence is Deafening
For years, many of Netflix’s shareholder letters included a Nielsen chart showing its share of US TV time. The chart last appeared in the Q3 2025 letter. The Q4 letter kept a one-line boast about December’s record share. This quarter’s letter contains no Nielsen reference at all. First the chart went. Then the sentence.
Look further back and the same chart used to show Netflix visibly winning TV time. This one is from Q1 2024:
So what does the public data say? Nielsen’s latest reading, the April Gauge, had Netflix at 7.8% of US TV time. That’s below the 9.0% platform record from December, but up from March, and up 0.3 points from April last year. February and March were dented by a Super Bowl and a Winter Olympics pulling eyeballs to broadcast. Nothing in the public data screams share loss.
The company also announced that its What We Watched engagement report, published twice a year since 2023, will move to an annual cycle. The official rationale, from the letter: “The goal of separating the publication of the report from our earnings results is to keep the focus on our primary financial metrics – revenue and operating profit.”
There’s a pattern here. Netflix stopped reporting quarterly subscriber numbers when subscriber growth stopped being the story it wanted to tell. Now engagement disclosure shrinks just as engagement becomes the question.
It’s annoying, but it’s usually a rite of passage for companies conceding they’ve matured. That doesn’t mean they’ll do poorly for shareholders. It means the metrics they used to impress us won’t be as impressive anymore, simply because of how big they’ve become.
Take Apple. It stopped reporting iPhone unit sales in November 2018, right as unit growth stalled, and pointed investors to monetizing its installed base instead. The market punished the opacity first and adjusted later.
The Season Two Fight
On July 5, Bloomberg published the piece that lit the fuse: One Piece season two down more than 30%, Beef down more than 70%, The Night Agent down roughly 50% for season two and another 35% for season three, Avatar: The Last Airbender down 60% in week one. Within days, investors were worried Netflix had lost its edge.
Sarandos came to the call loaded and ready to address it head on:
“In aggregate, we are not seeing any material change in our second season viewing compared to season 1. Our second seasons are performing well within our bands of expectation… When we look across the entire portfolio, across all the regions, all the content categories, our season 2 falloff is actually slightly improved this year relative to last year. Now of course, you can pick any 5 data points to tell any story you want.”
He counted. Bloomberg’s piece used exactly five shows.
Buried in Shaw’s July 13 newsletter is the baseline that reframes everything: the average Netflix show loses more than 30% of its audience after its first season. Hold the five data points against that yardstick and One Piece looks ordinary. Beef and Avatar were genuine misses. The Night Agent’s season two underperformed, and season three’s drop was close to the norm. Five data points, two real outliers.
Nor is this a Netflix disease. The Rings of Power fell 71% between seasons, on Amazon. Wednesday and Poker Face shed viewers after multi-year gaps. Shaw’s own verdict: “This is not a Netflix phenomenon but an industry one… Netflix is just the only service that gives us the weekly data to judge it.” And on the popular theory that long gaps are fatal: “there is still no definitive evidence that time off hurts a show” as a rule. Bridgerton, Stranger Things, and Severance all came back from long breaks as big as ever, or bigger.
But one variable does keep showing up in the data: cadence.
Shows that returned within about a year grew their audiences, across five different platforms: The Pitt up 172%, Love Island USA up 143%, Landman up 103%, The Traitors up 66%, The Lincoln Lawyer up 19%. Netflix’s Night Agent season three, back after a shorter gap but carrying a weaker season two, fell 39%.
So I do think Netflix has a problem to fix. It’s not that they lost their touch or their magic. It’s their cadence. Sarandos closed his answer with “no changes in release strategies.” But the binge model isn’t what the data indicts. The two-year gap between seasons is. That question deserves a direct answer, and it didn’t get one.
That’s the full picture. An ad business compounding, a disclosure retreat worth watching, and a cadence problem worth fixing. What’s left is the decision: what 20 times earnings actually buys, and what I’m doing with my own position.
Valuation
Last July, Netflix traded at 48x forward earnings. Today it trades at roughly 19.8x, about 37% below its 10-year median of 31.7x. Same business, better numbers, less than half the multiple.
Source: Fiscal AI (get a 15% discount using this link)
The balance sheet tells you what management thinks of that. In April, the board authorized an additional $25 billion of buybacks on top of the $6.8 billion remaining at the end of Q1. In Q2, Netflix repurchased $4.7 billion of stock, its largest quarter of buybacks ever, leaving $27.1 billion of capacity.
So what returns can we expect from here? If the multiple stays parked at 20x, returns simply track earnings per share. And EPS should grow faster than revenue, for two reasons.
First, operating leverage. Revenue is guided to grow 13% to 14% this year, but the letter says the plan implies operating income growth of 20%+. Costs don’t scale one-for-one with revenue. Incremental ad dollars and price increases drop through at high margins.
Second, the buyback. That $4.7 billion in Q2 alone equals about 1.6% of the market cap (my math). With $27.1 billion of authorization and roughly $12.5 billion of annual free cash flow feeding it, Netflix can retire shares at a meaningful clip. And the cheaper the stock stays, the more shares each dollar buys.
Stack it up: if we assume high single digit to low-teens revenue growth, a few points of operating leverage, and 2% to 4% of shares retired each year. That gets you to low to mid-teens in returns without any help from the multiple.
Compound steadily,
Thomas
Disclaimer: This research report constitutes the author’s personal views only and is for educational purposes only. It is not to be construed as financial advice in any shape or form. From time to time, the author may hold positions in the below-mentioned stocks consistent with the views and opinions expressed in this article. Disclosure – I hold a position in Netflix at the time of publishing this article (this is a disclosure and NOT A RECOMMENDATION).