The Attention Landscape · established evidence
Money Buys Broadband Before It Buys a Platform: Income as the Hidden Segmentation Variable
Local-marketing strategy defaults to a geographic frame: reach the urban customer one way, the rural customer another. The data does not support geography as the primary divide. Pew Research Center puts home broadband subscription at 73% in rural households versus 86% in suburban ones, a 13-point gap. Broken out by income instead, the gap is nearly three times as wide: 57% of households earning under $30,000 a year subscribe to broadband, versus 95% of households earning over $100,000, a 38-point spread. Income, not geography, is the variable that most gates whether a household is online at all, and it keeps gating which platforms that household uses once it is. This piece lays out the evidence, established and specific, and the practical consequence: a business that segments its reach strategy by zip code alone is measuring the wrong axis, and is likely misallocating budget between audiences that were never the same reachability problem to begin with.
Two ways to slice the same population
Local and regional marketing plans routinely reach for one segmentation frame first: geography. Urban markets get one channel mix, suburban another, rural a third, on the working assumption that place is what determines how reachable an audience is online. It is a convenient assumption, because geography is easy to buy media against: zip codes, designated market areas, and store trade radii all come pre-built into ad platforms.
The population-level survey data does not support treating geography as the primary access variable. There is a second, cross-cutting axis, household income, and when the two are measured against the same outcome, the income gap is considerably larger than the geography gap. That is not a minor correction. It means a segmentation model built on place alone is optimizing against a smaller effect while ignoring a bigger one.
The income gap dwarfs the geography gap
Pew Research Center's national surveys give both numbers directly, measured against the same outcome: home broadband subscription. By geography, broadband subscription runs 73% in rural households, 77% in urban households, and 86% in suburban households, a spread of 13 points between the lowest and highest reading.
By income, the same outcome runs 57% for households earning under $30,000 a year versus 95% for households earning over $100,000, a spread of 38 points. The income-based gap is close to three times the size of the geography-based gap. If a business is choosing where to concentrate reach based on urban versus rural labels, it is reading the smaller of the two available signals.
Access is the floor. Adoption is the next fracture, and it is income-shaped too.
Having broadband is a precondition, not the whole picture. Once a household is online, which platforms it actually uses tracks income and education independently of age. Pew's social-media survey found YouTube adoption reaches 89% among households earning $100,000 or more, and Instagram adoption is likewise highest in that same top income bracket, at 58%.
The steepest gradient of any major platform measured runs by education rather than income directly: LinkedIn usage is 53% among adults with a bachelor's degree or higher, versus 10% among adults with a high school education or less. That is a second, compounding fracture layered on top of the access question. A campaign built to reach "everyone online" in a given zip code can still miss most of a platform's real audience if it does not account for the income and education skew inside that access.
Why the urban and rural frame persists anyway
Geography persists as the default frame partly because it is operationally convenient and partly because it correlates loosely with income. Rural areas do, on average, have lower median household incomes than suburban ones, so a geography-based plan captures some of the income effect indirectly. But a loose correlation is a weaker instrument than the variable itself, and the weakness shows up at the edges that matter most for targeting decisions.
A higher-income rural household looks, in broadband subscription and platform adoption, considerably more like a higher-income suburban household than like a lower-income household in the same rural county. A lower-income urban household can, by the same logic, have more in common with a lower-income rural household than with its own wealthier urban neighbors. Geography groups people who live near each other. Income groups people who behave alike online. Those are not the same grouping, and conflating them is the specific error this evidence corrects.
What this changes for channel allocation
The practical shift is to route budget and channel choice by an income signal, even an imperfect proxy such as service price point or area median income, rather than by geography label alone. In markets that skew toward the lower end of the income distribution, expect the broadband and platform gaps to genuinely suppress the digital reach ceiling; a channel plan that assumes near-universal platform coverage there is planning against data that does not hold for that segment.
In higher-income segments, the adoption gradients run the other way: LinkedIn, Instagram, and YouTube all reach a majority-to-near-universal share of $100,000-plus households, and a channel plan can lean into those platforms with more confidence. The point is not that rural or lower-income audiences are unreachable; 57% broadband subscription is still a majority. The point is that the reachability ceiling and the platform mix both move with income first, and a plan built on geography alone is measuring a proxy instead of the variable that actually gates the outcome.
What the evidence does not establish
These are national, population-level figures from Pew Research Center, a disclosed-methodology source, and the tier here is established for the topline numbers themselves. What is not established is how precisely these national splits translate to any single local market or industry vertical; a specific med-spa, contractor, or law firm's actual buyer pool may skew differently from the national average in either direction, and the only way to know is to measure that market directly rather than assume the national figure applies unchanged.
The data also describes subscription and adoption, not depth of engagement or purchase intent. A household with broadband and an Instagram account is reachable on that platform; whether it is an in-market buyer for a given service is a separate question these figures do not answer. Read the income and education gradients as the floor on reachability, not as a complete audience model.
The evidence
Key findings, with their sources
-
Home broadband subscription is 57% among households earning under $30,000 a year, versus 95% among households earning over $100,000, a 38-point gap.
established Pew Research Center, "Internet use, smartphone ownership, digital divides in the U.S.", Jan 2026, and "Home Broadband and Mobile Use" survey, Jan 2024.
-
Home broadband subscription runs 73% in rural households, 77% in urban households, and 86% in suburban households, a 13-point spread, roughly a third the size of the income gap.
established Pew Research Center, "Internet use, smartphone ownership, digital divides in the U.S.", Jan 2026.
-
YouTube adoption reaches 89% among households earning $100,000 or more; Instagram adoption is likewise highest in the $100K-plus bracket, at 58%.
established Pew Research Center, "Americans' Social Media Use", Jan 31 2024.
-
LinkedIn usage is 53% among adults with a bachelor's degree or higher, versus 10% among adults with a high school education or less, the steepest education gradient of any major platform measured.
established Pew Research Center, "Americans' Social Media Use", Jan 31 2024.
Reference
Glossary
- Broadband subscription gap
- The difference in home broadband subscription rates between two population segments, here measured separately by geography and by household income.
- Income gradient
- A pattern in which a behavior or adoption rate rises steadily as household income rises, distinct from any effect of geography or age.
- Education gradient
- A pattern in which a behavior or adoption rate tracks educational attainment. LinkedIn shows the steepest measured education gradient of any major platform.
- Reachability ceiling
- The practical upper limit on how much of a given audience segment a digital channel can physically reach, before any question of messaging, offer, or creative.
Straight answers
Frequently asked questions
Does income really matter more than geography for internet access?
By the available Pew data, yes. The income-based gap in home broadband subscription (57% versus 95%, a 38-point spread) runs close to three times the size of the geography-based gap (73% versus 86%, a 13-point spread).
Does this mean rural markets are unreachable digitally?
No. 73% of rural households still subscribe to home broadband. The finding is not that rural audiences are unreachable, it is that income within any geography predicts reachability more precisely than the geography label alone.
How should a local business use this?
Route channel and budget decisions by an income signal, such as service price point or area median income, rather than by zip code or urban/rural label alone, and expect platform mix to shift with that signal independently of age.
Is this data specific to my industry?
No, it is general US population data. Translating it to a specific vertical or local market requires direct measurement of that market's actual buyers, which this piece flags as the next step rather than assuming the national split applies unchanged.
Provenance
Sources
- Pew Research Center, "Internet use, smartphone ownership, digital divides in the U.S.", Jan 8, 2026 (established)pewresearch.org
- Pew Research Center, "Home Broadband and Mobile Use" survey, Jan 2024 (established)
- Pew Research Center, "Americans' Social Media Use", Jan 31, 2024 (established)
Every figure above is attributed to a real, dated source and tagged with its evidence tier. Where a claim could not be verified to a primary source, it is not stated as fact.