Connected TV Advertising Statistics 2026
By Axis Intelligence Research
Co-author: Alex Rivera | Last updated: August 26, 2026 | License: CC BY 4.0
A thousand hours of streaming on Roku produced $17.75 in advertising revenue during the quarter ended June 30, 2026 — up from $15.22 a year earlier. Streaming reached 48.6% of all U.S. television watch-time in May 2026, yet 2026 is the first year U.S. social video ad spend is projected to grow faster than connected TV.
Quick Answer
Connected TV advertising in 2026 is growing on price, not on hours. Axis Intelligence Research finds that Roku’s Streaming Hour Ad Yield (SHAY) reached $17.75 per 1,000 streaming hours in Q2 2026, a 16.65% increase, while streaming hours on the same platform rose only 7%. U.S. digital video ad spend is projected to pass $80 billion in 2026 at 11% growth, and roughly 34% of all American TV watch-time is now ad-supported streaming.
Key Findings
- Axis Intelligence Research finds that Roku’s Streaming Hour Ad Yield (SHAY) — advertising revenue per 1,000 streaming hours — reached $17.75 in Q2 2026 against $15.22 in Q2 2025, a 16.65% increase, on streaming hours that grew just 7%.
- Axis Intelligence Research calculates that ad-supported streaming accounts for approximately 34.0% of total U.S. television watch-time as of Q1 2026, derived from Nielsen’s Ad Supported Gauge: streaming held 46.6% of ad-supported TV, and ad-supported viewing represented nearly 73% of all TV.
- Magnite’s connected-TV contribution ex-TAC grew 36% year over year to $97.1 million in Q2 2026 while its non-CTV segment grew 2%, per the company’s August 5, 2026 results — a 34-point growth spread between streaming and open-web inventory inside one seller’s own book.
- Axis Intelligence Research measures a CTV Growth Premium (CGP) of 5.62 percentage points in Q2 2026: a dollar-weighted CTV panel grew 26.12% against 20.50% for the broad digital advertising panel over the same quarter.
- U.S. digital video advertising is projected to exceed $80 billion in 2026, growing 11%, and to clear 60% of total TV/video ad spend for the first time, according to IAB’s 2026 Digital Video Ad Spend & Strategy Report — with social video growing 13% and outpacing CTV for the first time.
How Much Is Connected TV Advertising Worth in 2026?
Nobody can size CTV precisely from primary sources, because most of the inventory sits inside companies that never break it out. Amazon folds Prime Video advertising into a single advertising-services line. Disney, Paramount and NBCUniversal report advertising across linear and streaming together. Netflix, which will book roughly $3 billion in advertising this year, publishes no advertising line at all.
What can be sized is the container. The IAB’s 2026 Digital Video Ad Spend & Strategy Report puts U.S. digital video ad spending above $80 billion in 2026, up 11% year over year — roughly 20% faster than the total ad market. Digital video here means CTV plus online video plus social video, and the IAB expects that combined category to exceed 60% of total TV/video ad spend for the first time this year.
| Metric | Value | Period | Source |
|---|---|---|---|
| U.S. digital video ad spend | >$80.0B | 2026 (projected) | IAB Digital Video Ad Spend & Strategy Report |
| Digital video YoY growth | 11% | 2026 (projected) | IAB |
| Digital video share of TV/video ad spend | >60% | 2026 (projected) | IAB |
| CTV growth rate | 11% | 2026 (projected) | IAB |
| Social video growth rate | 13% | 2026 (projected) | IAB |
| U.S. digital video ad revenue | $78.0B | FY 2025 | IAB/PwC Internet Advertising Revenue Report |
| U.S. digital video YoY growth | 25.4% | FY 2025 | IAB/PwC |
| U.S. total digital ad revenue | $294.6B | FY 2025 | IAB/PwC |
Two numbers in that table deserve to be read against each other. Digital video grew 25.4% in 2025. It is projected to grow 11% in 2026. The category did not stop working; it stopped being new.
Alex Rivera: The headline everyone will reach for is “$80 billion.” The headline a buyer should reach for is the growth deceleration from 25.4% to 11% in a single year, because that is the number that changes what you can negotiate. When a channel compounds at 25%, sellers hold rate. When it compounds at 11% and its nearest neighbor is compounding at 13%, the upfront conversation changes tone. This is the first year in CTV’s history where walking away from a deal is a credible position rather than a bluff.
Why 2026 broke CTV’s growth streak
The IAB’s own framing of the 2026 forecast is that social video, at 13%, notably outpaces CTV for the first time — driven by AI-powered personalization and creator-economy spend. CTV’s counterweight is sports: the NBA on Amazon, the NFL on Peacock and Netflix, and MLB spread across platforms.
That is a structural point worth naming plainly. CTV’s growth is now tied to live rights acquisition, which is lumpy, expensive and concentrated in a handful of buyers. Social video’s growth is tied to creator supply, which is effectively unlimited and costs the platform nothing to expand.
What Is the Streaming Hour Ad Yield (SHAY)?
The single hardest question in connected TV is what an hour of viewing is actually worth. Impressions are not disclosed. CPMs are negotiated privately. Reach figures are marketing.
Exactly one major CTV platform publishes both halves of the equation in the same document, which is why Axis Intelligence Research built a metric around it.
SHAY — Streaming Hour Ad Yield — measures the advertising revenue a connected-TV platform generates per 1,000 hours streamed on that platform in a given quarter.
Formula:
SHAY = advertising revenue (USD) ÷ ( streaming hours ÷ 1,000 )
Q2 2026 computation, every input shown:
Advertising revenue, Q2 2026 = $672,816,000 (Roku Form 8-K, Ex. 99.1, Aug 6 2026)
Streaming hours, Q2 2026 = 37,900,000,000 (same document)
SHAY = 672,816,000 ÷ (37,900,000,000 ÷ 1,000) = $17.7524 per 1,000 hours
Advertising revenue, Q2 2025 = $539,053,000 (same document, comparative column)
Streaming hours, Q2 2025 = 35,420,560,748 (37.9B ÷ 1.07; Roku states hours grew 7% YoY)
SHAY = 539,053,000 ÷ (35,420,560,748 ÷ 1,000) = $15.2186 per 1,000 hours
Change = +16.65%
| Reading | Advertising revenue | Streaming hours | SHAY (per 1,000 hrs) | Source |
|---|---|---|---|---|
| Q2 2026 | $672.8M | 37.9B | $17.75 | Roku Form 8-K, Ex. 99.1 |
| Q2 2025 | $539.1M | 35.4B (derived) | $15.22 | Roku Form 8-K, Ex. 99.1 |
| Change | +24.81% | +7.0% | +16.65% | Axis Intelligence Research |
The decomposition is the finding. Multiply the two growth rates and Roku’s reported advertising growth falls out exactly:
1.070 × 1.1665 − 1 = 24.81% vs. reported advertising revenue growth of 24.81%
A quarter of the growth came from more viewing. Three quarters came from each hour being worth more. Roku’s advertising gross margin moved with it, expanding to 62.4% from 56.0% a year earlier — the company attributes that to a mix shift toward higher-margin ad products, which is what monetizing the same hour harder looks like on a margin line.
Scope of the reading. SHAY is a platform-level yield measure, not a CPM. It divides all advertising revenue by all streaming hours, including hours on services that Roku does not monetize with its own inventory, so it understates the yield on Roku-sold inventory and overstates nothing. It is comparable across quarters for the same platform, and it is not comparable across platforms with different hour definitions. Roku is currently the only CTV platform of scale that discloses both inputs; when others do, the metric extends without modification.
Alex Rivera: Every planner has had the argument about whether CTV inflation is real. SHAY settles the direction, if not the magnitude. On the one platform where you can watch the arithmetic happen, the price of attention rose 16.65% while the supply of attention rose 7%. Hold your CTV budget flat into that and you buy 85.7% of last year’s yield-equivalent volume. That is not a rate card you can argue with, because it is not a rate card — it is two disclosed numbers dividing into each other.
How Much of American TV Viewing Is Actually Ad-Supported?
The Gauge is the number everyone quotes: streaming reached 48.6% of total U.S. TV watch-time in May 2026, up a full point, with broadcast at 19.2% and cable at 20.4%. Nielsen notes in the same release that the Gauge counts both ad-supported and non-ad-supported viewing, which makes it the wrong denominator for an advertiser.
The right one comes from a different Nielsen product. The Q1 2026 Ad Supported Gauge reports that ad-supported viewing held nearly 73% of all TV, and that streaming captured a record 46.6% of that ad-supported total.
ASST — Ad-Supported Streaming Share of Total TV — combines the two to answer the question a media plan actually asks.
ASST = (streaming share of ad-supported TV) × (ad-supported share of total TV)
= 46.6% × 73%
= 34.02% of total U.S. TV watch-time
| Category | Share of ad-supported TV (Q1 2026) | Implied share of total TV | Source |
|---|---|---|---|
| Streaming | 46.6% | 34.02% | Nielsen Ad Supported Gauge Q1 2026 |
| Broadcast | 28.2% | 20.59% | Nielsen Ad Supported Gauge Q1 2026 |
| Cable | 25.2% | 18.40% | Nielsen Ad Supported Gauge Q1 2026 |
| Non-ad-supported (all) | — | ~27% | Axis Intelligence Research |
Because Nielsen states ad-supported viewing as “nearly 73%” rather than a precise figure, 34.02% is a ceiling rather than a point estimate. Axis Intelligence Research publishes it as approximately 34% and will re-baseline when Nielsen publishes an exact quarterly figure.
The distribution-level picture explains where that third of American TV time sits.
| Distributor | Share of total U.S. TV watch-time, May 2026 | Monthly change | Source |
|---|---|---|---|
| YouTube | 13.8% | +0.4 pts | Nielsen Media Distributor Gauge |
| Disney | 10.0% | — | Nielsen Media Distributor Gauge |
| NBCU + Versant | 8.4% | +0.2 pts | Nielsen Media Distributor Gauge |
| Netflix | 8.0% | +0.2 pts | Nielsen Media Distributor Gauge |
| Prime Video | 4.5% | +0.3 pts | Nielsen Media Distributor Gauge |
| The Roku Channel | 3.1% | +0.1 pts | Nielsen Media Distributor Gauge |
| Paramount Streaming | 2.3% | +0.2 pts | Nielsen Media Distributor Gauge |
YouTube has held the largest share of American television among all distributors for three consecutive months and posted the single largest monthly share gain in May. The category that CTV buyers historically treated as adjacent inventory is the largest television distributor in the United States.
Alex Rivera: The 48.6% figure gets quoted in every deck and it is the wrong number for a media plan, because roughly a quarter of it is inventory nobody can buy. Ad-supported streaming at about a third of total TV time is the honest addressable figure, and it is still the largest single ad-supported block in American television — bigger than broadcast and cable individually. Plan against 34%, not 48.6%, and the reach math in your flowchart stops overpromising.
Which Sellers Are Capturing CTV Ad Dollars?
The clearest read on where streaming money is going comes from the intermediaries, because they publish the split that content owners refuse to.
Magnite’s Q2 2026 results report CTV contribution ex-TAC of $97.1 million, up 36%, against DV+ — mobile and desktop inventory outside connected TV — at $92.5 million, up 2%. Same company, same quarter, same salesforce, 34 points of growth spread.
| Segment | Q2 2026 contribution ex-TAC | YoY growth | Share of total | Source |
|---|---|---|---|---|
| CTV | $97.1M | +36% | 51.21% | Magnite Q2 2026 results |
| DV+ (non-CTV) | $92.5M | +2% | 48.79% | Magnite Q2 2026 results |
| Total | $189.6M | +17% | 100% | Magnite Q2 2026 results |
Axis Intelligence Research calculates that CTV crossed half of Magnite’s business this quarter for the first time — 51.21%, against 44.07% a year earlier, using the $71.4 million Q2 2025 CTV figure implied by the company’s stated 36% growth.
The buy side tells a harder story. The Trade Desk reported Q2 2026 revenue of $715 million, up 3%, with video including CTV at a low-50% share of spend on its platform, and guided Q3 to at least $650 million. A supply-side specialist growing CTV at 36% and a demand-side specialist growing total revenue at 3% in the same quarter is not a contradiction; it is a description of where pricing power now sits.
The CTV Growth Premium
CGP — CTV Growth Premium — measures how much faster connected-TV advertising revenue is growing than the broad digital advertising market in the same quarter.
CGP = (dollar-weighted YoY growth, CTV panel) − (YoY growth, broad digital panel)
Q2 2026 computation:
Roku advertising 672.8 vs 539.1 → +24.81%
Magnite CTV ex-TAC 97.1 vs 71.4 → +36.00% (Q2 2025 implied from stated growth)
Dollar-weighted CTV panel: (672.8 + 97.1) ÷ (539.1 + 71.4) − 1 = +26.12%
Broad digital panel, Q2 2026 (Axis registry) = +20.50%
CGP = 26.12 − 20.50 = 5.62 percentage points
The broad digital benchmark is the Q2 2026 panel growth published on the Axis Intelligence Research Digital Marketing Statistics 2026 page, covering Alphabet, Meta, Amazon, Pinterest and Reddit. Holding the same as-of quarter on both sides is the whole point of the comparison.
Scope of the reading. The CTV panel is two publicly reporting sellers that disclose a clean streaming advertising figure — one platform owner, one exchange. It excludes Netflix, Amazon’s video inventory, Disney, Paramount and NBCUniversal because none discloses a separable CTV advertising number, and inserting an estimate would break the arithmetic the metric exists to show. A 5.62-point premium says CTV is still outgrowing digital advertising overall in dollar terms. It is a smaller premium than the category’s reputation implies.
Alex Rivera: Roku’s own letter says the quiet part: nearly three-quarters of in-stream video ad spend on its platform now arrives through third-party demand-side platforms — Amazon DSP, The Trade Desk, Google DV360, Yahoo. A platform that owns the operating system, the home screen and the first-party graph is routing three of every four ad dollars through somebody else’s pipe. That is what a maturing channel looks like from the inside, and it is why the interesting margin question in CTV has moved from “who owns the eyeballs” to “who owns the auction.”
What Is Changing About How CTV Inventory Gets Bought?
The buying criteria flipped this year. IAB reports that targeting overtook content quality as the top criterion for TV/video investment, up 10 points year over year — and that the shift is being driven hardest by small and mid-size spenders, up 23 points, the same group most exposed to identity degradation in the open market.
Agentic AI moved from pilot to plan on a similar timeline. Across digital video buyers, IAB finds 21% live, 20% testing and 25% planning to use agentic AI for video campaigns in 2026, with a further 28% investigating.
| Adoption stage, agentic AI for digital video | Share of buyers | Source |
|---|---|---|
| Live | 21% | IAB 2026 Digital Video Ad Spend & Strategy Report |
| Testing | 20% | IAB |
| Planning to use in 2026 | 25% | IAB |
| Actively investigating | 28% | IAB |
Roku’s disclosure lands in the same frame from the sell side: advertising gross margin expanded 650 basis points year over year on a mix shift toward higher-margin formats, and spend from non-media-and-entertainment advertisers passed a third of Roku Experience advertising revenue — evidence that the buyer base is widening past the endemic streaming-services base that built the channel.
The data layer regulators are watching
The targeting that makes CTV work runs on household-level identity, and the most detailed federal account of how streaming services assemble it remains the FTC’s September 2024 staff report, A Look Behind the Screens, built on 6(b) orders issued to nine social media and video streaming companies. The Commission’s summary of that report recommended limiting data retention and sharing and restricting targeted advertising. No CTV measurement or identity standard has been federally mandated since, which means the addressability that buyers now rank above content quality rests on private agreements rather than rules.
Where Does Political Advertising Fit in the 2026 Cycle?
This is the gap in CTV that almost nobody prices, and the 2026 midterms are the first cycle where it matters at scale.
Roku disclosed that Q2 2026 political advertising on its platform exceeded the comparable quarter of the 2024 presidential cycle — a midterm quarter beating a presidential quarter — and that its political spend, like the industry’s, is weighted toward late Q3 and Q4.
Now hold that against the regulatory perimeter. The FCC’s political programming authority covers radio and television broadcast stations, cable television systems, direct broadcast satellite providers and satellite radio licensees. Reasonable access, equal opportunities, lowest unit charge and the online political file all attach to those categories. Streaming services are not among them.
The practical consequence is specific: a candidate ad running on a broadcast affiliate carries a lowest-unit-charge entitlement and a publicly inspectable file recording who bought what, when and for how much. The identical creative, targeted more precisely and delivered to the same household through a connected-TV app, carries neither. As political dollars migrate to where the audience went, the public record of political spending thins out — not because anyone repealed a disclosure rule, but because the rule was written for a delivery method that is now a minority of ad-supported TV time.
Alex Rivera: Media buyers should read the FCC scope as a pricing fact before reading it as a policy fact. On broadcast, the lowest-unit-charge window caps what a station can charge a candidate in the 45 and 60 days before an election. On CTV, there is no cap, no window and no file. That is precisely why political CTV inventory is going to clear at rates broadcast cannot legally match this autumn, and why Roku is telling shareholders to expect the back half of the year to be heavy.
Methodology
Collection. Every figure in this article traces to a primary document fetched on August 26, 2026: Roku’s Form 8-K Exhibit 99.1 filed with the SEC on August 6, 2026; Nielsen’s Q1 2026 Ad Supported Gauge (published June 23, 2026) and May 2026 Gauge and Media Distributor Gauge reports (published July 28, 2026); the IAB 2026 Digital Video Ad Spend & Strategy Report announcement of May 5, 2026; the IAB/PwC Internet Advertising Revenue Report Full Year 2025 announcement of April 16, 2026; Magnite’s Q2 2026 results release of August 5, 2026; The Trade Desk’s Q2 2026 results release of August 6, 2026; Netflix’s Q2 2026 shareholder letter as filed on Form 8-K; the FTC’s September 2024 6(b) staff report; and the FCC Media Bureau’s political programming scope. No aggregator, market-research resale or secondary summary was used for any figure.
SHAY formula. SHAY = advertising revenue ÷ (streaming hours ÷ 1,000), computed per platform per quarter. Q2 2026 reading: $17.7524 per 1,000 hours. Q2 2025 reading: $15.2186. The Q2 2025 hours denominator (35.42 billion) is derived from Roku’s disclosed 37.9 billion hours and stated 7% year-over-year growth; it is flagged as an Axis calculation in the dataset. Validation: multiplying hours growth by SHAY growth reproduces Roku’s reported advertising revenue growth to the second decimal (1.070 × 1.1665 = 1.2481 against 24.81%).
ASST formula. ASST = (streaming share of ad-supported TV) × (ad-supported share of total TV). Q1 2026 reading: 46.6% × 73% = 34.02%. Because Nielsen states the ad-supported share as “nearly 73%,” the reading is published as a ceiling and rounded to approximately 34% in prose.
CGP formula. CGP = dollar-weighted year-over-year growth of the CTV panel minus year-over-year growth of the broad digital advertising panel, same quarter. Q2 2026: 26.12% − 20.50% = 5.62 points. CTV panel: Roku advertising revenue and Magnite CTV contribution ex-TAC. The digital benchmark is the canonical Q2 2026 figure from the Axis Intelligence Research digital marketing panel.
Derived values. Five figures are Axis Intelligence Research calculations rather than direct disclosures, each carrying a method note in the dataset: Roku Q2 2025 streaming hours (35.42B), Magnite Q2 2025 CTV contribution ex-TAC ($71.4M), CTV share of Magnite total contribution ex-TAC (51.21% and 44.07%), the implied shares of total TV for broadcast and cable, and the SHAY, ASST and CGP readings themselves.
Currency and periods. All financial figures are U.S. dollars as reported. Quarters are calendar quarters ended June 30, 2026. Nielsen Gauge intervals follow the broadcast calendar; the May 2026 interval spanned April 27 through May 31, 2026. Nielsen notes that the Gauge and Media Distributor Gauge reflect total TV viewing rather than the currency ratings used for advertising sales, and that Gauge reporting has not yet migrated to ARF DASH-based universe estimates.
Verification. All arithmetic was recomputed in two independent passes, floating-point and exact-fraction, outside the drafting process. Every numeric token in this article was matched against a dataset row before publication.
About This Dataset
connected-tv-advertising-statistics-2026.csv contains 109 observations covering connected-TV advertising revenue and margins, streaming hours, platform and distributor share of U.S. TV watch-time, ad-supported viewing share, programmatic CTV supply and demand metrics, digital video ad spend projections, buyer criteria and agentic AI adoption, and the regulatory perimeter for political advertising. Each row carries source_org, source_document, source_url, retrieved_date, is_primary, axis_calculated and method_note.
Coverage: Q2 2025 through Q2 2026 for platform financials; Q1 2026 for ad-supported viewing; May 2026 for distributor share; full-year 2025 and 2026 projections for category spend.
Geography: United States for viewing, spend and regulatory data; global for Roku and Magnite corporate financials.
License: CC BY 4.0. Free to use, share and adapt with attribution.
Citation line: Axis Intelligence Research, Connected TV Advertising Statistics 2026, 2026.
Cite This Research
APA Axis Intelligence Research, & Rivera, A. (2026, August 26). Connected TV advertising statistics 2026: Ad yield, platform share and spend data. Axis Intelligence. https://axis-intelligence.com/connected-tv-advertising-statistics/
MLA Axis Intelligence Research, and Alex Rivera. “Connected TV Advertising Statistics 2026: Ad Yield, Platform Share and Spend Data.” Axis Intelligence, 26 Aug. 2026, axis-intelligence.com/connected-tv-advertising-statistics/.
Chicago Axis Intelligence Research and Alex Rivera. “Connected TV Advertising Statistics 2026: Ad Yield, Platform Share and Spend Data.” Axis Intelligence, August 26, 2026. https://axis-intelligence.com/connected-tv-advertising-statistics/.
Copy any format above and paste directly.
Frequently Asked Questions
Is SHAY the same thing as a CTV CPM?
No. A CPM prices one thousand ad impressions; SHAY prices one thousand hours of viewing across an entire platform, including hours the platform does not monetize with its own inventory. The two move together but measure different things. SHAY is useful precisely because CPMs are negotiated privately and never disclosed, while both SHAY inputs sit in an SEC filing anyone can open. Treat SHAY as the direction and magnitude of platform-level ad inflation, and your own account data as the price.
Why does streaming’s 48.6% share of TV overstate what a CTV buyer can reach?
Because Nielsen’s Gauge counts every streamed minute, ad-free subscription tiers included. An hour of ad-free Netflix and an hour of The Roku Channel both land in that 48.6%, but only one of them has inventory. Nielsen’s Ad Supported Gauge separates them: streaming holds 46.6% of ad-supported TV, and ad-supported viewing is nearly 73% of all TV, which puts genuinely addressable streaming at roughly 34% of total American TV time.
If CTV is growing 11% and social video 13%, should budget shift to social?
Not on those two numbers alone. Category growth rates tell you where incremental industry dollars are flowing, not where a given advertiser’s next dollar performs best. What the 2026 IAB reading does rule out is the argument that CTV is the uncontested fastest-growing video channel — for the first time it is not. The defensible read is that CTV’s growth is now tied to live sports rights, which are lumpy and concentrated, while social video’s is tied to creator supply, which expands at effectively zero marginal cost to the platform.
Why is Magnite’s CTV segment growing 36% while The Trade Desk’s total revenue grows 3%?
They sit on opposite sides of the same auction and count different things. Magnite’s figure is streaming inventory it represents on the sell side, where supply is expanding as more services turn on ad tiers. The Trade Desk’s figure is total platform revenue across every channel including open-web display, which is where the softness is. Their CTV and audio channels grew double digits in the same quarter. A single-channel growth rate and a blended company growth rate are not comparable measurements.
Do FCC lowest-unit-charge rules protect political advertisers buying connected TV?
No. FCC political programming authority — reasonable access, equal opportunities, lowest unit charge and the online political file — attaches to broadcast stations, cable systems, DBS providers and satellite radio licensees. Streaming and CTV apps sit outside that perimeter. A candidate buying broadcast in the 45 days before a primary gets a statutory rate cap and a public file entry; the same candidate buying the same household through a CTV app gets neither, which is one reason political CTV inventory clears above regulated broadcast rates.
What share of CTV ad spend now moves programmatically through third-party DSPs?
On Roku, nearly three-quarters of in-stream video ad spend arrived through third-party demand-side platforms in Q2 2026, including Amazon DSP, The Trade Desk, Google DV360 and Yahoo, with Magnite, Google Ad Manager and FreeWheel on the supply side. That is one platform’s disclosure rather than an industry figure, but it is the largest CTV platform that publishes the split, and it establishes that owning the operating system no longer means owning the transaction.
Does the Fox acquisition of Roku change CTV inventory pricing?
It changes who controls it, and the transaction has not closed. Fox and Roku announced a definitive agreement on June 15, 2026 valuing Roku at approximately $22 billion, with closing expected in the first half of 2027. Roku suspended earnings calls and financial guidance because of the pending deal, which means the SHAY series loses its forward outlook even though the historical inputs continue to be filed. The combination would place a major broadcast rights holder in direct control of a home screen used by more than half of U.S. broadband households.
Where does YouTube sit in connected TV statistics — is it CTV or social video?
Both, depending on who is counting, and that is a real measurement problem rather than a definitional quibble. Nielsen counts YouTube viewing on a television screen in the Gauge, where it held 13.8% of all U.S. TV watch-time in May 2026 — the largest share of any distributor for three straight months. The IAB classifies YouTube inside social video for spend purposes. The same viewing hour can therefore appear as television in an audience report and as social in a spend report, which is one reason CTV market-size figures from different publishers rarely reconcile.
How reliable is the Q2 2025 streaming-hours figure used in the SHAY comparison?
It is derived, not disclosed, and flagged as such. Roku reported 37.9 billion streaming hours in Q2 2026 and stated growth of 7% year over year; 35.42 billion is what that implies for Q2 2025. Because Roku rounds the growth rate to a whole percentage point, the derived denominator carries the rounding, and the Q2 2025 SHAY reading of $15.2186 should be read as accurate to roughly a cent. The direction and rough magnitude of the 16.65% yield increase survive the full plausible rounding range.
Related Axis Intelligence research: Cord-Cutting Statistics 2026 · Digital Marketing Statistics 2026 · Google Ads Statistics 2026 · Facebook Ads Statistics 2026 · YouTube TV Statistics 2026 · Tubi Statistics 2026