The SubscriptionShame Q2 2026 Report.

Two numbers, one real gap.

Why "average American subscription spend" has two different answers.

If you searched for that phrase this quarter, you likely hit both numbers within the first page of results, with no explanation of why they disagree. Here's the honest answer: they're measuring different things.

CNET's 2026 State of Subscriptions survey (commissioned from YouGov Plc, n=2,522 US adults, 2,074 of whom pay for at least one subscription, fielded April 29–May 1, 2026) asked people directly: roughly how much do you spend on subscriptions per month? The answer, $111/month and $1,332/year, is a recall-based, top-of-mind estimate. It's real data, honestly collected, but it's the same kind of number that every underestimation study warns about: people remember Netflix and Spotify; they forget the iCloud storage bump, the app they trialed in March, the gym's annual fee that hit once.

C+R Research, Bango, and Empower's figures ($219–$273/month) come from methodologies designed to catch the subscriptions people forget: asking respondents to review statements, prompting for bundled services (Apple One, carrier perks, insurance add-ons), or aggregating known billing data. These are closer to a full accounting, not a guess.

Put plainly: ask someone what they spend and you get $111. Make them actually add it up and it's closer to $250. That roughly 2× gap is not a data error. It's the finding. It's the same gap our own calculator exists to close: type in what you actually pay for, rate how often you use it, and the guessing stops.

What CNET's fresh survey also found.

Source: CNET, "US Adults Are Throwing Away Over $250 a Year on Unused Subscriptions" (published Jul 6, 2026), survey conducted by YouGov Plc, weighted to be representative of US adults 18+.

What we verified ourselves this quarter.

Unlike the modeled category breakdowns below, everything in this section is original, dated, first-party research: pricing we checked directly against each company's own pages and policy documents between April and July 2026, not survey extrapolation.

Where the subscription dollars actually go.

Carried forward from Q1, unchanged this quarter. See methodology below for why.

Waste share is modeled by combining published survey data with our pricing database, not measured from calculator usage. Same methodology and figures as the Q1 report; see "How this report was put together" below.

The database grew this quarter.

SubscriptionShame's pricing database now tracks 167 services, up from 158 at the end of Q1. Additions include SiriusXM, Audible, The Wall Street Journal, Discord Nitro, DoorDash DashPass, Uber One, Walmart+, and YouTube TV, each priced and verified directly against the provider's current US pricing page. We also launched a Charge Decoder directory this quarter (free lookups for cryptic statement charges like APPLE.COM/BILL and AMZN Digital) after noticing that some of the top search results for these charges were low-quality or outright scam pages posing as billing support.

How this report was put together.

Read this before citing any number above. This report blends three kinds of numbers, and we label which is which rather than let you guess:

1. Fresh primary survey data (cited directly): CNET's 2026 State of Subscriptions survey, conducted by YouGov Plc (n=2,522, fielded Apr 29–May 1, 2026, published Jul 6, 2026). We quote its figures exactly as published and note plainly that they are self-reported estimates, not audited totals.
2. Earlier published US surveys (also cited directly): C+R Research 2024 Subscription Survey (n=1,000+ US adults), Bango Subscription Wars 2024 (n=5,000 across 5 markets, US weighted), Empower 2025 Subscription Spending Report. These figures use fuller-accounting methodologies and remain the source for the $219–$273/month range.
3. Our own first-party research (new this quarter): the price-hike and hidden-fee findings above, each verified against the company's own pricing or policy pages on the date shown, and each backed by a full cancellation walkthrough on our blog.
4. Modeled estimates (unchanged from Q1, clearly labeled): the category breakdown above is constructed by combining survey data with our pricing database. It has not been updated this quarter because we still do not have real, aggregated calculator-usage data to replace it with; see the note below.

Why there's still no real usage data. The SubscriptionShame calculator has always run entirely in your browser: nothing you type is sent to a server, stored in a database, or tracked with analytics. That's a deliberate privacy choice, and it means we have never had a mechanism to measure real, aggregate usage patterns. The Q1 report's promise to deliver that "from Q2 onward" was made before that constraint was fully accounted for. We're not publishing invented numbers to fill the gap. If a future report adds a fully anonymous, aggregate-only measurement (no individual tracking, ever), we will say so explicitly and explain exactly what is and isn't collected before it ships.

We also deliberately exclude a category of source that dominates this search topic: secondary "subscription tracker" blogs that repackage the same handful of studies while blending self-reported and audited figures into a single number with no methodology note. If you see $111 and $273 quoted as the same statistic somewhere, that's what happened.

If a published statistic on this page goes stale or you spot a math error, please use the contact form and we will fix it within 48 hours.

CITE THIS REPORT

Free for journalists and researchers.

Interview requests, custom data pulls, or media additions: please use our contact form.