Cookie consent tools meter pageviews, and a spike tips the tier

A consent banner costs nothing until the campaign works. Six consent platforms were read off their own pricing pages on 2026-09-16 and they do not meter the same unit: two count pageviews, one sessions, one banner views, one monthly visitors, and one does not count traffic at all. The free ceilings are low enough that one post landing well clears them in a week.

What the banner is actually doing before it is a line item

The consent layer records a choice about non-essential storage before that storage happens. The GDPR defines consent as “any freely given, specific, informed and unambiguous indication of the data subject’s wishes” (Article 4(11)), and Article 7 requires that “it shall be as easy to withdraw as to give consent”. The EDPB’s Guidelines 05/2020 and California’s CCPA are the reference points vendors build against.

The meter matters because the consent record is a precondition for the tags. Google’s consent mode documentation describes two implementations, basic, which blocks tags until consent, and advanced, which loads tags with defaults and adjusts to the consent state. Tags with built-in consent checks collect according to that signal, so a consent script that stops serving takes the measurement of the campaign with it.

This is about billing mechanics, not legal advice. What a specific site needs is a question for a lawyer.

The meter is not the same unit at every vendor

VendorUnit statedFree or entry ceiling as printedDomainsPage read
CookieYesPageviews/monthFree 5,000; Basic $10/mo 100,000Per domain; one free plan per accountcookieyes.com/pricing/
CookiebotSubpages (unique URLs). “Does not charge based on page views or usage”Free up to 50 subpages; Lite EUR 7/mo up to 501 on Free and Litecookiebot.com/en/pricing/
UsercentricsSessions/monthFree 1,000; Essential EUR 7/mo 1,5001 to Plus, 3 on Pro, 10 on Businessusercentrics.com/pricing/
iubendaPageviews/monthEssentials EUR 4.99/site/mo, 25,000; free plan states no numberPer siteiubenda.com/en/pricing
TermlyBanner views: “visitors who view your consent banner”Free 10,000; Starter $10/site/mo 50,000; Pro+ unlimited1 website; more per licensetermly.io/pricing/
OsanoMonthly visitorsFree 5,000; Plus $199/mo 30,0001 on Free, 3 on Plusosano.com/plans/cookie-consent

All six pages were fetched on 2026-09-16. Osano’s own /pricing page states no plan number and routes to a demo, so its figures come from its self-service plans page. A dated snapshot, not Watchdog data and not on the re-verification schedule.

What a vendor counts as one pageview

CookieYes is the one of the six that documents its unit properly. Its Pageview Pricing page, updated 23 June 2026, states that it “registers a pageview every time a webpage containing the CookieYes code loads”, filters known crawlers and malicious bots, and counts “all pageviews on a website, whether or not a website visitor clicks or interacts with the cookie banner”. So a returning visitor who already consented still counts, and a page where the banner never renders still moves the meter.

The other five leave gaps. iubenda defines its unit as “the total number of times a web page is loaded/reloaded” and says nothing about repeat visitors or bots. Usercentrics answers the repeat case with a window rather than a page: “A session represents each time a visitor loads your website with our compliance script for a 30-minute period”, each refresh after 30 minutes counting as new. Termly does not say whether a suppressed banner counts, and Osano defines nothing. None of the six says whether a single-page app changing routes without a page load increments the counter. Those absences are the finding.

The unit is not cosmetic for a spiky pattern. Pageviews and banner views count arrivals; sessions and visitors collapse repeat activity by one person into one unit. An audience arriving repeatedly moves a pageview meter far faster.

Why a social spike breaks a monthly meter differently than growth does

Work it against a printed ceiling. CookieYes’ Free plan allows 5,000 pageviews per month. Every traffic figure below is an assumption of this example, not a measurement.

Assume a site averaging 120 pageviews a day, so 3,600 over 30 days, inside the ceiling. A post lands on day 12 and delivers 4,000 extra pageviews over three days. Days 1 to 11 consume 1,320; days 12 to 14 add 360 of baseline plus the 4,000, so the total reaches 5,680, crossing the ceiling partway through day 14. The remaining 16 days, another 1,920 assumed pageviews, all fall past the limit. The allowance was spent on a Tuesday while the campaign still points at the page.

Ordinary growth behaves nothing like it. From 3,600 a month at 10% growth the site reaches 3,960, then 4,356, then 4,792, crossing 5,000 in the fifth month. Four invoices of warning. A single post gives none.

The reset basis is mostly unstated. CookieYes documents it: the count “resets every month based on your billing cycle date”. Cookiebot checks subpages at the end of the billing cycle. Usercentrics, iubenda, Termly and Osano print monthly figures without saying which day the month turns.

What the vendor says happens at the ceiling

  • CookieYes. Either outcome, on one setting. With extra pageviews enabled the banner “will remain active”, overage at $0.30 per 1,000. Without it, alerts fire at 80%, 95% and 100%, and at 100% “your cookie banner will no longer be displayed on your website and no new pageviews will be tracked”. Not available on the Free plan.
  • iubenda. Overage on paid plans: past the quota “you will be charged EUR 0.05 per 1,000 pageviews”, so “everything continues to run with a small extra charge”. Silent on the free plan.
  • Usercentrics. Exceed the session limit and “your plan will automatically be adjusted in accordance with our Terms & Conditions”.
  • Cookiebot. If the scanner finds more subpages than the plan allows, “your subscription will automatically upgrade to the next appropriate tier”.
  • Termly. Nothing stated past 10,000 on Free or 50,000 on Starter.
  • Osano. Nothing stated past the allowances.

The two shapes are not equally bad live. A quiet bill is recoverable: the banner keeps serving and the cost surfaces on an invoice. A script that stops is not, because the traffic that triggered it is the traffic you lose the record of. On a lapse, Usercentrics states the setup stays intact but logs cannot be accessed and the banner is not displayed; Cookiebot the same after a trial. iubenda states that on a plan change “your data and settings stay put”.

The second meter nobody reads: domains, subdomains and scans

Each of the six also charges on something other than traffic. CookieYes prices per domain, iubenda per site. Cookiebot bills further domains at the next tier up, Termly sells extra websites per license with subdomain scanning on Pro+, and Usercentrics stays at one domain until Pro. The trap needs no spike: give a campaign its own landing host and a one-domain plan is over its allowance at ten visitors. None of the six addresses a staging host running the same script.

There is a scan meter too. CookieYes prints a pages-per-scan limit beside the pageview allowance: 100 on Free, 600 on Basic, 4,000 on Pro, 8,000 on Ultimate. That caps how much of the site is examined for cookies, not how much is served.

The check to run before the post goes out

  1. Read the usage figure in the consent tool’s own dashboard, not an estimate from analytics, and record the date.
  2. Read the ceiling off the vendor’s pricing page the same day, because the pages change.
  3. Find the reset date, not just the word monthly, and count the days left in the cycle.
  4. Compare the remainder against what the last comparable post produced, from the site’s own analytics.
  5. Check where the campaign’s landing host sits against the domain allowance.
  6. Read what the vendor states happens at the cap, and if it turns on a setting, confirm it now.

Where the rest of this pattern is written up

Several quotas on the same page empty on the same traffic event. Adjacent pieces cover the others: session recorders meter the same spike, privacy-first analytics reporting social traffic as direct, monitoring the page a campaign points at, and response-metered form tools. Disclosure: followedapp is published by the team behind RecurPost, and RecurPost is covered vendor #34 under the same rules as every other vendor.

FAQ

Do consent platforms bill by pageview or by visitor?

The unit is printed on each pricing page and it varies. Of the six read on 2026-09-16, CookieYes and iubenda meter pageviews, Usercentrics sessions on a 30 minute window, Termly banner views, Osano monthly visitors, and Cookiebot meters neither, sizing plans on unique URLs.

What happens the moment a free plan’s limit is passed?

Outcomes run from an automatic charge to a banner that stops appearing. iubenda charges overage on paid plans, CookieYes either bills $0.30 per 1,000 extra pageviews or, without that setting, stops displaying the banner until the next cycle, and Usercentrics and Cookiebot move the subscription up a tier. Termly’s and Osano’s pages state no answer. Establish which applies before a campaign: tags gated behind the consent signal do not fire without it, so losing the layer mid-spike costs the measurement of the traffic that caused it.

Sources

Scroll to Top