Some earnings reports sink a stock because of a bad number. And then there are reports like the one Netflix delivered on Thursday evening: record revenue, earnings per share above expectations, an operating margin that beat the company’s own guidance, and the largest share buyback in its history. The stock fell as much as 9% after hours to a 52-week low — its weakest level since September 2024.
The trigger was not a number in the report. The trigger was a number that will disappear from future reports.
Netflix announced that starting in the first quarter of 2027, it will publish its “What We Watched” engagement report only once a year instead of twice. The sentence itself is unremarkable: “Starting in Q1 2027, we will move from a six-month to a yearly snapshot.” A line about publishing cadence. It moved the stock more than the record revenue did.
That sounds like a footnote for media reporters. It is, in fact, one of the most instructive market reactions of the summer — because it demonstrates that disclosure is not decoration on an equity story. It is part of the valuation.
What Netflix actually reported
Start with what was in the report, because the notion that Netflix delivered a bad quarter is simply wrong.
Second-quarter 2026 revenue rose 13.4% to $12.56 billion — the highest quarterly revenue in company history, though it missed consensus of $12.58–12.59 billion by a hair. Net income came in at $3.401 billion, earnings per share at $0.80, a penny ahead of estimates. Operating income climbed 11% to $4.19 billion.
The 33.4% operating margin was below last year’s 34.1% — but above the company’s own 32.6% guidance. That distinction matters: on profitability, Netflix beat its own promise rather than breaking it.
Two figures did disappoint. Free cash flow fell by a third to $1.53 billion from $2.27 billion a year earlier, explained in part by taxes tied to the Warner Bros. termination fee. And the growth rate: on an FX-neutral basis, revenue grew 11%, down from 12% in the prior quarter. That is the second consecutive quarter of deceleration.
The real stumbling block was the outlook. For the third quarter, Netflix guided to $12.86 billion in revenue, up 11.7% — Wall Street had penciled in roughly $13 billion. Guided EPS of $0.82 also fell short of the $0.84 consensus. Two misses landing in the same release tend to weigh more heavily than either alone. Full-year guidance was narrowed to $51.0–51.4 billion from $50.7–51.7 billion.
That leaves a line on the page nobody can argue away: roughly 16% growth in Q1, 13.4% in Q2, a guided 11.7% in Q3. Three quarters, one direction.
The retreat from the data
And it is precisely into that deceleration that Netflix announces it will publish less.
The company’s stated rationale: it wants to decouple the engagement report from earnings day so investors focus on revenue and operating profit. Taken on its own, that is a defensible argument. A company is entitled to decide which metrics it puts at the center.
The problem is the context. Netflix already stopped reporting membership numbers on a regular basis in 2025 — for decades the single most important metric in streaming and the metronome of every Netflix earnings season. After that, the semi-annual engagement report was the last hard operating metric outsiders had. Now it is being halved.
Consider what Netflix does not report at all: no average revenue per account, no gross margin by segment, and above all no churn. Without churn, customer lifetime value cannot be calculated from the outside with any rigor. Analysts have complained for some time that an already thin set of operating metrics keeps getting thinner. Anyone trying to verify that the Netflix machine is running well has very few instruments left.
Then there is the timing. The move follows Bloomberg reporting that Netflix originals suffer significant audience declines in their second seasons — the “sophomore slump” debate. Netflix explicitly rejects that narrative. Co-CEO Ted Sarandos told analysts: “Our Season 2 fall-off is actually slightly improved this year relative to last year.” Improved, in other words, not worse.
That may well be true. It is probably true. But it creates an awkward configuration: management says engagement is healthy — and from 2027 will publish less often the data that would let you check. The combination is the problem. Neither statement is a problem on its own.
The arithmetic nobody enjoys saying out loud
Why does engagement matter so much here? Because one simple calculation reveals where the growth now comes from.
In the first half of 2026, Netflix members watched more than 97 billion hours — by the company’s own account the highest half-year total ever. Growth, however, was just 2%, after 1.5% the year before. To be fair: against competition from the Olympics and the football World Cup.
Two percent more hours. And 13.4% more revenue.
That gap is the entire story. If people watch only 2% more but revenue rises at more than six times that rate, then growth is coming almost entirely from two other sources: higher prices and advertising. Netflix raised prices in the first half — Co-CEO Greg Peters on the results: “Our first half price changes … have gone well. The results are consistent with our expectations.” And the advertising business is set to roughly double to about $3 billion annually in 2026.
Both are working. But both are finite.
An engine with a limited tank
Price increases on flat usage are an engine with a limited tank: at some point, willingness to cancel rises. Advertising is cyclical, and against $51 billion in total revenue its roughly 6% share is still small. Netflix’s own reporting also shows that viewing hours per member are declining year over year.
The one metric that would reveal how much runway this price-and-advertising engine has left is engagement. And that is the one moving to once a year.
This is not an accounting scandal, and nothing here suggests one. It is something subtler and, for investors, more common: a business whose growth driver has quietly shifted from volume to price — while the volume data gets harder to see.
Why disclosure is part of the valuation
Here lies the real lesson of the session, and it extends well beyond Netflix.
Netflix once traded above 100 times earnings. Investors paid that premium not for current profits but for the belief that the best years lay ahead. Today the forward multiple is roughly 19 times, measured against 2026 consensus EPS of about $3.51. The stock is down more than 40% from a year ago and roughly 20% year to date. Thursday’s close before the print was $74.35; after hours it traded near $67.60.
The crucial point: this decline has almost never been a decline of the business. Profits are growing, the subscriber base is enormous, the advertising business is only just being built. What fell is the price investors are willing to pay for the future. That is multiple compression, and it has been the quiet protagonist of this stock for a year. Add an “acquisition hangover” — Netflix’s co-CEOs called the Warner Bros. Discovery deal “no longer financially attractive” at the price required to match, yet the valuation discount persisted even after the uncertainty resolved — plus the long-term worry that artificial intelligence upends content production.
In precisely this phase, disclosure stops being a side issue. As long as a company grows at an obvious double-digit clip, it has little to prove; the price writes the story by itself. Once growth cools toward 11%, every claim has to be checkable. A company whose statements cannot be verified gets a discount — not out of market malice, but because uncertainty has a price. When a growth story matures, transparency becomes the equity story. Cut it, and you trade a growth multiple for a value multiple.
Analysts repriced immediately: Goldman Sachs cut its target to $94 from $110, Morgan Stanley sits at $90, KeyBanc at $92, JPMorgan at $85. The average is around $90, and most houses still rate the stock a buy. Nobody is calling for collapse. Everybody is modeling less.
What it means for US investors
For American investors, the read-across runs across the whole communication services complex. Disney, Warner Bros. Discovery, Comcast’s NBCUniversal, Paramount and Roku are all fighting for the same attention and increasingly the same ad dollars — and Netflix is now pushing into that ad market hard. If its ad revenue doubles to $3 billion, that money does not come from nowhere; it comes from budgets that legacy media and connected-TV platforms have long considered theirs. Amazon’s Prime Video and Apple TV apply the same pressure from the other side, with balance sheets that do not need streaming to pay for itself.
There is also a useful analogue in Spotify: another subscription business that has leaned on price increases and advertising to grow as user growth normalizes. The playbook is the same, and so is the question — how many times can you raise the price before usage answers back?
On taxes, the mechanics are straightforward and worth remembering in a year like this one. Netflix pays no dividend, so this is purely a capital gains story: long-term rates of 0%, 15% or 20% depending on income, plus the 3.8% net investment income tax for higher earners. Positions held under a year are taxed as ordinary income. And for anyone sitting on a Netflix loss in a taxable account, this is the classic setup where tax-loss harvesting deserves a look — provided you respect the wash-sale rule and, more importantly, that the tax tail does not wag the investment dog.
One more takeaway that has nothing to do with Netflix: go through the companies in your portfolio and count how many verifiable operating metrics they still publish — and whether that number has risen or fallen over the past few years. It is one of the most underrated questions in equity analysis.
The case for Netflix
You can tell this story the other way around, and the counterarguments are strong.
First, the quarter was objectively good. An EPS beat, a margin above guidance, record revenue — that is not the picture of a company in trouble. Second, Greg Peters is substantively right when he pushes back: “All hours are not created equal. All hours don’t provide the same kind of value to the business.” An hour on the ad-supported tier monetizes very differently from an hour on premium. Raw hours really are a crude compass for revenue — stare only at them and you will miss the target. Fittingly, Peters noted that six of the ten best new-member sign-up days over the past five years came from live events, even though live accounts for only about 1% of viewing hours while consuming more than 5% of content spend.
Third, the advertising business is doubling, and the gap between revenue per member on the ad tier and the ad-free standard tier is slowly closing — “That gap is narrowing,” as Peters put it. Fourth, Netflix repurchased $4.7 billion of its own stock in the quarter, the largest buyback in company history, with roughly $27 billion of authorization remaining. Fifth, content spend is rising only about 10% in 2026, slower than revenue — Sarandos frames it as principle: “We grow the content spend slower than revenue.”
And finally the valuation itself. Roughly 19 times forward earnings for a company growing 11–13% with a 33% operating margin is not an absurd price. A substantial share of the bad news is already in the stock. CFO Spencer Neumann summed up the house view: “We don’t manage the business on a quarter-to-quarter basis. Our goal is to sustain healthy revenue and profit growth.”
The proof is coming — just less often
The irony of Thursday evening is that Netflix got exactly what it asked for and was punished for exactly that. The company wanted attention shifted away from viewing hours and toward revenue and operating profit. The market promptly looked at revenue and operating profit — and found 11.7% growth ahead.
The backdrop was remarkable. As Netflix’s numbers landed, Asia sold off hard on AI chips Friday morning. The Nikkei 225 lost 5.2%, its steepest drop since March, and the MSCI Asia Pacific fell 2.7%; Kioxia collapsed 16%, Advantest and SoftBank around 9% each, and even TSMC fell more than 3% despite record profits. The market is transfixed by whether AI capex will pay off — and largely missed that a very different kind of warning came out of Los Gatos. Not: demand is cracking. But: soon you will get to check less often.
The next genuine test remains engagement. If the next snapshot confirms management’s account, this week’s skepticism was overdone and a 19x forward multiple will look like a gift. If it does not, the market smelled it correctly today. Either way, that proof will now be furnished once a year instead of twice.
For investors, none of this automatically says “sell.” It says this: from 2027, owning Netflix is somewhat more an investment in trusting management and somewhat less one in verifiable data. Plenty of portfolios can carry that comfortably. You should simply know you are making the trade — because the market will price it whether you notice or not.
Try TradingView Free for 30 Days
Plus get a $15 discount on your first subscription through this link.
