The Data Behind the Personas
Last updated 2026-09-09 · 3 studies · 45 verified evidence rows
Everything the persona studio reads from, printed in full so you can check its arithmetic rather than trust it. Every cell below is a published figure. Nothing here is modelled, combined or inferred: the combining happens in the tool, and it shows its working there.
Every measured figure, one surface
Pick a question, pick a lens. Platforms, how people get online, where they get news including AI chatbots, and the national baseline underneath all of it. Every cell is a published figure with its source and sample size attached, and where a source does not publish a cut it says so rather than filling the gap.
Who ever uses each social platform.
| All | 18 to 29±5.6 | 30 to 49±3.4 | 50 to 64±3.6 | 65 and over±3 | |
|---|---|---|---|---|---|
| YouTube | 84% | 95%+11 | 92%+8 | 85% | 64%-20 |
| 71% | 68%-3 | 80%+9 | 74%+3 | 57%-14 | |
| 50% | 80%+30 | 62%+12 | 40%-10 | 19%-31 | |
| TikTok | 37% | 63%+26 | 44%+7 | 30%-7 | 12%-25 |
| 36% | n/p | n/p | n/p | n/p | |
| 32% | n/p | n/p | n/p | n/p | |
| 32% | 37%+5 | 40%+8 | 30% | 20%-12 | |
| 26% | 48%+22 | 35%+9 | 16%-10 | 6%-20 | |
| Snapchat | 25% | 58%+33 | 31%+6 | 13%-12 | 4%-21 |
| X | 21% | 33%+12 | 25%+4 | 16%-5 | 10%-11 |
| Nextdoor | 13% | n/p | n/p | n/p | n/p |
| Threads | 8% | 15%+7 | 10% | 6% | 3%-5 |
| Bluesky | 4% | 6% | 5% | 3% | 2% |
| Bereal | 3% | n/p | n/p | n/p | n/p |
| Truth Social | 3% | 1% | 3% | 5% | 4% |
Social Media Fact Sheet · published 2025-11-20 · n=5,022 · ±1.9pp overall · “n/p” means that source does not publish that cut. Nothing on this surface is estimated.
One lens at a time
Pick a single dimension and read the published cells straight. Add a second and it tells you no study measured the overlap, which is the truthful answer at this level. The studio on the main page is where combinations get estimated, with the arithmetic shown.
Build an audience · Data responds live
All US adults · ever use each platform
n=5,022 · margin of error ±1.9pp
Every figure: Pew Research Center, Americans' Social Media Use 2025, surveyed Feb 5 to June 18 2025, n=5,022. The right-hand column is the difference from all US adults. Bars are the share who say they EVER use a platform, which is reach, not time spent.
Which group uses which platform
Eight platforms across gender, age, income and education. Bar length is the value, the small figure beneath it is the gap from the national average, and each column carries its own margin of error.
| Platform | All adults | 18 to 29±5.6 | 30 to 49±3.4 | 50 to 64±3.6 | 65 and over±3.0 |
|---|---|---|---|---|---|
| YouTube | 84 | 95%+11 | 92%+8 | 85%+1 | 64%-20 |
| 71 | 68%-3 | 80%+9 | 74%+3 | 57%-14 | |
| 50 | 80%+30 | 62%+12 | 40%-10 | 19%-31 | |
| TikTok | 37 | 63%+26 | 44%+7 | 30%-7 | 12%-25 |
| 32 | 37%+5 | 40%+8 | 30%-2 | 20%-12 | |
| 26 | 48%+22 | 35%+9 | 16%-10 | 6%-20 | |
| Snapchat | 25 | 58%+33 | 31%+6 | 13%-12 | 4%-21 |
| X | 21 | 33%+12 | 25%+4 | 16%-5 | 10%-11 |
| Threads | 8 | 15%+7 | 10%+2 | 6%-2 | 3%-5 |
| Bluesky | 4 | 6%+2 | 5%+1 | 3%-1 | 2%-2 |
| Truth Social | 3 | 1%-2 | 3%0 | 5%+2 | 4%+1 |
% of US adults who say they EVER use each platform. Bar length is the value; the small figure beneath is the difference from all adults. ± is the published margin of error for that column. Pew Research Center, Americans' Social Media Use 2025, n=5,022, surveyed Feb 5 to June 18 2025. Columns are read independently: there is no published figure for any combination of two columns.
The same eight platforms by race and ethnicity
Published in the same survey, kept in its own table because the sample sizes are much smaller and the error bars much wider. The Asian sample is 211 people at plus or minus 8.9 points, so a nine-point difference in that column may be nothing at all. Pew notes its Asian adult estimates represent English speakers only.
- White
- Adults who report being only one race, White, and are not Hispanic. Pew's own category, quoted rather than renamed.
- Black
- Adults who report being only one race, Black, and are not Hispanic. Pew's own category and capitalisation.
- Hispanic
- Hispanic adults of any race, so this group overlaps none of the others by construction and is not a residual.
- Asian
- Adults who report being only one race, Asian, and are not Hispanic. Pew states these estimates represent English speakers only, so non-English-speaking Asian adults are absent from this column.
| Platform | Whiten=3,304 ±2.3 | Blackn=512 ±6 | Hispanicn=757 ±5 | Asiann=211 ±8.9 |
|---|---|---|---|---|
| YouTube | 82% | 85% | 88% | 92% |
| 70% | 74% | 74% | 62% | |
| 45% | 54% | 62% | 58% | |
| TikTok | 28% | 53% | 57% | 31% |
| 23% | 37% | 56% | 54% | |
| 27% | 18% | 22% | 44% | |
| Snapchat | 24% | 29% | 31% | 19% |
| X | 18% | 26% | 23% | 32% |
| Threads | 6% | 18% | 10% | 12% |
| Bluesky | 4% | 2% | 4% | 3% |
| Truth Social | 4% | 2% | 2% | 6% |
Where they live, and how they lean
Two dimensions the same survey publishes that most persona work ignores. Community type separates WhatsApp more sharply than income does: 44% urban against 17% rural. Party barely moves YouTube or Snapchat, and almost entirely determines Bluesky and Truth Social.
| Platform | Urbann=1,394 ±3.6 | Suburbann=2,334 ±2.8 | Ruraln=1,235 ±3.8 | Rep / lean Repn=2,234 ±2.8 | Dem / lean Demn=2,446 ±2.8 |
|---|---|---|---|---|---|
| YouTube | 85% | 87% | 79% | 84% | 85% |
| 71% | 71% | 71% | 72% | 70% | |
| 55% | 54% | 37% | 49% | 53% | |
| TikTok | 42% | 37% | 32% | 33% | 40% |
| 44% | 34% | 17% | 25% | 38% | |
| 29% | 30% | 18% | 22% | 32% | |
| Snapchat | 25% | 26% | 23% | 26% | 25% |
| X | 23% | 22% | 17% | 24% | 19% |
| Threads | 11% | 9% | 4% | 6% | 10% |
| Bluesky | 4% | 5% | 2% | 1% | 8% |
| Truth Social | 2% | 4% | 4% | 6% | 1% |
How they get online at all
The layer that decides what you can build, not just where to show up. 16% of US adults are smartphone-only: they own a smartphone and have no home broadband. That is 34% of adults in households under $30,000, 28% of Hispanic adults and 27% of those with a high school education or less. Blank cells mean that fact sheet does not publish that cut, which is why the party columns are empty for internet and broadband.
| Measure | All | 18 to 29 | 30 to 49 | 50 to 64 | 65 and over | White | Black | Hispanic | Asian | Under $30K | $30K to $70K | $70K to $100K | $100K and over | High school or less | Some college | College graduate | Urban | Suburban | Rural |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Uses the internet | 96% | 99% | 99% | 96% | 90% | 96% | 94% | 97% | 99% | 91% | 96% | 98% | 99% | 93% | 98% | 99% | 96% | 98% | 94% |
| Has home broadband | 78% | 71% | 87% | 79% | 70% | 81% | 71% | 68% | 86% | 54% | 75% | 88% | 94% | 62% | 81% | 92% | 75% | 84% | 71% |
| Owns a smartphone | 91% | 97% | 96% | 90% | 78% | 91% | 85% | 93% | 96% | 82% | 89% | 96% | 97% | 84% | 93% | 96% | 91% | 93% | 87% |
| Owns a cellphone of any kind | 98% | 99% | 99% | 98% | 95% | 98% | 98% | 98% | 98% | 95% | 98% | 99% | 99% | 97% | 98% | 99% | 97% | 99% | 97% |
| Owns a cellphone but not a smartphone | 7% | 2% | 3% | 7% | 16% | 7% | 12% | 5% | 3% | 13% | 9% | 3% | 2% | 13% | 5% | 2% | 7% | 5% | 9% |
| Smartphone-only: owns a smartphone but has no home broadband | 16% | 27% | 11% | 15% | 17% | 13% | 19% | 28% | 11% | 34% | 19% | 10% | 4% | 27% | 15% | 6% | 19% | 12% | 20% |
Which platforms they open every day
Reach says a platform can be used. Daily use says it is a habit, and it is the closer answer to where an audience actually spends time. A different survey with a different sample, so it is kept separate rather than blended. The reversals are the interesting part: men out-use women on YouTube daily 58 to 39, while women lead Facebook 60 to 44.
| Group | YouTube | TikTok | X | |
|---|---|---|---|---|
| All US adults | 52% | 48% | 24% | 10% |
| 18 to 29 | 49% | 66% | 47% | 18% |
| 30 to 49 | 58% | 54% | 28% | 10% |
| 50 to 64 | 54% | 45% | 17% | 9% |
| 65 and over | 45% | 27% | 5% | 4% |
% who visit or use each platform at least once a day, combining "several times a day" and "about once a day". A SEPARATE survey from the reach figures above: 5,123 US adults, February 24 to March 2 2025. Only these four platforms were asked about. Estimates for Asian adults represent English speakers only.
The country a persona lives in
No persona is credible floating free of its baseline. A household earning $120,000 reads differently once you know the median is $81,600. Retrieved from USAFacts, which harmonises 70-plus federal agencies, with the agency of record named on every row.
Money
Women earned $0.82 for every dollar men made in a typical week
Gender pay gap · Q2 2026
Bureau of Labor Statistics, via USAFactsHousehold
65.2% of households own their home, about 2 in 3
Homeownership rate · 2025
US Census Bureau, via USAFactsabout $1,487 per month
Median rent, including utilities · 2024
US Census Bureau and Department of Housing and Urban Development, via USAFacts835,900, which is 1.3% of all married-couple households
Households led by a same-sex married couple · 2024
US Census Bureau, via USAFactsPeople
74.1 million, which is 23%
People aged 5+ speaking a language other than English at home · 2024
US Census Bureau, via USAFactsWork
USAFacts publishes 94 distinct questions, each expanded across states and counties into roughly 29,500 answer pages. Eleven were taken here, selected for what a persona actually needs. The remainder is a known, mapped source for later passes rather than an unexplored pile. Collected by direct fetch on 2026-09-09; usafacts.org permits ClaudeBot in its robots.txt and every figure above appears in the served HTML, so no paid scraper was used or needed.
Which source answers which layer
Four of the six layers below are answered by free public research, most of it federal. Paid tooling is justified for collecting what people say and type, never for demographics, and nothing at any price answers the last row.
| Layer | The question | Source | Cost |
|---|---|---|---|
| Who they are | Age, gender, income, education, race, community | Pew Research Center, US Census, American Community SurveyProbability samples with published margins of error. No commercial audience tool improves on this for US demographics, and most are inferring what Pew measures. | free |
| How they spend time | Hours per day on an activity, by demographic | American Time Use Survey (BLS)Federal time-diary survey, roughly 6,100 respondents in 2025, microdata downloadable at no cost. A stronger answer to time spent than any commercial panel, because it is a diary rather than a recollection. | free |
| What they pay | Spending by category and demographic | Consumer Expenditure Survey (BLS)Annual means, standard errors and shares across 17 demographic characteristics, plus public-use microdata back to 1980. This is the salary and spending layer, already published. | free |
| What they type | Search demand, queries, SERP composition | DataForSEOJustified: no free source gives query-level volume at scale. Keyword volume understates real demand, so pair it with Search Console rather than treating it as the whole picture. | paid |
| What they say | Posts, comments, community threads | mcpscraper, Bright DataJustified for collection only. These retrieve text that is already public; they do not supply demographics, and nothing they return establishes why somebody decided anything. | paid |
| Why they decided | Motivation, trust, trade-offs | Purpose-built surveys and interviewsNo tool at any price answers this. Behavioural data shows where a person went, never why. This is the layer where every persona on the internet quietly starts inventing, and where ours marks traits inferred. | free |
What changed, and what was thrown out
2026-09-09 · Pew fact sheet corpus
- Pulled all three Pew Internet & Technology fact sheets through the machine-readable .md endpoints their robots.txt publishes for AI clients. 44 tables extracted structurally rather than read by eye.
- Reconciliation first: all 136 reach cells already shipped were compared against the fresh pull. 136 of 136 matched exactly, zero drift. That is what licensed expanding from this source rather than re-deriving from the PDF.
- Reach grid went from 136 cells to 242, now dense: 11 platforms by 22 segments with no gaps.
- Added Threads (8%), Bluesky (4%) and Truth Social (3%), first published in the 2025 wave.
- Added two dimensions Pew publishes that were not being used: community type (urban, suburban, rural) and party lean.
- Added the access layer: internet use, home broadband, cellphone and smartphone ownership, and smartphone dependency. 16% of US adults are smartphone-only, rising to 34% under $30,000 and 28% among Hispanic adults.
- Every sample size and margin of error re-verified against Pew's own methodology table, which is published as an image. All 17 previously shipped figures confirmed exact; five new segments added from it.
- Access-layer figures are deliberately NOT combined across traits. They sit near the ceiling, where multiplying odds adds error without adding information, so each is shown as published.
2026-09-09
- First published. Three studies, 42 verified evidence rows, one persona.
- Pew age breakdowns verified across two independent reads and reconciled against published overall figures.
- Pew gender breakdowns and WhatsApp overall excluded after failing that reconciliation; recorded in the do-not-assert list.
- Re-verified the same day against the primary report PDF (Americans' Social Media Use 2025, published 2025-11-20) instead of the JavaScript fact-sheet table. The audit reversed the earlier conclusion: the OVERALL figures were the faulty read, not the gender splits. Corrected TikTok 32% to 37%, WhatsApp to 32%, Reddit 24% to 26%, Snapchat 26% to 25%. Gender data restored, corroborated verbatim by the report for Instagram (women 55%, men 44%).
- Added the full reach matrix (8 platforms x 13 segments across gender, age, income and education) with the published sample size and margin of error for every segment, to power the interactive explorer.
Figures that fail verification are recorded internally and never published, so a later refresh cannot quietly reintroduce a number that was already rejected once.
Sources
Every figure on this page and in the studio comes from one of the following. Each entry states what the study actually measured, its sample and field dates, and when it was last checked.
- [1]Americans' Social Media Use 2025 (full report, PDF)
Pew Research Center · accessed 2026-09-09
The primary source for every reach and daily-use figure on this page. Published November 20, 2025. Reach data: survey of 5,022 US adults, February 5 to June 18, 2025, margin of error ±1.9pp. Daily-use data: a separate survey of 5,123 US adults, February 24 to March 2, 2025. Read directly from the PDF after the interactive fact sheet produced figures that would not reconcile.
- [2]Social Media Fact Sheet
Pew Research Center · accessed 2026-09-09
The full crosstabs for the same survey: 11 platforms by age, gender, race, income, education, community type and party, plus the 2012-2025 trend. Read through Pew's machine-readable .md endpoint, which its robots.txt publishes for AI clients. All 136 cells that overlapped the PDF extraction matched exactly, which is what licensed the rest of the grid being taken from here.
- [3]Internet, Broadband Fact Sheet
Pew Research Center · accessed 2026-09-09
Internet use and home broadband subscription, each by age, race, gender, income, education and community type, with trend lines back to 2000. Published November 20, 2025 from the same NPORS survey. The source for the access layer: 96% of US adults use the internet, 78% subscribe to home broadband.
- [4]Mobile Fact Sheet
Pew Research Center · accessed 2026-09-09
Cellphone and smartphone ownership by seven dimensions, and smartphone dependency, defined as owning a smartphone without a home broadband subscription. Published November 20, 2025. The source for the smartphone-only figure of 16% nationally, which rises to 34% among adults in households under $30,000.
- [5]Social Media Use 2025: Methodology
Pew Research Center · accessed 2026-09-09
Where every sample size and margin of error on this site comes from. Total sample 5,022 at ±1.9pp; subgroups range from White adults at n=3,304 (±2.3pp) to Asian adults at n=211 (±8.9pp). Pew publishes this table as an image, so it was read from the image rather than parsed.
- [6]National Public Opinion Reference Survey (NPORS) methodology
Pew Research Center · accessed 2026-09-09
Address-based sampling with a web, mail and phone protocol. 2,349 completed online, 2,331 on paper and 342 by phone. The published unweighted sample size and margin of error for every segment on this page comes from its appendix, which is why each column carries its own ± figure.
- [7]Answers (94 questions, harmonised from 70+ federal agencies)
USAFacts · accessed 2026-09-09
Nonpartisan nonprofit founded by Steve Ballmer in 2017, which collects and standardises data from more than 70 federal agencies and names the agency behind each figure. Used here for the national baseline: income, wages, debt, rent, homeownership, poverty, population and household composition. Every row carries the originating agency as well, because the agency is the source of record and USAFacts is the route. Retrieved by direct fetch, permitted by their robots.txt.
- [8]Digital 2026 Global Overview Report
DataReportal, with GWI panel data · accessed 2026-09-09
Source of the time-per-day figures. Panel-based, self-reported, global and all-user, so it is graded derived rather than measured wherever it appears against a US persona.
- [9]Car Buyer Journey Study, 16th annual (summary PDF)
Cox Automotive · accessed 2026-09-09
Source of every vehicle-purchase figure. Surveyed 2,300 US consumers who bought a new or used vehicle in the prior 12 months, fielded autumn 2025, published January 13, 2026. Measures the journey for buyers as a whole and is not cross-cut by platform use, parental status or model.
No paid audience-intelligence tool was used. Everything here is public research, read at the source rather than through an aggregator.
One page per question
Which platforms they use2025 data
How many people use them2025 data
Where they get news2025 data