Connected TV statistics

Almost every CTV number you will meet falls into one of three families. The first is penetration: the share of households that own at least one connected device. Leichtman Research Group's household surveys put this at roughly 87% of US TV homes. Penetration tells you the size of the potential audience, but nothing about how much that audience watches.

The second family is viewing share: the proportion of total television time spent on streaming rather than broadcast or cable. Nielsen's monthly report, The Gauge, is the reference here. In its May 2025 edition streaming reached 44.8% of US TV viewing and, for the first time, exceeded broadcast and cable added together. Viewing share is the number that tells you where attention has actually moved.

The third family is platform-reported scale: account, household or user totals published by the companies that run the operating systems. Roku reports more than 90 million streaming households in its shareholder letters; Titan OS told TechCrunch in December 2025 that it had 18 million users. These figures are real disclosures, but they are measured by the platform itself, defined in the platform's own terms, and they are not comparable with each other without care.

Readers often notice that two respectable sources give different figures for what sounds like the same thing. Usually both are correct and they are measuring different units. A household that owns a smart TV, a Roku stick and a games console counts once in a penetration survey, but may count as three active devices in a manufacturer's numbers and as one, two or three accounts on different platforms.

Time periods differ too. Nielsen publishes monthly snapshots, and streaming share moves with the calendar — summer months, when broadcast seasons end, tend to lift it. A figure from a single month should be cited with that month attached. Annual surveys such as Deloitte's Digital Media Trends ask people about their behaviour, which captures intent and perception rather than metered minutes.

Geography is the final trap. Most widely quoted CTV statistics are American. European operating systems such as Titan OS, which ships on Philips, Sony, JVC, AOC and Vestel televisions according to EU-Startups and The Streaming Wars, operate in markets with different pay-TV structures. Applying a US viewing-share figure to Germany or Spain will mislead.

Media planners use viewing share to justify moving budget from linear television to streaming. Product teams at app publishers use penetration and platform scale to decide which operating systems to build for first. Investors read platform-reported households and engagement hours to judge whether an OS business can monetise its install base through advertising and revenue share.

Journalists and analysts use the same data to describe industry change — the shift in which Philips moved all of its 2026 televisions from Google TV to Titan OS, reported by heise online, is one example of a platform decision that only makes sense against the backdrop of CTV scale and advertising economics.

In every use, the right question is the same: does this number measure the thing my decision depends on? Penetration supports reach arguments. Viewing share supports attention arguments. Platform scale supports distribution arguments. Mixing them is how most errors in CTV presentations begin.

Nielsen's The Gauge splits total US television usage into streaming, cable, broadcast and other categories, and then breaks streaming down by distributor such as YouTube, Netflix and others. When you read a release, start with the category totals, because they answer the structural question — how much of TV time is streaming — and only then look at the distributor ranking, which answers a competitive question. Note that the denominator is all television usage on the TV set, which means time spent on phones and laptops is excluded. A service with large mobile viewing can therefore look smaller in The Gauge than its total audience suggests. Note the month every time; the May 2025 crossover figure of 44.8% is a specific moment in a seasonal series, and later months can be higher or lower without changing the underlying trend.

Outside the United States, CTV statistics are scattered across national audience measurement bodies, regulator reports and company disclosures. In Europe, independent television operating systems are a growing part of the picture: Titan OS reports 18 million users across Europe and Latin America, and heise online reported that all 2026 Philips models switched from Google TV to Titan OS. These numbers describe distribution — which software sits on the screen — rather than viewing share. When a European figure for streaming's share of viewing is needed, look first for the country's official audience measurement provider and its published methodology, and cite it with country and period. Avoid filling the gap with US numbers; pay-TV penetration, public-service broadcasting and language markets all differ, and the result would be misleading.

Every number on this page links to the organisation that published it: an independent research firm, an official product or pricing page, a public filing, or official documentation. We do not publish our own estimates as facts, and where a figure is reported by the company it describes, it is labelled company-reported. Prices are taken from official vendor pages and change often, so open the linked source before relying on one. If a widely repeated figure could not be traced to a primary source, we left it out. Entries are reviewed when the underlying sources publish new editions, and the update date at the top of the page shows the latest review.