Skip to main content

Cookie Banner Ghosting: Why Analytics Loses 15-60% of Your Data

· 50 min read
Rafael Jimenez
Rafael Jimenez
Founder & CEO at Sealmetrics

TL;DR — Many visitors never answer the cookie banner — they ignore it. Ghosting, not rejection, is the quiet reason cookie-based analytics loses 15-60% of data.

Introduction​

The biggest threat to your analytics isn't cookie rejection—it's cookie banner ghosting.

While marketers obsess over rejection rates (users clicking "Reject All"), the quieter drain on analytics data is the large share of visitors who simply ignore cookie consent banners entirely. They don't accept. They don't reject. They just... ignore it.

This phenomenon, called "banner ghosting" or "decision avoidance," represents basic human psychology: when faced with an unwanted decision, people procrastinate or avoid it entirely. Cookie consent banners trigger this avoidance behavior at scale.

The math nobody runs:

  • A large group ignores the banner — no decision, no consent, no tracking
  • A second group rejects outright — explicit refusal, no tracking
  • Only the accepters are measured — and how big that group is depends entirely on who your visitors are

Where it lands: cookie-based analytics loses 15-60% of your visitor data. The spread is not noise — it is driven by your sector, the strength of your brand and where your traffic comes from. A recognised consumer brand whose visitors mostly arrive direct sits near the 15% end. A site buying cold traffic in a privacy-sensitive market sits near the 60% end. Either way, your real capture rate is somewhere between 40% and 85%, not 100%.

This isn't a rounding error—it's a structural hole in your business intelligence. Companies running Google Analytics 4, Adobe Analytics, or any cookie-based platform are making strategic decisions on a fraction of their actual traffic, and never a random fraction: the visitors who go missing are systematically different from the ones who stay.

The solution: Cookieless analytics platforms like Sealmetrics don't display consent banners because they don't require cookies. No banner means no ghosting, no rejection, no consent-driven data loss. Because no personal data is stored, the dataset falls outside the material scope of the GDPR (Recital 26), so no Article 6 legal basis is needed at all — and because nothing is written to or read from the device, the ePrivacy consent rule (Article 5(3)) is never triggered.

This article examines the psychology of banner ghosting, sizes the combined impact of ghosting plus rejection, and explains how cookieless analytics eliminates both problems simultaneously.

Key Takeaways​

  • A large share of visitors ignore cookie banners entirely without making any decision (banner ghosting)
  • Combined with outright rejection, cookie-based analytics loses 15-60% of visitor data — where you land depends on sector, brand strength and traffic sources
  • Banner ghosting is psychological avoidance behavior, not technical blocking
  • Cookie-based analytics cannot solve this problem—the banner itself causes the behavior
  • Cookieless analytics like Sealmetrics eliminates the banner, removing consent-driven data loss entirely

The Banner Ghosting Phenomenon​

Banner ghosting represents the single largest source of analytics data loss in 2025, yet it receives far less attention than cookie rejection. Understanding why users ghost banners is essential to recognizing that cookie-based analytics has no solution.

What is Banner Ghosting?​

Banner ghosting occurs when a website visitor:

  1. Sees the cookie consent banner
  2. Does not click "Accept All"
  3. Does not click "Reject All"
  4. Does not customize cookie preferences
  5. Continues using the site with the banner present

The user neither consents nor refuses—they simply ignore the decision entirely. During this state, cookie-based analytics tools cannot track the user because no consent has been granted. The user is invisible.

Scale of the Problem​

Where ghosting runs hottest:

Germany: the highest ghosting rates in Europe, driven by privacy awareness combined with sheer decision fatigue. German visitors are also the most likely to reject when they do engage.

France: after years of consent banner exposure, French users have developed textbook "banner blindness"—they see the popup but don't process it as requiring action.

Spain: Spanish internet users increasingly treat cookie banners as temporary annoyances to be ignored rather than genuine choices to be made.

United Kingdom: UK users demonstrate similar avoidance patterns to their EU counterparts despite Brexit, as UK cookie regulations remain aligned with EU standards.

European Union overall: ghosting is the norm rather than the exception. Nordic countries trend highest (high privacy awareness), Southern and Eastern Europe somewhat lower (emerging privacy awareness).

United States: materially lower than the EU, due to the absence of GDPR-style regulations and less intrusive consent mechanisms — but rising as California's CPRA and other state laws expand cookie consent requirements.

A caution before you reach for a calculator: ghosting rates are a statement about your banner, not about your dataset. How much analytics data you actually lose is a different number, and a smaller one. We work through why below.

Psychology of Decision Avoidance​

Banner ghosting is not a technical phenomenon—it's psychological. Users ghost banners due to well-documented cognitive biases:

Decision fatigue: Modern web users encounter 5-15 cookie banners per day. Each banner demands a privacy decision: accept tracking, reject tracking, or configure custom preferences. After the first few banners, users experience decision fatigue and begin ignoring subsequent banners to conserve mental energy.

Status quo bias: Humans prefer maintaining current state over making changes. When a cookie banner appears, the status quo (from the user's perspective) is "no banner on screen." The easiest way to maintain this state is to ignore the banner and continue browsing. Clicking "Accept" or "Reject" requires active decision-making, which conflicts with status quo bias.

Choice overload: Many cookie banners present complex choices: "Necessary cookies (required), Functional cookies (optional), Analytics cookies (optional), Marketing cookies (optional), Third-party cookies (optional)." Users face 5-10 binary decisions simultaneously. Choice overload leads to decision paralysis—so users make no decision at all.

Temporal discounting: The privacy implications of cookies (data collection, tracking, profiling) are abstract and future-oriented. The benefit of accessing website content is immediate and concrete. Users discount future privacy costs and prioritize immediate access, leading them to ghost the banner and proceed without deciding.

Reactance: Psychological reactance occurs when individuals perceive their freedom is being restricted. Cookie consent banners restrict the user's freedom to access content. Rather than comply with the restriction by making a choice, users experience reactance and refuse to engage with the banner as an assertion of autonomy.

Banner blindness: Years of exposure to intrusive popups, ads, and consent banners have trained users to visually filter out rectangular overlays on websites. Users literally do not process the banner as requiring action—their brain categorizes it as visual noise to be ignored.

Duration of Ghosting State​

How long do users remain in ghosting state?

Research on consent banner behavior shows:

Single-session ghosters (75-80%): Most users who ghost a banner do so for the entire session. They browse multiple pages, potentially convert, and leave—all while ignoring the banner. If they return days later, they encounter the banner again and often ghost again.

Multi-session ghosters (15-20%): Some users ghost across multiple sessions over days or weeks. They've trained themselves to ignore your banner specifically. Cookie-based analytics never captures these users unless they eventually accept or reject.

Permanent ghosters (5-10%): A small segment never interacts with cookie banners on any site. They've developed complete banner blindness. Cookie-based analytics will never track these users.

Median ghosting duration: 3-5 minutes (single session) to indefinitely (never decides)

For cookie-based analytics, ghosting is functionally equivalent to rejection. Whether a user clicks "Reject All" or ignores the banner, the outcome is identical: zero data captured.

Technical Manifestation of Ghosting​

When a user ghosts a cookie banner, here's what happens technically:

  1. User visits website: Browser loads HTML, CSS, JavaScript
  2. Analytics script loads: Google Analytics (or similar) JavaScript executes
  3. Consent check: Script checks for consent cookie (finds none)
  4. Banner displays: Consent management platform shows banner
  5. User ignores banner: No interaction occurs
  6. Analytics script waits: Cannot set tracking cookies without consent
  7. User browses site: Views multiple pages, clicks buttons, maybe converts
  8. No tracking occurs: Every pageview, every action, every conversion is invisible
  9. User leaves: Session ends without ever being tracked

From the analytics platform's perspective, this user never existed. They're not in your visitor count, not in your session data, not in your conversion reports. Ghosting creates a complete blind spot.

Critical distinction: Banner ghosting is different from technical blocking (ad blockers, privacy extensions, browser settings).

Technical blocking:

  • User actively installs software (uBlock Origin, Privacy Badger)
  • Blocks analytics scripts from loading
  • Affects 10-15% of users
  • Demonstrates active privacy preference

Banner ghosting:

  • User passively ignores consent banner
  • Analytics scripts load but cannot set cookies
  • Affects a far larger group than blocking does
  • Demonstrates decision avoidance, not privacy preference

Both result in data loss, but ghosting affects several times more users than technical blocking and represents passive avoidance rather than active privacy protection. This means cookieless analytics solves banner ghosting but not technical blocking (nothing can track users who block JavaScript entirely).

Combined Impact: What Ghosting Actually Costs You​

Understanding banner ghosting fundamentally changes how we calculate analytics data loss. The problem isn't just users who click "Reject"—it's the group who never click anything, combined with the group who explicitly reject.

The Real Data Loss Math​

Traditional analysis (incomplete): count the people who clicked "Reject All," call that your data loss, move on. This assumes every visitor makes a decision. Most don't.

What actually determines the number: three variables, none of which are the banner's design.

  1. Your sector. A privacy-sensitive vertical — finance, health, anything a visitor would rather not be seen browsing — pushes acceptance down hard.
  2. Your brand strength. People accept cookies from brands they already know and trust. An unfamiliar domain gets ghosted.
  3. Your traffic sources. Direct and branded-search visitors arrive with intent and accept at higher rates. Cold bought traffic does not.

Where that lands: 15-60% of your data, gone. Equivalently, a capture rate somewhere between 40% and 85%. The range is wide because the underlying businesses are genuinely different, not because the measurement is fuzzy.

Why the Loss Is Smaller Than the Non-Acceptance Rate​

Here is the step most write-ups skip, including earlier versions of this one. The share of visitors who don't accept the banner is not the share of data you lose. Two things sit in between:

Consent Mode modelling. Google Consent Mode v2 sends cookieless pings for non-consenting users and models the missing conversions and sessions from them. That modelled data is an estimate rather than an observation — you cannot segment it, you cannot trust it at low volumes, and you should not present it to a board as measured fact — but it is not nothing. It recovers part of the gap.

Repeat visits. A visitor who ghosts on Monday may accept on Thursday. Ghosting is a per-visit behaviour, not a permanent state, so a slice of the ghosted population folds back into your measured data over time.

Net of both effects, the honest figure for a cookie-based setup is the 15-60% band — not the far larger number you get by naively summing ghosters and rejecters. The problem is real without the inflation, and inflated numbers are the fastest way to lose an argument with a sceptical CFO.

Site-by-Site: Why the Loss Lands Between 15% and 60%​

Country is a weaker predictor than the profile of your business. Three sketches:

Profile A — strong consumer brand, mostly direct traffic

Sector:      retail / DTC, a name people recognise
Traffic:     direct + branded search
Acceptance:  comparatively high (visitors already trust the brand)
Data loss:   near 15%   →   capture rate near 85%

This is the best case, and it is still one visitor in seven who never appears in your reporting.

Profile B — mid-market B2B, mixed acquisition

Sector:      SaaS / professional services
Traffic:     organic, paid search, social referrals
Acceptance:  moderate, and inconsistent between channels
Data loss:   middle of the band

The dangerous case, because the loss is uneven across channels — which distorts comparisons between them rather than just shrinking the totals.

Profile C — cold paid traffic, privacy-sensitive market

Sector:      finance, health, legal, anything sensitive
Traffic:     bought, cold, no prior brand relationship
Acceptance:  low
Data loss:   near 60%   →   capture rate near 40%

Here you are making decisions on a minority of your traffic, and the minority is self-selected.

Why Ghosting Makes Rejection Statistics Misleading​

Industry discussions often cite "87% rejection rate in Germany" or "75% rejection rate in France." These statistics are rejection rate among users who interact with the banner, not rejection rate among all visitors — and neither one is your data loss rate.

Misleading stat: "87% of users reject cookies in Germany" What it actually says: 87% of the users who bothered to engage with the banner rejected. It is silent on the larger group who engaged with nothing at all.

Two different denominators, routinely conflated:

Rejections / (Acceptances + Rejections)   =   rejection rate among deciders
Non-accepters / All visitors              =   your consent gap
Measured shortfall vs reality             =   your actual data loss (15-60%)

Focusing on rejection rates understates the behaviour. Summing ghosting and rejection overstates the damage. Ghosting is often larger than rejection and receives no attention because ghosting users generate no event at all — there is no "reject" click to count.

Sample Bias: Who Accepts Cookies?​

The users who accept cookies are not representative of your total audience. They're a biased sample with distinct characteristics:

Cookie accepters tend to be:

  • Less privacy-aware (don't understand tracking implications)
  • Less technically sophisticated (don't use privacy tools)
  • More trusting of websites (brand loyalty or naivety)
  • In a hurry (click "Accept All" to dismiss banner quickly)
  • Younger or older (youth: less concerned; elderly: less aware)

Cookie ghosters and rejecters tend to be:

  • More privacy-aware (understand tracking, actively avoid)
  • More technically sophisticated (use privacy tools, read policies)
  • More skeptical of websites (question data practices)
  • More patient (willing to ignore banner or carefully reject)
  • Middle-aged professionals (prime awareness demographic)

Business consequences of sample bias:

Your analytics shows behavior of less privacy-aware, less technically sophisticated users who trust your brand. This is not representative of your market.

Example: You A/B test a feature. Variant A shows 5% conversion among cookie accepters. Variant B shows 6% conversion. You deploy B site-wide.

Problem: Privacy-aware users — the ones you're not tracking — might hate variant B because it requires more data entry, but you don't see their behavior. When you deploy B to 100% of traffic, overall conversions drop because the tested sample was unrepresentative.

Sealmetrics eliminates sample bias: with no banner in the way, Sealmetrics reports on your actual audience—privacy-aware and privacy-indifferent visitors alike. A/B tests are valid. Marketing attribution is accurate. Strategic decisions represent reality.

Business Impact Example: E-commerce​

Scenario: Fashion e-commerce site, 100,000 monthly visitors, 2% conversion rate (actual), €75 average order value. Mixed acquisition, so it sits mid-band — call it a 50% capture rate.

Reality (partly invisible to cookie-based analytics):

  • Visitors: 100,000
  • Orders: 2,000
  • Revenue: €150,000

What Google Analytics 4 shows:

  • Tracked visitors: 50,000 (50,000 invisible)
  • Tracked orders: ~1,000 (1,000 orders from ghosters/rejecters invisible)
  • Tracked revenue: ~€75,000 (€75,000 invisible)
  • Apparent conversion rate: 2% (coincidentally correct, but based on a biased sample)

Note what is not wrong here: the conversion rate. Loss that were uniform would mostly shrink your numbers without distorting your ratios, and you could live with that. The damage comes from the loss being uneven, which is what the next three decisions show.

Business decisions based on incomplete data:

  1. Marketing attribution: Google Ads shows 200 conversions for €10,000 spend (€50 CPA). LinkedIn shows 40 conversions for €4,000 spend (€100 CPA). LinkedIn looks twice as expensive, so the company cuts its budget.

    Reality: search visitors arrive with intent and accept the banner more often (say 60% captured); LinkedIn's B2B audience is more privacy-aware and accepts less (say 40%). Google actually drove ~333 conversions, LinkedIn ~100. True CPA: Google €30, LinkedIn €40. LinkedIn is still the more expensive channel — but by a third, not by double, and it was cut on the strength of a gap that was mostly measurement artefact.

  2. Product performance: Product A shows 1,000 views and 20 purchases (2% conversion). Product B shows 800 views and 25 purchases (3.1%). A looks like the volume driver, B like a small high-converting niche.

    Reality: Product A is bought by returning brand loyalists arriving direct, who accept the banner readily (~75% captured). Product B attracts comparison shoppers arriving cold from price aggregators, who don't (~45% captured). Actual figures: Product A ~1,333 views and ~27 purchases; Product B ~1,778 views and ~56 purchases. The conversion rates were right. The volume ranking was backwards — B is the bigger product, and inventory was planned against the wrong one.

  3. Landing page priorities: Analytics shows the campaign landing pages carrying almost all measured revenue, so the team pours budget into more of them.

    Reality: the comparison and specification pages that cold, privacy-aware researchers use before buying are systematically under-measured, because that audience is exactly the audience that ghosts banners. The pages that look like dead weight are doing work you can't see.

Cost of decisions based on a half-blind sample: €50,000-100,000 annually in misallocated marketing spend, wrong inventory, and suboptimal site design.

Sealmetrics solution: see all 100,000 visitors, all 2,000 orders, all €150,000 of revenue. Marketing attribution is accurate. Product demand is visible. Decisions are based on your whole audience rather than the half of it that clicks "Accept."

Banner ghosting is not a technical problem that cookie-based analytics platforms can fix through better engineering. It's a psychological problem caused by the banner itself. As long as a consent banner exists, ghosting will occur.

Failed Solutions​

Cookie-based analytics vendors have attempted multiple approaches to reduce ghosting and rejection:

1. Consent Mode (Google Analytics 4)

Google's "consent mode" attempts to provide degraded analytics for non-consenting users:

  • Tracks aggregate data without individual user identifiers
  • Uses modeling to estimate behavior of non-consenting users
  • Provides partial data instead of no data

Why it falls short:

  • Still requires a consent banner (ghosting continues)
  • Modelled data is estimated, not observed — you can't segment it or drill into it
  • Modelling quality degrades badly at low traffic volumes
  • Legal uncertainty (some DPAs consider consent mode insufficient)
  • It narrows the gap rather than closing it: a real shortfall remains, and it is the part of the 15-60% band you cannot see into

2. Progressive Consent / Delayed Banners

Some sites delay banner display hoping users engage with content first:

  • Show banner after 10-30 seconds on site
  • Show banner after user scrolls 50% down page
  • Show banner on second pageview, not first

Why it fails:

  • Users still ghost delayed banners at much the same rate, regardless of timing
  • Creates compliance risk (tracking before consent)
  • Regulators (especially CNIL) explicitly prohibit pre-consent tracking
  • Ghosting is deferred, not eliminated

3. Banner UX Optimization

Endless tweaking of banner design to increase acceptance:

  • Make "Accept" button more prominent (dark patterns)
  • Hide "Reject" in submenus (illegal in most EU jurisdictions)
  • Use persuasive copy ("Help us improve your experience")
  • Simplify choices to binary Accept/Reject

Why it fails:

  • Dark patterns are illegal under GDPR and face fines
  • Users still ghost even optimized banners (decision avoidance, not design issue)
  • Regulators require "Reject" to be as prominent as "Accept"
  • Optimisation nudges acceptance up by a few points; it does not move you out of the 15-60% loss band

4. Incentivized Consent

Offering incentives for accepting cookies:

  • "Accept cookies to get 10% discount"
  • "Accept cookies to access premium content"
  • "Accept cookies to enter giveaway"

Why it fails:

  • Legally problematic (consent must be "freely given" under GDPR)
  • Creates compliance risk if incentive is seen as coercive
  • Users still ghost (offer is irrelevant to decision avoiders)
  • May lift acceptance somewhat, but ghosting barely moves

5. AI-Powered Consent Prediction

Using machine learning to predict which users will accept/reject:

  • Show different banner variants to different users
  • Optimize timing and messaging per user segment
  • Attempt to reduce ghosting through personalization

Why it fails:

  • Still requires banner (ghosting continues)
  • Users resent personalized manipulation
  • Regulatory scrutiny on profiling for consent decisions
  • Marginal improvement at the edges doesn't solve the underlying data loss problem

Fundamental Impossibility​

Cookie-based analytics faces an unsolvable paradox:

  1. The ePrivacy Directive requires consent to store or read information on a user's device — Article 5(3), which is what cookies do
  2. Consent requires a banner to obtain user agreement
  3. Banners trigger ghosting due to psychological decision avoidance
  4. Ghosting prevents tracking because no consent was granted
  5. No tracking means no data for a large slice of your visitors

This cycle cannot be broken within cookie-based architecture. The moment you display a consent banner, a substantial share of users will ghost it. No amount of optimization, design changes, or technical workarounds can eliminate decision avoidance behavior.

The only solution is to eliminate the banner entirely—which requires eliminating cookies entirely.

Why Cookieless Analytics Solves Ghosting​

Cookieless analytics platforms like Sealmetrics break the paradox by removing cookies from the equation:

  1. No cookies used for analytics tracking
  2. No consent required, because nothing is stored on or read from the device (ePrivacy Article 5(3) is never engaged) and no personal data is stored (Recital 26)
  3. No banner displayed to users
  4. No ghosting possible (nothing to ghost)
  5. Every visitor measured, with no consent mechanism in the way

Banner ghosting is eliminated because the banner itself is eliminated. Users cannot avoid a decision they're never asked to make.

Psychological advantages of no banner:

  • No decision fatigue (no decision required)
  • No status quo disruption (content accessible immediately)
  • No choice overload (no choices presented)
  • No temporal discounting (no privacy decision to defer)
  • No reactance (no freedom restriction)
  • No banner blindness (no banner to ignore)

Sealmetrics removes the psychological triggers that cause ghosting in the first place. Users access your site without friction. Your analytics sees the whole audience. And the compliance position rests on architecture rather than on a consent record: no personal data stored, nothing written to the device.

How Sealmetrics Eliminates Both Ghosting and Rejection​

Sealmetrics solves the combined problem of banner ghosting and cookie rejection — together, the 15-60% hole in your dataset — by eliminating the root cause: the consent banner itself.

Technical Approach: Cookieless Tracking​

Sealmetrics uses session-based tracking that doesn't require cookies:

Session-ID Tracking (Primary Method):

  1. Visitor arrives: User lands on your website
  2. Session identifier computed: Sealmetrics JavaScript derives an ephemeral session identifier from general device characteristics — it is not unique to a person (different visitors can produce the same value) and cannot identify an individual
  3. No browser storage: The identifier is never stored on the device — no cookies, no LocalStorage, no SessionStorage
  4. Session tracking: All pageviews during this browser session use the same identifier
  5. Automatic expiry: The identifier expires with the session and is never used to correlate visits over time — each new entrance counts as new, independent data (privacy by design)
  6. No consent required: nothing is stored on or read from the user's device, so the ePrivacy consent rule (Article 5(3)) is never triggered — and because no personal data is retained, there is no Article 6 legal basis to choose in the first place

Isolated Hits Tracking (Fallback Method):

For the small share of visits where a session identifier cannot be computed:

  1. Server-side inference: Sealmetrics tracks each pageview as isolated hit
  2. Pattern recognition: Server logic infers session continuity based on:
    • Pageview timing (views within 30 minutes likely same session)
    • Referrer patterns (internal referrers suggest continued session)
    • Navigation flow (homepage → product → checkout suggests single journey)
  3. Conservative attribution: When uncertain, treats hits as separate sessions
  4. Maximum privacy: Zero browser storage, pure server-side analysis

Why This Approach Eliminates Ghosting​

No banner = no ghosting. Users cannot ignore a decision they're never presented with.

When a user visits a site using Sealmetrics:

  1. Page loads immediately (no banner delay)
  2. Content is accessible instantly (no friction)
  3. Sealmetrics measurement occurs in the background (no user awareness required)
  4. No decision required from user (no cognitive load)
  5. Every visit measured (no consent-driven data loss)

From the user's perspective, the site "just works." From the business perspective, analytics "just works." No banner. No ghosting. No rejection. No data loss.

Why This Approach Eliminates Rejection​

No consent request = no rejection. Users cannot reject what they're not asked to approve.

Cookie-based analytics requires obtaining consent before tracking. This creates the rejection problem: a meaningful share of the users who engage with the banner choose "Reject All."

Sealmetrics doesn't request consent because it doesn't use cookies and doesn't store personal data — no IP addresses, not even hashed ones.

Rejection is impossible when no consent mechanism exists. There is no banner to reject, so that source of loss disappears along with ghosting.

The argument runs on two separate tracks, and it is worth keeping them apart, because they answer different regulators.

Track one — ePrivacy Article 5(3). This is the rule that actually mandates cookie banners. It requires consent to store information on, or gain access to information stored in, a user's terminal equipment. Sealmetrics writes nothing to the device: no cookies, no LocalStorage, no SessionStorage, and it reads nothing back. The obligation is never triggered, so there is nothing to consent to.

Track two — GDPR material scope, Recital 26. The data protection principles do not apply to anonymous information — information which does not relate to an identified or identifiable natural person. Because Sealmetrics stores no personal data, the resulting dataset falls outside the material scope of the Regulation.

This is why we do not claim legitimate interest. It is tempting to reach for Article 6(1)(f), and plenty of vendors do. But invoking any Article 6 basis presupposes that you are processing personal data and merely have a good reason for it. That concedes the entire point. If no personal data is stored, no legal basis is needed — asserting one would weaken the position, not strengthen it.

The one place Article 6(1)(f) does legitimately appear is narrower: the transient, in-memory handling of an IP address for security and anti-abuse purposes, which Recital 49 addresses directly. That address is never written to storage and never reaches the analytics dataset.

Regulatory guidance: CNIL (French data protection authority) has confirmed that audience measurement meeting specific criteria can operate without consent. Note that CNIL does not certify or approve individual tools, and neither does any other supervisory authority — no such scheme exists for analytics software. See our CNIL self-assessment for how we assess ourselves against those criteria.

DPO reviews: Sealmetrics has passed vendor privacy reviews by Data Protection Officers at multiple EU enterprises evaluating its consentless architecture.

Data Capture Comparison​

Cookie-based analytics (Google Analytics 4):

Total visitors:      100,000
Ghosting + rejection: the missing slice
Consent Mode:         models part of it back, as estimates
Data capture rate:    40-85%, depending on sector,
                      brand strength and traffic sources
Data loss:            15-60%

Cookieless analytics (Sealmetrics):

Total visitors:      100,000
No banner displayed: 100,000 (all measured)
No ghosting possible: not applicable
No rejection possible: not applicable
Data capture rate:   full, minus the JavaScript blockers
                     that no analytics tool can see
Consent-driven loss: none

Impact: depending on where your site sits in that band, Sealmetrics surfaces roughly 1.2x to 2.5x the data a cookie-based tool reports — and, more importantly, eliminates the sample bias, which is the part that makes decisions go wrong rather than merely go small.

Real-World Implementation​

Migration from Google Analytics 4 to Sealmetrics:

Step 1: Install Sealmetrics tracking code (2 minutes)

<script src="https://t.sealmetrics.com/t.js?id=YOUR_SITE_ID" defer></script>

Step 2: Remove consent banner code (1 minute)

  • Delete Cookiebot, OneTrust, or custom consent code
  • Remove conditional analytics loading logic

Step 3: Remove Google Analytics 4 code (1 minute)

  • Delete GA4 gtag.js scripts
  • Remove GA4 configuration

Total implementation time: 4 minutes

Result:

  • No consent banner
  • Every visit measured
  • No personal data stored, so no consent record to maintain
  • Real-time analytics
  • No consent-driven data loss

User experience improvement:

  • Faster page loads (no banner script overhead)
  • Immediate content access (no banner friction)
  • No decision fatigue (no choice required)
  • Better mobile experience (no screen-covering popup)

Business intelligence improvement:

  • Substantially more visitor data, depending on where your site sits in the band
  • Accurate marketing attribution
  • Valid A/B test samples
  • True conversion rates
  • Representative analytics (not biased toward cookie accepters)

Business Impact of a 15-60% Data Gap​

Operating with cookie-based analytics in the current environment means making strategic decisions on a partial, self-selected sample of your visitors. The consequences cascade through every business function — and they get worse, not better, as the gap gets more uneven between channels.

Marketing Attribution Collapse​

Scenario: Mid-sized SaaS company, €100,000 monthly marketing budget. Capture rates vary sharply by channel — email and branded search convert consenting visitors readily; LinkedIn's B2B audience does not.

Cookie-based analytics shows:

Google Ads: €30,000 spend, 150 conversions = €200 CPA
LinkedIn Ads: €20,000 spend, 40 conversions = €500 CPA
SEO content: €15,000 spend, 60 conversions = €250 CPA
Email marketing: €10,000 spend, 80 conversions = €125 CPA

Decision: Cut LinkedIn (appears most expensive by a factor of four), increase Email (appears cheapest)

Reality with Sealmetrics, once each channel is measured on the same footing:

Google Ads: €30,000 spend, 250 conversions = €120 CPA   (60% captured)
LinkedIn Ads: €20,000 spend, 100 conversions = €200 CPA (40% captured)
SEO content: €15,000 spend, 100 conversions = €150 CPA  (60% captured)
Email marketing: €10,000 spend, 107 conversions = €94 CPA (75% captured)

Actual optimal decision: LinkedIn is genuinely the most expensive channel, but at €200 CPA rather than an apparent €500 — well inside the range where it earns its budget. Trim it if the economics demand; don't kill it on a number that was never real.

Cost of the wrong decision: €25,000-40,000 annually in misallocated marketing spend

This is the mechanism that matters. LinkedIn looks disproportionately expensive because its audience — B2B professionals — is more privacy-aware and consents less often than Google searchers. Uneven loss between channels doesn't shrink your report, it reorders it. The channel that looks worst may simply be the channel whose audience reads cookie banners.

A/B Test Statistical Invalidity​

Scenario: E-commerce checkout optimization test

Test design:

  • Variant A: 3-step checkout (control)
  • Variant B: 1-page checkout (experimental)
  • Required sample: 1,000 conversions per variant for 95% confidence
  • Expected runtime: 2 weeks at full capture — roughly 4 weeks at a 50% capture rate

Cookie-based analytics results:

Variant A: 1,043 conversions, 3.2% conversion rate
Variant B: 1,127 conversions, 3.5% conversion rate
Winner: Variant B (1-page checkout)
Statistical significance: p < 0.05 (appears valid)

Decision: Deploy 1-page checkout site-wide

Reality with complete data (Sealmetrics):

Variant A: 2,300 conversions, 3.4% conversion rate
Variant B: 2,150 conversions, 3.1% conversion rate
Actual winner: Variant A (3-step checkout)
Statistical significance: p < 0.05 (reversed result)

What happened: the consenting sample preferred the 1-page checkout. The unmeasured half — ghosters and rejecters, who skew more privacy-aware and security-conscious — preferred the 3-step checkout, because entering payment details on a dedicated page felt safer. Note that the two variants weren't even captured at the same rate, which is exactly why the tracked totals looked closer than they were.

Cost of deploying wrong variant: 3.4% → 3.1% conversion rate = roughly a 10% drop in conversions = €150,000 annual revenue loss for a €1.5M/year e-commerce site

Time waste: 4 weeks to run the test on a half-sample, 2 weeks to discover variant B underperforms in production, 2 weeks to roll back and re-test — 8 weeks gone

With Sealmetrics: a 2-week test on the full audience, correct winner identified first time, six weeks saved and the €150k avoided.

Product Roadmap Misallocation​

Scenario: SaaS product team prioritizing features based on analytics

Cookie-based analytics shows (a sample biased toward less privacy-aware users):

Feature usage:
- Zapier integration: 15% of users
- Email reports: 45% of users
- Mobile app: 10% of users
- API access: 5% of users
- Custom dashboards: 25% of users

Decision: Prioritize email reports (highest usage), drop API from the roadmap (lowest usage, 9x behind)

Reality with Sealmetrics:

Actual feature usage:
- Zapier integration: 26% of users (tech-savvy, high ghosting rate)
- Email reports: 34% of users (less tech-savvy, low ghosting rate)
- Mobile app: 17% of users (high ghosting rate on mobile)
- API access: 12% of users (developers, highest ghosting rate)
- Custom dashboards: 24% of users (roughly accurate, little ghosting bias)

What happened: privacy-aware users — the ones who ghost banners — are exactly the users who reach for technical features like the API and Zapier. Under-representing them makes technical features look like dead weight.

Email reports really is the most-used feature; that part survives contact with complete data. What doesn't survive is the margin. API access looked like a rounding error at 5%, nine times behind the leader, which is what justified cutting it. At 12% it is a quarter of your user base and under three times behind — a very different conversation, and one nobody got to have.

Cost of the wrong roadmap: six months of development aimed at a gap that wasn't there, and a segment written off on a number that was an artefact of who accepts cookies.

With Sealmetrics: the ranking and the margins both hold up, so the prioritisation debate happens on real numbers.

Revenue Reporting Inaccuracy​

Scenario: Board meeting, CEO presents revenue metrics

Cookie-based analytics shows:

Monthly revenue: €150,000 (tracked conversions only)
Conversion rate: 2.5%
Average order value: €95
Top product: Product A (€45,000 tracked revenue)

Board decision: Product A is the star performer, allocate more marketing budget to it

Reality with Sealmetrics:

Actual monthly revenue: €300,000 (blended capture rate around 50%)
Actual conversion rate: 2.5% (rate was coincidentally correct, volume was not)
Actual average order value: €95 (correct)
Top product: Product B (€140,000 actual, €56,000 tracked)
Product A: €60,000 actual, €45,000 tracked

What happened: Product B is bought by enterprise researchers with high privacy awareness — barely two in five consent. Product A is bought by impulse buyers who accept the banner about three times in four. Both products were under-reported; only one was under-reported enough to lose its place. Analytics named Product A the top performer when Product B was generating well over twice its revenue.

Cost of wrong strategic decision: Marketing budget misallocation to Product A instead of Product B, potentially €100,000-200,000 annually in opportunity cost

Competitive Intelligence Blind Spots​

Scenario: SaaS company analyzing competitive positioning

Cookie-based analytics shows:

Users visiting from competitors:
- 500 visits from competitor A site
- 200 visits from competitor B site
- 100 visits from competitor C site

Decision: Competitor A is the main threat at 2.5x competitor B, focus positioning against them

Reality with Sealmetrics:

Actual competitive referrals:
- 625 from competitor A (comparison sites, low ghosting)
- 500 from competitor B (privacy-focused competitor, high ghosting among its users)
- 250 from competitor C (technical competitor, high ghosting among developers)

What happened: people researching privacy-focused alternatives ghost banners at the highest rates of anyone. Competitor B isn't a distant second at 40% of A's volume — it is within 20% of it, effectively neck and neck.

Cost: a positioning strategy built around a comfortable lead that doesn't exist, and messaging that never seriously engaged the competitor actually taking your traffic

Psychological Drivers of Banner Ghosting (Deep Dive)​

Understanding why users ghost banners helps explain why the problem is unfixable within cookie-based architectures and why eliminating the banner is the only solution.

Average user encounters: 8-15 cookie consent banners per day across different websites. Each requires a decision.

Psychological research: Humans have limited decision-making capacity. After making several decisions, decision quality degrades and people begin avoiding decisions entirely (decision fatigue).

Cookie banner fatigue pattern:

  • Banner 1-2: User reads and decides (accept/reject)
  • Banner 3-5: User skims and decides (probably rejects)
  • Banner 6-10: User ignores (ghosting begins)
  • Banner 11+: Complete banner blindness (automatic ghosting)

By the time a user reaches your website (likely not their first site of the day), they've already made multiple cookie consent decisions and have entered decision fatigue state. Your banner is ignored reflexively.

Why "better banners" don't work: Optimizing your banner doesn't address that users are exhausted by all banners, not specifically yours. Decision fatigue is cumulative across all sites, not site-specific.

Status Quo Bias and Loss Aversion​

Status quo bias: Humans prefer maintaining current state over changing state, even when change might be beneficial.

User perception of cookie banner:

  • Current state: "I want to read this article / buy this product"
  • Banner appears: "I need to make a privacy decision before I can proceed"
  • Status quo preservation: Ignore banner, continue toward goal

From the user's perspective, the path of least resistance is ignoring the banner and accessing the content they came for. Clicking "Accept" or "Reject" requires deviating from the goal path.

Loss aversion: Users frame the banner as a potential loss (lose time making decision, lose privacy by accepting, lose functionality by rejecting). When faced with potential loss, humans procrastinate or avoid the decision.

Why "incentivized consent" doesn't work: Offering rewards for accepting cookies attempts to reframe the decision as a gain, but research shows loss aversion is stronger than equivalent gain attraction. Users still avoid the decision to avoid the potential loss.

Choice Architecture and Paralysis​

Choice overload: When presented with too many options, humans experience decision paralysis and make no choice at all.

Typical cookie banner choices:

  1. Accept All (cookies enabled)
  2. Reject All (cookies disabled)
  3. Customize Preferences:
    • Necessary cookies (required)
    • Functional cookies (optional)
    • Analytics cookies (optional)
    • Marketing cookies (optional)
    • Personalization cookies (optional)
    • Third-party cookies (optional)

Users face 6-10 binary decisions simultaneously. Most users don't understand the technical differences between cookie categories. Choice overload leads to decision paralysis → ghosting.

Regulatory requirements: EU regulators (especially CNIL) require that "Reject All" be as prominent and easy to access as "Accept All," and that users be able to customize preferences. This prevents simplified "Accept / Learn More" patterns and forces choice complexity.

Why "simplified banners" don't work: GDPR and regulatory guidance mandate complexity. You cannot legally simplify to a single "Accept" button without equivalent "Reject" prominence. Required complexity creates choice overload regardless of UX design.

Temporal Discounting and Hyperbolic Time Preferences​

Temporal discounting: Humans value immediate rewards more than future rewards, and discount future costs more than immediate costs.

Cookie banner decision framework:

  • Immediate cost: Time spent reading banner, making decision, clicking button
  • Immediate benefit: Access to website content
  • Future cost: Privacy implications of cookies (tracking, profiling, data breaches)
  • Future benefit: Customized experience, supporting free content

Users heavily discount future costs (privacy invasion months later) and prioritize immediate benefit (access content now). The rational choice, given temporal discounting, is to ignore the banner and access content immediately.

Why privacy warnings don't work: Explaining that cookies track users across sites, build profiles, and share data with third parties describes future costs. Temporal discounting means users don't weigh these future costs as heavily as immediate content access benefit.

Psychological Reactance​

Reactance theory: When individuals perceive their freedom is being restricted, they experience psychological reactance—motivation to restore freedom, often by resisting the restriction.

Cookie banner as freedom restriction: Users perceive the banner as restricting their freedom to access website content without making a privacy decision. This restriction triggers reactance.

Reactance response to cookie banners:

  • Ignore banner entirely (passive resistance)
  • Immediately click "Reject All" without reading (active resistance)
  • Leave website (ultimate resistance)
  • Use ad blockers or privacy tools to prevent banners (systemic resistance)

Ghosting the banner is a form of passive reactance—users refuse to engage with the banner as an assertion of autonomy. The more prominent or persistent the banner, the stronger the reactance response.

Why "gentle reminders" don't work: Making the banner reappear after scrolling or after certain time periods increases reactance. Users feel the site is nagging them, strengthening their resolve to ignore it.

Banner blindness: After years of exposure to intrusive web elements (ads, popups, modal overlays), users have trained their brains to visually filter out rectangular overlays.

Cognitive filtering process:

  1. User's visual system detects rectangular overlay
  2. Pattern recognition: "This is an ad / banner / popup"
  3. Automatic categorization: "Irrelevant to my goal"
  4. Visual filtering: Brain suppresses awareness of overlay
  5. Attention remains on underlying content

Cookie consent banners trigger the same cognitive filtering as ads and popups. Users' brains have literally trained themselves not to process the banner as requiring attention.

Why "attention-grabbing design" doesn't work: Making banners more colorful, animated, or attention-grabbing actually reinforces banner blindness. Users recognize "attention-grabbing overlay" as a pattern to be ignored.

The Unsolvable Nature of Ghosting​

All five psychological mechanisms (decision fatigue, status quo bias, choice overload, temporal discounting, reactance) are hardwired human cognitive biases. They cannot be designed away.

Cookie-based analytics cannot solve ghosting because:

  1. Banner fatigue is cumulative across all sites (not site-specific)
  2. Status quo bias applies to any interruption (not design-specific)
  3. Choice complexity is legally mandated (not optional)
  4. Temporal discounting is cognitive bias (not education issue)
  5. Reactance increases with banner prominence (persistence backfires)
  6. Banner blindness is learned behavior (not reversible)

The only solution is eliminating the banner entirely, which requires eliminating cookies entirely, which requires cookieless analytics like Sealmetrics.

Implementation: Migrating to Cookieless Analytics​

Eliminating banner ghosting and cookie rejection requires eliminating the consent banner, which requires eliminating cookies, which requires migrating to cookieless analytics.

Step 1: Quantify Current Data Loss​

A. Install Sealmetrics in parallel with existing analytics (5 minutes)

<!-- Keep existing Google Analytics -->
<script async src="https://www.googletagmanager.com/gtag/js?id=GA_MEASUREMENT_ID"></script>
<!-- Add Sealmetrics -->
<script src="https://t.sealmetrics.com/t.js?id=YOUR_SITE_ID" defer></script>

B. Run both systems for 7-14 days

C. Compare visitor counts:

Week 1 Results:
Google Analytics 4: 26,244 visitors
Sealmetrics: 58,320 visitors
Data loss (GA4): 55% (32,076 visitors invisible)

The invisible visitors are ghosters and rejecters in some
mix you cannot determine from the outside — the ghosters
generate no event to count. What you can determine is the
total, which is the number that matters.

Whatever figure you get, expect it to land between 15% and 60%. If it lands outside that, check your setup before you celebrate or panic: a suspiciously small gap usually means Consent Mode modelling is inflating your GA4 side, and a suspiciously large one usually means a tagging error.

D. Calculate business impact:

Invisible conversions: ~245 per week (extrapolated)
Invisible revenue: ~€18,375 per week
Annual invisible revenue: ~€955,000

This revenue exists, but your analytics can't attribute it to marketing channels, can't analyze user behavior, can't optimize conversion funnel.

A. Review Sealmetrics privacy practices:

  • ✅ No cookies used
  • ✅ No IP addresses stored (not even hashed)
  • ✅ No cross-site tracking
  • ✅ No third-party data sharing
  • ✅ Temporary session identifiers only, never written to the device
  • ✅ Nothing stored on or read from the device, so ePrivacy Article 5(3) is not engaged
  • ✅ No personal data stored, so the dataset sits outside GDPR material scope (Recital 26)
  • ✅ Data hosted in Dublin, Ireland

B. Consult with DPO (if applicable):

  • Share Sealmetrics' privacy documentation and DPA
  • Review the analysis of why no Article 6 basis is required
  • Obtain DPO sign-off for consentless operation

C. Update privacy policy:

Old text (cookie-based):
"We use cookies to analyze website traffic. You can accept or reject cookies 
via the consent banner. For more information, see our cookie policy."

New text (cookieless):
"We use Sealmetrics, a cookieless analytics service, to understand website 
usage and improve user experience. Sealmetrics does not use cookies, does not 
store IP addresses, and does not collect personal data. Nothing is stored on or 
read from your device, and session identifiers are temporary and are never 
written to your browser. Because no personal data is retained, this measurement 
falls outside the scope of the GDPR. Questions: privacy@yourcompany.com."

A. Identify consent management code:

  • Cookiebot
  • OneTrust
  • Custom banner implementation

B. Remove banner scripts:

<!-- DELETE THESE -->
<script id="Cookiebot" src="https://consent.cookiebot.com/uc.js" data-cbid="YOUR-ID" type="text/javascript" async></script>

<!-- DELETE CONDITIONAL LOADING -->
<script type="text/javascript">
function loadAnalyticsWithConsent() {
  if (Cookiebot.consent.statistics) {
    // Load GA4
  }
}
</script>

C. Remove cookie policy page (or simplify drastically):

  • Delete detailed cookie tables
  • Delete consent management instructions
  • Keep privacy policy (simplified)

D. Verify banner removal:

  • Test site in multiple browsers
  • Test from multiple EU countries (VPN)
  • Confirm no cookie banner appears
  • Verify Sealmetrics tracking works immediately

Step 4: Remove Google Analytics 4​

A. Remove GA4 tracking code:

<!-- DELETE THIS -->
<script async src="https://www.googletagmanager.com/gtag/js?id=GA_MEASUREMENT_ID"></script>
<script>
  window.dataLayer = window.dataLayer || [];
  function gtag(){dataLayer.push(arguments);}
  gtag('js', new Date());
  gtag('config', 'GA_MEASUREMENT_ID');
</script>

B. Export historical data (if needed):

  • GA4 API export to BigQuery or CSV
  • Download key reports for historical reference
  • Archive in data warehouse

C. Redirect team to Sealmetrics:

  • Update bookmarks
  • Update dashboard links
  • Train team on Sealmetrics interface

Step 5: Verify Full Data Capture​

A. Server log comparison:

Week after migration:
Server logs (Apache/Nginx): 59,140 unique visitors
Sealmetrics: 58,320 visitors
Capture rate: 98.6% (difference likely bots/scrapers)

B. Cross-device testing:

  • Test on Chrome (desktop + mobile)
  • Test on Safari (desktop + mobile)
  • Test on Firefox (desktop + mobile)
  • Test with privacy extensions (uBlock Origin, Privacy Badger)
  • Test with strict browser privacy settings
  • Verify all tests appear in Sealmetrics real-time

C. Conversion tracking validation:

  • Complete test purchases/signups
  • Verify appear in Sealmetrics immediately
  • Check attribution (source/medium)
  • Compare to payment processor data

Success metrics:

  • ✅ 98-99% of legitimate traffic tracked (excluding bots)
  • ✅ No consent banner visible
  • ✅ Full conversion tracking operational
  • ✅ Team adopted Sealmetrics interface
  • ✅ DPO sign-off documented

Step 6: Analyze New Insights​

A. Discover previously invisible traffic:

Month 1 with Sealmetrics vs Last Month with GA4:

Total visitors: 115,860 (was 52,140) → 2.2x
LinkedIn referrals: 4,900 (was 1,850) → 2.6x
Mobile traffic: 69,516 (was 28,713) → 2.4x
Conversions: 2,860 (was 1,304) → 2.2x

The uneven multipliers are the finding, not the headline number. LinkedIn and mobile climb further than the site average because their audiences consented least — which is precisely the distortion that was quietly reshaping your channel reporting.

B. Correct marketing attribution:

  • Channels previously undervalued (LinkedIn, privacy-focused referrers) now show true performance
  • Channels previously overvalued (some Google search terms) now show accurate performance
  • Reallocate budget based on complete data

C. Identify UX issues affecting ghosters/rejecters:

  • Previously invisible user segments now visible
  • Discover friction points affecting the privacy-aware segment you were never measuring
  • Optimize for your whole audience, not just the part of it that clicks "Accept"

D. Run valid A/B tests:

  • Reach statistical significance sooner, on more data
  • Representative samples (not biased toward cookie accepters)
  • Results hold in production (tested on actual user base)

Migration Timeline​

Week 1: Parallel Testing

  • Day 1: Install Sealmetrics alongside GA4
  • Day 2-7: Compare data, identify discrepancies
  • End of week: Quantify data loss (typically somewhere in the 15-60% band)

Week 2: Preparation

  • Day 8-9: DPO review and approval
  • Day 10-11: Update privacy policy
  • Day 12-14: Export GA4 historical data

Week 3: Migration

  • Day 15: Remove consent banner code
  • Day 16: Remove GA4 code
  • Day 17-21: Verify 100% tracking, team training

Week 4: Optimization

  • Day 22-28: Analyze new data, correct attribution, optimize based on complete visitor insights

Total time: 4 weeks from start to full optimization

Actual implementation time: 10-15 minutes (installation + removal)

Rest of time: Validation, training, analysis

Frequently Asked Questions​

Enough that ghosting, not rejection, is usually the larger group. A substantial share of visitors ignore consent banners entirely, making no decision at all — neither accepting nor rejecting. Rates run highest in Germany and the Nordics, lower in Southern and Eastern Europe, and materially lower in the US. Be careful what you infer from the figure, though: ghosting rates describe your banner, not your dataset. The data loss that reaches your reports is 15-60%, which is smaller than the raw non-acceptance rate.

Why do users ghost banners instead of rejecting?​

Users ghost banners due to psychological decision avoidance: decision fatigue (exhausted from multiple daily banners), status quo bias (prefer not interrupting their goal), choice overload (too many cookie options), temporal discounting (prioritize immediate content access over future privacy), psychological reactance (resisting the forced decision), and banner blindness (trained to ignore rectangular overlays). Ghosting requires zero effort while clicking "Reject" requires effort.

No. Cookie rejection occurs when users actively click "Reject All" or customize preferences to reject cookies. Banner ghosting occurs when users ignore the banner entirely without interacting. From an analytics perspective, both result in no tracking for that visit, but ghosting typically affects the larger group and represents passive avoidance rather than an active privacy choice. Ghosters are also harder to quantify, because they generate no event — there is no click to count.

What's the total data loss from ghosting plus rejection?​

15-60% of your data, with a capture rate of 40-85%. The spread depends on your sector, the strength of your brand and where your traffic comes from: a recognised consumer brand whose visitors arrive direct sits near 15%, while a site buying cold traffic in a privacy-sensitive market sits near 60%.

Don't reach that number by adding ghosting and rejection together — that overstates it. Two things sit between non-acceptance and actual data loss. Consent Mode v2 models part of the unconsented traffic back into your reports (as estimates, not observations). And ghosting is a per-visit behaviour, so some visitors who ignored the banner on one visit accept on a later one.

No. Banner ghosting is caused by the existence of the consent banner itself, not by banner design flaws. Cookie-based analytics requires cookies, which require consent under GDPR, which requires a banner, which triggers psychological decision avoidance (ghosting). No amount of UX optimization, consent mode implementation, or banner design improvement can eliminate decision avoidance behavior. The only solution is eliminating the banner entirely by using cookieless analytics.

How does Sealmetrics eliminate banner ghosting?​

Sealmetrics doesn't use cookies and stores nothing on the device, so no consent is required and no banner is displayed. No banner means no ghosting is possible—users cannot avoid a decision they're never asked to make. Measurement begins on page load, with no user interaction required.

Yes, when it is genuinely cookieless and genuinely stores no personal data. Two separate rules are in play. The ePrivacy Directive's Article 5(3) is the one that mandates cookie banners, and it applies to storing or reading information on the user's device — Sealmetrics does neither. The GDPR governs personal data, and its own Recital 26 puts anonymous information outside its material scope — so with no personal data stored, no Article 6 legal basis is required.

Note what we are not saying: we do not claim legitimate interest. Invoking Article 6(1)(f) would concede that personal data is being processed. CNIL has confirmed that audience measurement meeting specific criteria can operate without consent, but no supervisory authority certifies or approves individual analytics tools — no such scheme exists. Sealmetrics has passed vendor privacy reviews by DPOs at multiple EU enterprises, which is a customer assessment, not a regulatory endorsement.

What happens to my historical Google Analytics data?​

Historical data remains in Google Analytics 4 for as long as Google retains it (14 months for free accounts, longer for GA360). You can export historical data via GA4 API or reports before migration. Sealmetrics begins capturing fresh data from installation forward. Most businesses maintain GA4 access for historical reference while using Sealmetrics for all current and future analytics.

How long does it take to migrate from GA4 to Sealmetrics?​

Implementation takes 10-15 minutes: install Sealmetrics tracking code (2 minutes), remove consent banner scripts (5 minutes), remove GA4 code (3 minutes), update privacy policy (5 minutes). Total migration timeline including validation and team training is typically 2-4 weeks, but the technical implementation is less than 15 minutes of developer time.

Will I lose conversion tracking without cookies?​

No. Sealmetrics tracks conversions, goals, and custom events without requiring cookies. Conversion tracking covers every visit, not just the visitors who accepted a banner. Marketing attribution (source, medium, campaign) is captured via referrer and UTM parameters without consent requirements. ROI calculation becomes more accurate because you see all conversions, not just those from cookie accepters.

What about users who block JavaScript?​

Users who block JavaScript entirely (0.2-0.5% of visitors) cannot be tracked by any JavaScript-based analytics, including Google Analytics and Sealmetrics. This is an unavoidable technical limitation shared by both. The difference is what sits on top of it: Sealmetrics loses only the JS blockers, while cookie-based analytics loses the JS blockers plus the banner ghosters and the rejecters — a 15-60% gap on top of the same floor.

Can I use Sealmetrics with Google Tag Manager?​

Yes, Sealmetrics can be installed via Google Tag Manager or directly in HTML. However, direct HTML installation is simpler (one line of code) and avoids GTM overhead. Since Sealmetrics doesn't require consent management integration (no banner, no conditional loading), GTM provides minimal benefit. Most users install directly in the HTML <head> section for simplicity and performance.

Does Sealmetrics track across subdomains or multiple domains?​

Sealmetrics tracks across subdomains automatically (example.com, blog.example.com, shop.example.com all tracked as single property). For multiple separate domains (example.com and other-example.com), you need separate Sealmetrics projects. This is intentional for privacy—Sealmetrics doesn't enable cross-site tracking that would raise GDPR concerns.

What if a competitor also uses Sealmetrics?​

Sealmetrics projects are completely isolated. Competitor using Sealmetrics has zero impact on your analytics. Each project has unique project ID and separate data storage. Unlike Google Analytics (where Google has access to all GA data across all sites), Sealmetrics data is private to each project owner. No cross-site tracking, no data sharing between projects.

How does Sealmetrics handle returning visitors?​

Sealmetrics uses session-based tracking. Each browser session gets a temporary identifier computed in the browser from general device characteristics — it is not unique to a person, is never stored on the device, and expires with the session. When a user returns days later, that visit is simply counted as a new, independent entrance. Sealmetrics does not correlate visits over time and does not identify returning visitors — there is no cross-session history or profile, which is exactly what keeps it privacy-safe (no long-term tracking).

Can users opt out of Sealmetrics tracking?​

Yes. Although no consent is required, sites using Sealmetrics should still provide an opt-out mechanism in their privacy policy (email address or form for objection requests). Additionally, users blocking JavaScript or using browser privacy modes (incognito) are not tracked. Sealmetrics respects Do Not Track browser signals when enabled.

Conclusion​

Cookie banner ghosting—not cookie rejection—is the primary cause of analytics data loss in 2025.

While industry discussions focus on rejection rates (users clicking "Reject All"), the larger and quieter group is the visitors who simply ignore cookie consent banners entirely. These users neither accept nor reject—they ghost the banner and continue using your site. During this ghosting state, cookie-based analytics cannot track them.

Combined with explicit rejection, that puts total data loss at 15-60%, depending on your sector, the strength of your brand and where your traffic comes from — a capture rate of 40-85%. The missing visitors aren't privacy extremists running ad blockers. They're ordinary people engaging in predictable psychological decision avoidance, which is exactly why they're missing in a systematic, non-random pattern rather than an even one.

The root cause is the consent banner itself, not banner design or implementation. Banner ghosting is driven by hardwired cognitive biases:

  • Decision fatigue (exhausted from multiple daily banners)
  • Status quo bias (prefer not interrupting browsing goals)
  • Choice overload (too many cookie options create paralysis)
  • Temporal discounting (prioritize immediate access over future privacy concerns)
  • Psychological reactance (resisting forced decisions)
  • Banner blindness (learned visual filtering of rectangular overlays)

Cookie-based analytics cannot solve banner ghosting because the banner is required to obtain legally valid consent under GDPR, and the banner's existence triggers the psychological avoidance mechanisms that cause ghosting. No amount of UX optimization, consent mode, or banner redesign can eliminate decision avoidance behavior.

Cookieless analytics eliminates both ghosting and rejection by eliminating the consent banner entirely. Sealmetrics doesn't use cookies and stores no personal data, so the ePrivacy consent rule is never triggered and no GDPR legal basis is required. No banner means no ghosting is possible, no rejection is possible, and every visit is measured from the moment it lands on your site.

Business impact of a 15-60% data gap:

  • Marketing attribution skewed by uneven loss between channels → €25,000-40,000 annual misallocation
  • A/B tests run on an unrepresentative sample, wrong variants deployed → six-figure revenue loss
  • Product roadmap ranking features by how consent-friendly their users are → months of wasted development
  • Strategic decisions made on a self-selected slice of your audience → opportunity cost nobody ever invoices you for

The asymmetry is the argument: migrating to Sealmetrics takes 15 minutes of implementation and two to four weeks of validation. Continuing as you are costs a slice of every decision you make, indefinitely.

For businesses operating in the EU, this isn't really a tooling preference. Losing between 15% and 60% of your visitors is survivable; losing them non-randomly is what quietly bends your conclusions. You cannot optimize what you cannot measure. You cannot attribute revenue you cannot see. And you cannot correct a bias you don't know the size of.

Sealmetrics measures every visit, eliminates banner ghosting, eliminates cookie rejection, and does it without storing personal data — which is what removes the consent question rather than merely answering it.

Stop guessing at the missing 15-60%. Try Sealmetrics free for 14 days and see your complete visitor base for the first time.

Additional Resources​

Legal Resources:

Psychological Research:

  • Decision Fatigue Studies (Baumeister et al.)
  • Status Quo Bias Research (Samuelson & Zeckhauser)
  • Choice Overload Theory (Iyengar & Lepper)
  • Psychological Reactance Theory (Brehm)

Related Reading:

Official Resources: