The Middleman Tax.
Why 30% of every ad dollar never reaches a human.
Also known as "ad platform fees" or the programmatic supply-chain skim. The Middleman Tax is the roughly 30% of every ad dollar absorbed by exchanges, DSPs (demand-side platforms), SSPs (supply-side platforms), and resellers between a business's budget and a real person seeing the ad. It's not a line item on any invoice. It's the hidden tax on modern advertising — and the number has been confirmed by two independent industry audits.
Where the money actually goes
For every $1,000 you put into a major ad platform, roughly $700 reaches a human who could act on it. The other $300 is absorbed before delivery, across a chain of intermediaries:
- ~15% — Auction rake. Exchanges and DSPs take a cut every time your bid clears. You pay for the privilege of competing.[1]
- ~10% — Ad-tech "supply chain" fees. SSPs, verification vendors, fraud tools, brand-safety scanners, viewability measurers. Each layer adds a fee. [2]
- ~5–10% — Viewability failure. The ad renders but isn't seen — off-screen, auto-scrolled, in a background tab. You paid, the exposure didn't happen. [3]
- ~5%+ — Bot fraud and invalid traffic. Industry research reports 10–15% of impressions come from bots, click farms, or invalid traffic. Even after detection, a share still gets billed. [4]
The ANA audit labeled a portion of this skim the "unknown delta" — the gap between what the advertiser paid and what the publisher received that could not be reconciled in supply-chain logs. [1]
The figures vary by vertical — high-CPC (cost-per-click) categories like auto and mortgage see extraction rates above 40%.
What $100 actually buys
A single $100 spent in programmatic, traced from budget to human:
- 1$30–$40 — skimmed immediately by the layered ad-tech stack (the Middleman Tax).
- 2$60–$70 — left for "working media" (actual ad placement).
- 3Of that, another 30–50% lost to bot fraud, invalid traffic, and non-viewable impressions.
- 4Net: ~$35 of the original $100 reaches a real human.
You paid for $100 worth of reach. You got roughly a third of it.
Tax vs. Waste — two leaks, not one
Most audits catch one, not both:
Tax
What middlemen charge you
Transaction fees, platform skim, unknown delta.
Waste
What never delivered in the first place
Bots, non-viewable impressions, invalid traffic.
Programmatic stacks suffer from both simultaneously, and they compound: you pay a middleman to broker fraudulent traffic. That's why aggressive audits report effective extraction rates of 60%+ on a dollar-in, human-out basis.
What that costs a local business
Independent restaurant
$2,000 / month
- Annual spend
- $24,000
- Skim to middlemen
- ~$7,200
- Reaches real humans
- ~$16,800
- Equivalent verified deliveries
- ~120,000
Regional chain · 13 locations
$6,000 / month
- Annual spend
- $72,000
- Skim to middlemen
- ~$21,600
- Reaches real humans
- ~$50,400
- Equivalent verified deliveries
- ~360,000
Industry waste benchmarks
Combines WordStream CPC (cost-per-click) benchmarks [5], DoubleVerify IVT (Invalid Traffic — bots, click farms, non-human sources) rates [4], and the ANA-confirmed ~30% programmatic skim [1]. Estimated total waste = (industry IVT rate) + (platform skim baseline). Actual waste varies by campaign, platform, and season.
| Industry | Typical CPC¹ | Invalid-traffic rate² | Notable dynamic | Est. total waste³ |
|---|---|---|---|---|
| Legal (personal injury) | $50 – $150 | ~14% | Highest CPCs in digital; heavy competitor click-fraud | ~45–55% |
| Insurance (auto & life) | $10 – $50 | ~13% | Affiliate / comparison-shopping bots inflate IVT | ~40–50% |
| Mortgage / lending | $20 – $40 | ~13% | Rate-comparison bots; strict disclosure adds verification fees | ~40–50% |
| Automotive (dealers) | $3 – $15 | ~12% | Competing dealers click each other; high used-vehicle CPCs | ~35–45% |
| Local services (plumbing, HVAC — heating & cooling, electrical) | $4 – $20 | ~13% | Pay-per-call fraud; competitor call-bombing | ~40–50% |
| Real estate | $2 – $6 | ~11% | Luxury listings attract scraper fraud | ~35–45% |
| Healthcare / wellness | $3 – $8 | ~12% | Extra brand-safety vendor fees on regulated verticals | ~35–45% |
| Financial services (business-to-consumer) | $5 – $20 | ~12% | Heavy verification-vendor stack | ~35–45% |
| Gaming / app installs | Cost-per-install (CPI) varies | ~20% | SDK (software development kit) fraud, install farms, click-flooding | ~50–60% |
| B2B SaaS (business software) | $3 – $10 | ~11% | High LinkedIn cost-per-thousand (CPM); smaller fraud exposure | ~30–40% |
| Travel / hospitality | $1 – $4 | ~12% | OTA (online travel agency) take-rates layer on top of ad-tech skim | ~30–40% |
| E-commerce / retail | $1 – $3 | ~10% | Scale efficiencies; lowest relative waste | ~28–35% |
Methodology. Industry benchmarks above combine publicly reported cost-per-click (WordStream), invalid-traffic rates (DoubleVerify), and the ANA/ISBA ~30% programmatic-skim baseline. WilDi does not publish private audit data; specific per-campaign waste varies with platform, creative, and time of year.
Refreshed against source reports April 2026. Verify against the current-year WordStream and DoubleVerify benchmarks before syndication.
Why the Middleman Tax exists in the first place
Programmatic advertising was designed to be an open auction — hundreds of buyers bidding on billions of impressions in milliseconds. The infrastructure to coordinate this (exchanges, DSPs, SSPs, verification vendors, brand-safety tools) was never free. Each layer charges a fee. Over 15+ years, those fees compounded into a structural 30%+ take.
The business owner never sees the breakdown. The dashboard shows:
- How much you spent
- How many "impressions" were "delivered"
- An inferred "cost per thousand"
It does not show:
- What percent of your spend the exchange kept
- How many of those impressions were human
- Whether anyone was looking when the ad rendered
The tax is invisible by design.
How WilDi Maps replaces it
WilDi Maps is not an exchange. It's not a DSP. It doesn't bid for your budget against anyone.
WilDi is spatial infrastructure — a physical mesh of real phones carried by real drivers through real streets. You lease a corridor (a Tunnel), a neighborhood (a Zone), or a nation-wide window (a Background). When a driver crosses your geography during your window, a message is delivered to their phone and acknowledged with a tap.
That tap is a verified delivery. It's the only thing you pay for. Not an impression. Not a probabilistic "likely reached." Not a cookie-based inference. A real human, in a real car, acknowledging your ad.
The model is deterministic, not probabilistic. Because WilDi controls the infrastructure end-to-end — the driver app, the verification layer, the operator dashboard — there's no open exchange for fraud to exploit. No middleman to broker the delivery. No unknown delta to reconcile.
Every dollar deploys. Nothing is absorbed by an exchange. No auction. No bidding. No "programmatic supply chain." Just a fixed cost per verified delivery.
Frequently asked
Is the Middleman Tax a real thing or marketing language?
What about direct-sold deals? Don't those avoid the tax?
Aren't some of these fees necessary — fraud detection, brand safety, viewability?
Is WilDi the only way to avoid the Middleman Tax?
Direct-sold and insertion-order deals also bypass the exchange but still layer in trafficking, viewability, and verification vendor costs. Extraction is smaller (5–15%) but never zero.
WilDi is the only option that offers the flexibility and measurement of programmatic with the directness of physical media, billed at a fixed rate perconfirmed delivery — the driver taps to acknowledge your message, so there's no estimation, no rotation cycle, no glance assumed. Every unit you pay for is a unit that landed.
Does the tax apply to social media ads too?
Sources
- 1
ANA Programmatic Media Supply Chain Transparency Study
Association of National Advertisers with PwC · 2017 (original) / 2023 (update)
The foundational US audit of programmatic supply-chain leakage; introduced the 'unknown delta' concept.
www.ana.net - 2
ISBA Programmatic Supply Chain Transparency Study
Incorporated Society of British Advertisers with PwC · 2020
The UK parallel of the ANA audit; tracked 31 advertisers and 47 publishers across 267 campaigns.
www.isba.org.uk - 3
MRC Viewable Impression Measurement Guidelines
Media Rating Council · 2014, ongoing updates
The industry-standard viewability definition (≥50% of pixels in view for ≥1 continuous second).
mediaratingcouncil.org - 4
DoubleVerify Global Insights Report — Invalid Traffic & Fraud Benchmarks
DoubleVerify · 2024
Annual quantification of bot fraud, sophisticated-invalid-traffic (SIVT), and non-human exposure rates across programmatic.
doubleverify.com - 5
WordStream Google Ads Industry Benchmarks
WordStream · 2024
Annual cost-per-click and click-through-rate (CTR) benchmarks across industries; tracks year-over-year CPC inflation on major platforms.
www.wordstream.com
Related reading
About
What WilDi operates →
Veteran-owned traffic-infrastructure company. Three deployment primitives, one rate card.
Pricing
Operator plans →
Flat monthly fee plus fixed per-delivery rate. Early Adopter lock-in during the Jacksonville pilot.
Tool
Run the Waste Audit →
Plug in your monthly ad spend. See exactly how much is being skimmed.
Stop paying the tax. Own the infrastructure.
Lease a corridor, a neighborhood, or a window. Fixed cost per verified delivery. No auction. No middleman. No bots.
Enterprise or agency? Book 20 minutes with Mike (CEO) →