From the archive / Audience & data

Deloitte's 2025 survey found young viewers rank social first

A vendor survey of 3,595 US consumers found Gen Z favours social video discovery over streaming, self-reported and US-only.

Visual published with the cited source for this record: Deloitte's 2025 survey found young viewers rank social first
Visual published with the cited source, shown for identification of the record. Credit: media.deloitte.com. Source ↗ · Image record · Owner review pending.

What was announced or recorded

On 25 March 2025, the professional services firm Deloitte published 2025 Digital Media Trends, the nineteenth edition of a survey it commissions annually on US media habits. Deloitte states the fieldwork ran in October 2024 among 3,595 US consumers, with results weighted to recent Census figures. This is a vendor survey, not a metered panel: every figure in it is a respondent's own report of behaviour, gathered once and then published five months later.

What the documents establish

The report states that Generation Z respondents spend 54% more time on social platforms and user-generated video than the average consumer, and 26% less time on TV and film, roughly 50 minutes more and 44 minutes less per day by Deloitte's own framing. It reports that 56% of Gen Z say social media content is more relevant to them than traditional TV and film, that 56% of younger respondents have watched a show or film on a subscription service after hearing about it from a creator online, and that two-thirds of younger respondents use free ad-supported streaming. A companion page, the Digital Media Monitor, describes the underlying programme as a Census-representative survey run every six months among US consumers aged fourteen and over, and states plainly that its figures are 'survey-based data, not measured behavioral data.' That caveat, from Deloitte itself, is the load-bearing fact for anyone citing the headline numbers.

Announced versus delivered

There is no plan-versus-result gap here, because a survey wave is a completed act of measurement, not a promise. The real distinction is between what the questions can show and what they cannot: self-reported minutes are recall estimates, not logged time, and a single US sample says nothing about audiences elsewhere. The finding that younger viewers discover shows through creators before watching them on a subscription service supports a pipeline from short-form clips to paid platforms, but the survey cannot say what share of that discovered viewing ends up as a vertical-drama app purchase rather than a film or a series.

What to watch

Editorially, the number worth tracking across future editions is whether Deloitte starts asking separately about paid short-episode apps, rather than folding them into 'social platforms and user-generated content.' Until it does, a reader cannot isolate vertical drama's share of the behaviour the survey is describing.

  • Is a cited figure from a metered panel or a self-report survey, and does the source say so itself?
  • Does the sample cover the audience a claim is being applied to, or only US consumers of one age range?
  • Does 'social video' in the finding include paid episodic apps, or only ad-funded clip platforms?

Read as what it is, a commissioned annual survey with its own stated caveats, the 2025 edition is good evidence that younger US respondents describe social discovery as central to their viewing; it is not a measured audience count, and it draws no line around vertical drama specifically.

Sources & evidence

2025 Digital Media Trends: Social platforms are becoming a dominant force in media and entertainment ↗

www.deloitte.com · Official source
Source date: 25 Mar 2025 · Checked: 16 Sept 2026

  • States the report's publication date, sample size, survey period and the Gen Z/millennial social-versus-streaming findings.
Digital Media Monitor ↗

www.deloitte.com · Official source
Source date: Not stated by source · Checked: 16 Sept 2026

  • Deloitte's own description of the survey programme's cadence, sample and its statement that the figures are self-reported, not measured, data.