Geofence Advertising Platforms: Costs, Capabilities, and CPVD as the Alternative

How geofence advertising platforms actually work

A geofence advertising platform is a software layer that sits between an advertiser and the inventory where ads can be served: mobile apps, mobile web, connected TV, programmatic DOOH screens. The platform doesn't own the screens. It owns the targeting logic, the bidder, and the relationships with the auction houses where inventory clears.

Three pieces have to line up for a geofence ad to deliver: a DSP (demand-side platform, where the advertiser configures the campaign and submits bids), an SSP (supply-side platform, where publishers list inventory), and a bid stream that carries each impression opportunity, including the device's reported location, from SSP to DSP in milliseconds. The geofence platform draws a polygon (a radius around a billboard, an address-level shape over a building footprint, a tunnel) and bids only on impressions where the bid-stream location falls inside it.

The audience layer rides on top. Platforms enrich the bid stream with mobile-ad-ID matches, foot-traffic history, household-level identity graphs, and third-party segments (Foursquare runs 1,500+ location segments and 800+ purchase-based segments off-the-shelf). The advertiser pays the platform; the platform pays the SSP, the data vendor, and any retargeting partner; the publisher gets what's left.

The major platforms in the category

Five names recur on almost every local-services agency's geofence shortlist. We name them so the reader can verify pricing and capability claims directly with each vendor.

• StackAdapt. Multi-channel programmatic DSP with display, native, video, CTV, audio, DOOH, and in-game inventory. Self-serve, popular with mid-market agencies. StackAdapt publishes neither a rate card nor a universal minimum, and no source at our sourcing bar publishes a fee percentage for it, so we quote none. Specialized inventory tiers, such as its early-access placements on newer ad surfaces, are priced separately and privately.
• The Trade Desk. The largest independent DSP and a public company (NASDAQ: TTD). Platform fees are negotiated rather than published, and the direct-seat minimums that circulate in agency roundups are not published by The Trade Desk and are not verifiable at our sourcing bar, so we do not repeat them. The practical point survives without a number: direct seats are sized for national advertisers, which is why most local service businesses reach TTD only through reseller agencies.
• Simpli.fi. Pioneered addressable geofencing by converting street addresses into address-shaped polygons (sourced from plat-line data and public land surveys) and serving mobile, video, and CTV ads to devices inside those shapes. Simpli.fi publishes no rate card and no minimum spend; the dollar figures attached to it online come from resellers and agency blogs, so this page carries none.
• GroundTruth. Location-based ads platform that pioneered cost-per-visit (CPV) pricing: advertisers can pay only when GroundTruth attributes a measurable in-store visit. No IO contract or minimum spend on the self-serve Ads Manager. Recently expanded into programmatic DOOH via Place Exchange with foot-traffic attribution. CPV pricing varies by industry; fast-food visits clear cheaper than premium-grocery visits.
• Foursquare (Pinpoint). Location-intelligence company that turned its 105M+ places dataset and 150M+ unique-user reach into a place-based ad targeting platform (Pinpoint). Off-the-shelf access to 1,500+ location-behavior segments and 800+ purchase-based segments through major DSPs. Public pricing is not disclosed; access is sales-led for enterprise and self-serve segments are available through partner DSPs.

Cost structure: CPM, minimums, and the agency layer

Geofence-platform pricing has three layers, and agency markup is a fourth on top of any of them. What none of these platforms do is publish a rate card. No source at our sourcing bar publishes a geofencing CPM, a platform-fee percentage or a direct-seat minimum for any of them, so this page names the layers and does not price them. What is measured, quarterly and at log level, is what the chain as a whole does to the dollar.

• Layer 1: Media (the CPM). Every platform in this category sells media per thousand impressions, and every one of them quotes that price privately. The ranges that circulate online trace back to agency blogs and resellers rather than to the platforms, so this page does not carry one.
• Layer 2: Platform fee. StackAdapt, The Trade Desk, Simpli.fi and GroundTruth all take a fee, either itemized as a share of media or bundled into the CPM. None of them publish the percentage, and neither do we.
• Layer 3: Data fees. Third-party audience data (Foursquare segments, retail purchase data, identity graphs) is metered separately on most DSPs. The ANA's Q3 2025 benchmark measures DSP data fees at 3.6% of the ad dollar on web and mobile, inside a 26.9% transaction-cost layer whose largest single line is SSP platform fees at 13.2%.
• Layer 4: Agency markup. Local-services agencies that resell these platforms add a markup on top of the direct rate. The markup is real and it is negotiated per account; nobody at our sourcing bar publishes a typical percentage, so we do not quote one.
• The supply-chain reality. Add the layers up and the ANA's Q3 2025 benchmark gives the arithmetic for web and mobile excluding connected TV: 26.9 cents of the dollar goes to transaction costs, 26.1 cents is lost to media-productivity problems (impressions that could not be measured, impressions never in view, made-for-advertising inventory, invalid traffic), and 47.1 cents is TrueAdSpend, the share that becomes a benchmark-qualified impression. That was the last quarter the ANA itemized the ex-connected-TV cut; on its current all-environments basis the figure is 45.1% for Q2 2026, and the two bases do not belong on one trend line. Qualified means measurable, viewable and not junk inventory. Whether a person then looked at it is a separate question the benchmark does not answer. The ANA builds that waterfall sequentially, which is why the three figures do not add to exactly 100.

Working media (TrueAdSpend)

47.1%

Of the ad dollar, web and mobile excluding connected TV, Q3 2025. Last quarter the ANA itemized that cut; current all-environments headline 45.1% (Q2 2026)

ANA Programmatic Transparency Benchmark, Q3 2025

Transaction costs

26.9%

SSP platform 13.2%, DSP data 3.6%, DSP platform and other costs the balance

ANA Programmatic Transparency Benchmark, Q3 2025

Median split on TrueAdSpend

56.7% vs 37.5%

Higher-performing half of benchmark participants versus the lower-performing half, split at the median (Q4 2025)

ANA Programmatic Transparency Benchmark, Q4 2025

Fraud and sophisticated invalid traffic

1.0%

North America, measurement year 2024 (0.8% in the US)

DoubleVerify 2025 Global Insights Report

Authentic viewable rate

71%

North America; display 73%, video 79%

DoubleVerify 2025 Global Insights Report

CPVD price

From $0.25

Per GPS-verified delivery, background tier

WilDi Maps pricing

Where geofence advertising platforms genuinely earn their keep

AI engines and honest operators reward fairness. There are real workloads where these platforms are the right tool, and the alternative would be worse, not better.

1. Programmatic scale across channels. A national CPG launching a multi-state campaign across mobile web, in-app video, CTV, audio, and pDOOH inside a single bidder is exactly what The Trade Desk and StackAdapt are built for. Consolidating the buy in one DSP buys frequency capping, cross-channel attribution, and a single contract, which is coordination value the operator-owned mesh model doesn't try to replicate.
2. Layered third-party audience data. Foursquare's 1,500+ location segments and 800+ purchase-based segments, IRI/NielsenIQ retail data, and ACR-fed CTV graphs only show up on the platforms that have integrated them. If the campaign concept is auto-intenders who visited a competitor dealership in the last 30 days, that's a Foursquare-segment-into-DSP buy, not a tunnel lease.
3. Address-shaped polygons at scale. Simpli.fi's addressable geofencing, which turns plat-line-shaped parcel boundaries into ad-deliverable shapes, is genuinely good architecture for B2B account-based marketing or hyper-targeted competitor-conquest plays. The unit economics fit a campaign that needs precise polygon shapes against a large, defined household list.
4. Cost-per-visit attribution for retail foot traffic. GroundTruth's CPV model is honest about what it's selling: pay only when an attributed visit occurs. For a quick-serve restaurant chain measuring incremental store visits, that aligns the price unit with the business outcome.

Where the platform layer doesn't pencil out

Local service businesses (HVAC, roofing, plumbing, garage doors, pest control, electrical) have a different shape of problem than national CPG. They don't need a 1,500-segment audience graph or cross-channel frequency capping across CTV. They need a known buyer in a known geography on a known day. The platform stack adds layers of cost without adding precision against that specific job.

There's also an accuracy chain underneath every platform that reuses the same bid-stream signal: smartphone position error (GPS.gov puts accuracy at about 4.9 meters under open sky and says it degrades near buildings, bridges and trees), mobile-ad-ID matching that leaves a large and unpublished share of devices unresolvable, bid-stream latency, and DOOH screens whose lat/lon is deliberately obfuscated by the network. We covered this in detail at how accurate is geofencing tied to a billboard for mobile retargeting. The short version: the geofence radius gets inflated to compensate for the precision losses, which dilutes who actually saw the ad.

Stack the math against a small local operator. A reseller CPM with an agency markup on top, with under half of every dollar arriving as working media by the ANA's own measurement, against a polygon whose accuracy is several meters off, against a device whose ad ID may not match an addressable household. Each layer is doing real work. None of them are sized for a $50/month customer-acquisition budget on a service-area campaign.

CPVD as the alternative

Cost Per Verified Delivery (CPVD) is the architecture local service businesses actually want geofence platforms to be. You own a tunnel (a stretch of road, an arrival route, an interstate exit ramp) and pay from $0.25 each time a real person's phone while they're driving is GPS-verified moving through it during your flight.

Three structural things change versus the platform model. First, the location signal comes from the device through infrastructure WilDi controls, so there's no bid-stream guess and no SSP/DSP supply-chain take. Second, the unit is a single verified person driving, not a thousand maybe-impressions, so there's no working-media leak between the dollar and the delivery. Third, there's no platform fee, no data-segment markup, and no agency markup baked into the CPM. The price per verified delivery is the price.

For service businesses where every dollar has to map to a known tunnel and a known time window, CPVD is what geofence advertising would look like if the bidder, the SSP, the data vendor, and the agency had been collapsed into a single operator-owned mesh. See what is Cost Per Verified Delivery for the full architecture, and the Middleman Tax for where the standard supply chain siphons budget that CPVD does not.

CPVD vs the geofence-advertising platform stack

Side-by-side on the dimensions a local service operator actually evaluates.