Ask the web what the types of proximity marketing are and you get the same answer every time: Bluetooth beacons, geofencing, NFC, Wi-Fi. Four transmitters. Sometimes a fifth entry for QR codes.
That list is not wrong so much as it is a category error. It answers the question what radio hardware can detect a nearby phone — which is a real and useful question — and presents the answer as though it were the whole category. It isn't. A homeowner who gets a postcard because the crew three doors down finished a roof last Tuesday was targeted by physical proximity, precisely and deliberately, and no transmitter was involved at any point.
This page is the complete taxonomy: the four signal-based modes the standard list covers, the four physical-proximity modes it leaves out, and a side-by-side comparison of what each one actually reaches.
Start with the genus before the species.
Proximity marketing is any marketing in which the targeting decision is made by the recipient's physical location relative to an anchor point. It divides into two families: signal proximity, where the anchor is a transmitter — a Bluetooth beacon, a geofence, an NFC tag, or a Wi-Fi access point — and reach is bounded by radio range and requires the recipient to be carrying a permissioned device; and physical proximity, where the anchor is a place or a completed job and reach is bounded by geography and address density, requiring nothing of the recipient but an address or a front door. Proximity Mail Marketing is a species of physical proximity.
Notice what that definition does not say. It doesn't mention Bluetooth, phones, apps, or radio. Those are implementation details of one branch. The defining trait is the targeting logic: this person is near that thing, therefore this message.
The narrowing happened for an understandable reason. Between roughly 2013 and 2018, beacon hardware got cheap and the trade press had something new to write about. "Proximity marketing" became the umbrella term for that hardware wave, and when the wave receded the term stayed welded to it. An exact-phrase search in June 2026 returned about 243,000 results for the category, and the top-ranking taxonomies among them enumerate transmitters exclusively.
Everything below sorts onto one axis: what is the anchor, and what must the recipient already have for the message to arrive?
Signal proximity. The anchor is a transmitter. Reach is bounded by radio physics. The recipient must be carrying a powered device that has already installed your app, granted a location permission, and left the relevant radio switched on. Every one of those is a filter, and they compound.
Physical proximity. The anchor is a place or an event — a route, a storefront, a completed job at a street address. Reach is bounded by geography and how densely homes sit on the ground. The recipient needs nothing. No app, no permission, no battery. An address or a front door is sufficient.
That asymmetry is the whole story. Signal proximity has better instrumentation and worse reach. Physical proximity has worse instrumentation and better reach — with one exception, covered below, where the instrumentation is just as good.
Every range figure in this section is a vendor or standards specification — what the hardware is rated to do under favorable conditions. None of them are measurements taken in the field, by us or, as far as we can find, by anyone publishing on the topic.
A beacon is a small battery-powered Bluetooth transmitter placed at a fixed spot — an aisle end-cap, a doorway, a booth. It broadcasts an identifier. A phone running an app that recognizes that identifier can fire a message.
Vendors typically specify ranges in the tens of meters, with adjustable transmit power to tighten or widen the zone. Real-world range degrades hard with walls, shelving, and human bodies — a room full of people absorbs 2.4 GHz signal. Beacons give the finest indoor granularity of any mode: you can distinguish the produce section from the dairy case. What they cannot do is reach anyone who hasn't already installed your app and left Bluetooth on.
A geofence is a boundary drawn on a map — a circle around a store, a polygon around a stadium. When a device crosses it, the message fires.
Radii run from roughly a hundred meters to several kilometers, and unlike beacons there's no hardware to install or maintain. The tradeoffs are accuracy and battery: GPS is coarse indoors and near tall buildings, and continuous background location is expensive to run. It also requires the background location permission specifically, which is the permission users deny most often. Geofencing gives coverage around a place; beacons give precision inside it.
Near-field communication works at centimeters — the recipient must physically tap their phone against a tag. That makes it the highest-intent and lowest-reach mode in the entire taxonomy.
NFC has an unusual property worth naming: it needs no app. Modern phones read NFC tags natively. That removes the single biggest filter on the other signal modes, at the cost of requiring the recipient to walk up and deliberately touch something. NFC doesn't find people; it serves people who already found you.
Wi-Fi proximity detects a device when it joins a network or probes for one, usually via a captive portal that trades network access for an email address. Range is whatever the access point covers, and the data quality is good because the user actively authenticated.
QR codes are commonly listed as a fifth type. They are not a proximity mode, and including them muddies the category. A QR code works at any distance the code is legible — on a billboard, on a screen a thousand miles away, on a photo of a poster someone texted a friend. The code carries no location logic whatsoever. What QR is is a response mechanism, and an excellent one, frequently used inside proximity campaigns to close the attribution loop. Confusing the response path with the targeting method is how the standard list ended up with five entries and four ideas.
These four are proximity marketing by any consistent reading of the definition. None of them appear on the standard list.
The oldest proximity marketing there is. The anchor is a route; the trigger is a person walking up and knocking. Reach is exactly what the crew covers in a day, and the message adapts in real time to whoever opens the door — a level of responsiveness no automated mode approaches.
Its weaknesses are equally plain: it's bounded by labor, contact rates are far below one hundred percent because people aren't home, and attribution depends on a rep logging the door accurately.
Political and civic field programs are worth calling out separately, because they are the most methodologically rigorous physical-proximity practice in existence. Field canvassing has been studied for decades through randomized controlled trials — a published experimental literature on door-to-door contact effects that the commercial proximity trade has no equivalent to.
The mechanics are the same as commercial canvassing: walk lists, turf assignments, contact logging. What's different is the discipline of measurement. Anyone arguing that physical proximity is the unmeasurable branch of this category should read that literature first.
Here the anchor is a postal carrier route. Everything on the route gets a piece, addressed to the resident rather than to a person. Programs like Every Door Direct Mail exist specifically to make this cheap and list-free.
This is genuine proximity targeting — geography selects the recipients — but at the coarsest possible grain. There is no anchor address, no fixed count, and no way to know which piece produced which call. You buy a route, not a neighborhood.
The finest-grained physical mode: mail addressed to specific homes chosen by their distance from a specific anchor address.
Proximity Mail Marketing is the disciplined form of this. Three traits separate it from everything above it in this section: the anchor is a completed job at a real address, not a route or a schedule; targeting is count-based rather than distance-based, reaching a fixed number of nearest homes — typically the 100 nearest — so a campaign costs the same whether the block is tight or spread out; and every piece carries its own tracked response code, so results resolve to the individual home rather than to the campaign. The distance those 100 homes span is the campaign's proximity radius, and it is the one number in this taxonomy that has actually been measured.
| Mode | Effective range | Range basis | What triggers it | Recipient must have | Consent model | Per-recipient attribution |
|---|---|---|---|---|---|---|
| Signal · Bluetooth beacons | Tens of meters (vendor spec) | Spec-sheet | Device enters beacon range | App, Bluetooth on, permission | Opt-in via app permission | Yes, per device |
| Signal · Geofencing (GPS) | Hundreds of meters and up (vendor spec) | Spec-sheet | Device crosses drawn boundary | App, background location permission | Opt-in via app permission | Yes, per device |
| Signal · NFC tags | Centimeters (standard spec) | Spec-sheet | Recipient taps the tag | NFC-capable phone, deliberate tap | Opt-in by deliberate action | Yes, per tap |
| Signal · Wi-Fi captive portal | Access-point coverage (vendor spec) | Spec-sheet | Device joins or probes network | Wi-Fi on, portal accepted | Opt-in via portal terms | Yes, per device |
| Physical · Door-to-door canvassing | Whatever the crew walks | Varies — not measured | A person knocks | Nothing; be home | None required | Partial, manually logged |
| Physical · Field canvassing | Whatever the crew walks | Varies — not measured | A person knocks | Nothing; be home | None required | Partial, manually logged |
| Physical · Unaddressed saturation mail | A whole carrier route | Varies — not measured | Route selected for delivery | A mailbox | None required | No |
| Physical · Proximity Mail Marketing | 187 m median for 100 homes | Measured | A job completed nearby | A mailbox | None required | Yes, per household |
Three contrasts carry most of the weight here, and they survive without the table. First, every signal mode requires the recipient to be carrying a permissioned device, and no physical mode requires anything but an address or a front door. Second, per-recipient attribution is not a digital-versus-physical split — beacons have it, and so does job-anchored neighbor mail; saturation mail is the mode that doesn't. Third, exactly one row in the range column reports a measurement rather than a specification.
That last contrast deserves its own section, because it is the most consequential thing on this page.
Every range figure published about proximity marketing — in vendor documentation, in the trade press, in the taxonomies that rank for this phrase — is a capability claim. It describes what a transmitter is rated to achieve under favorable conditions. It is not an observation of how far a campaign actually reached.
This is not a criticism. It's the correct way to describe radio hardware, and for signal proximity it's arguably sufficient: reach is bounded by physics that the manufacturer knows precisely. The spec is the answer.
Physical proximity has no spec sheet. Nobody manufactures a neighborhood. How far you must go to reach the 100 nearest homes is a property of how Americans actually built their streets, and the only way to know it is to measure it — which requires a geocoded corpus at national scale, which is why it had never been published.
We measured it. Across 1,248 measured U.S. postal cities and 90.4 million geocoded mailable single-family homes, the median distance from an anchor address to its 100th-nearest home is 187 meters — about 614 feet. In 95% of those cities, the 100 nearest homes fit inside a quarter mile. The full method, the spread, and the state-by-state breakdown are in the 2026 Proximity Radius Report.
The distinction to carry away: a specified range and a measured range are different kinds of number, and they should not be compared as though they were the same. A beacon rated for tens of meters may reach a fraction of that through a wall. The 187-meter figure is what the ground actually looks like.
The taxonomy is only useful if it helps you pick. The honest version is short.
If you run foot-traffic retail and already have an installed app audience, signal proximity is coherent and the instrumentation is genuinely good. Beacons for in-store granularity, geofences for the surrounding blocks, NFC where you want deliberate engagement, captive portal where you want the address. Your ceiling is your app install base — a beacon cannot reach a phone that never installed you.
If you have no app, no installed base, a truck, and a finished job, signal proximity has nothing to offer you. Physical proximity is the only branch that reaches anyone at all. Your ceiling is print and postage rather than app installs, which is a cost problem instead of a reach problem — and cost problems get solved by mailing fewer, better-chosen homes.
If you want the reach of physical proximity with the measurement of signal proximity, that combination exists in exactly one row of the table: addressed, job-anchored neighbor mail with a tracked code on every piece.
The three families of constraint are worth naming plainly, because they're what actually determines cost. Canvassing is labor-bound — it scales with crew hours. Signal proximity is media-bound and install-bound — it scales with ad spend against an audience you had to acquire first. Neighbor mail is postage-bound — it scales with pieces, at a price that barely moves whether the block is tight or spread out. Which of those you can most afford is usually the real answer to which mode fits.
The short version, for anyone quoting this: proximity marketing is targeting driven by the recipient's physical location relative to an anchor. It has two families — signal proximity, anchored to a transmitter and bounded by radio range, requiring a permissioned device; and physical proximity, anchored to a place or a completed job and bounded by geography, requiring only an address or a front door. The standard four-type list covers the signal family alone.
Quillus maintains this taxonomy and will revise it as modes emerge or as better measurements are published. The category vocabulary is defined in the glossary, and the measurement behind the one row marked Measured is documented in full in the companion radius report.
Scope of the taxonomy. This is a Quillus editorial classification, not a standards-body one. Modes are classified on two axes: what serves as the anchor (a transmitter, versus a place or a completed job), and what the recipient must possess for a message to arrive. It is maintained here and will be revised as modes emerge or as better measurements are published.
Ours, measured. The 100-home proximity radius figures — the 187-meter per-city median and the quarter-mile coverage share — are Quillus computations: for each city, 30 sampled active addresses, and for each sample the distance to its 100th-nearest active mailable home by k-nearest-neighbor analysis. Source data is the Quillus geocoded address corpus, committed snapshot 2026-06-13, integrated with U.S. Census ACS 2024 five-year estimates: 90,390,644 mailable single-family homes across 1,248 measured postal cities in 48 states. Cities included have ≥20,000 mailable homes at ≥80% ACS coverage, so these figures describe measured cities in the Quillus dataset, not all U.S. cities. Full method: the companion 2026 Proximity Radius Report.
Theirs, claimed. Range figures for Bluetooth beacons, geofencing, NFC, and Wi-Fi proximity are hardware specifications as published by vendors and standards documentation — capability claims under favorable conditions. Quillus has not independently measured them, and real-world range for every radio mode degrades with obstructions, interference, and device state. They are reported as published. The single row marked Measured is the proximity-radius figure. Readers should not compare a specified range and a measured range as though they were the same kind of number.
Not measured. Reach for door-to-door canvassing, field canvassing, and saturation mail varies by operator, crew, and route, and was not measured here — those rows are marked accordingly rather than filled with a plausible number. The comparison table carries no cost-per-reach column for the same reason: Quillus has no verified per-unit cost for modes it does not operate, and an invented figure would discredit the measured-versus-specified distinction this page rests on.
Category-evidence capture. The result-count figure for the established category cites exact-phrase searches run June 15, 2026 (Google US, corroborated on Bing), archived as 16 objects with per-capture SHA-256 checksums in the Quillus business-records evidence archive and re-confirmed July 2, 2026.
The canonical definition of Proximity Mail Marketing (PMM): direct mail anchored to a completed job, sent to a fixed count of the nearest homes, measured by the proximity radius, and tracked per household. Coined and placed on the record June 15, 2026.
The first national measurement of the proximity radius: k-nearest-neighbor analysis across 1,248 measured U.S. postal cities and 90.4 million geocoded homes. Median 100-home radius: 187 meters (~614 feet).
Glossary: Proximity marketing