- Browser cookie restrictions and privacy regulations are making cookie-based A/B testing increasingly unreliable
- Varify.io is architecturally cookieless — uses localStorage, not workarounds or first-party cookie fallbacks
- No consent banner needed, 100% of visitors tested, no tracking blocked by Safari ITP or Firefox ETP
- Native integration with GA4, BigQuery, Matomo — no parallel tracking, no data discrepancies
The cookie era in web experimentation is ending. Safari's Intelligent Tracking Prevention (ITP) caps first-party cookies at 7 days. Firefox Enhanced Tracking Protection blocks many tracking cookies by default. Chrome's Privacy Sandbox is reshaping how third-party data works. For A/B testing tools that rely on cookies for visitor recognition, this means shorter attribution windows, incomplete data, and growing compliance complexity.
Varify.io was built for this reality. It's architecturally cookieless — not as a workaround or fallback mode, but as a core design decision. Varify uses localStorage for variant assignment, never sets a single cookie, and delegates all measurement to your existing analytics (GA4, BigQuery, Matomo). The result: no consent banner needed, no ITP restrictions, 100% of visitors tested.
Why cookies are becoming a problem for A/B testing
Browser restrictions are tightening
Safari ITP: First-party cookies set via JavaScript are capped at 7 days. A visitor who returns after 8 days gets a new cookie — your A/B test sees them as a new user. This inflates unique visitor counts and dilutes test results.
Firefox ETP: Blocks third-party cookies by default and restricts some first-party cookie behaviors. Testing tools that rely on cross-domain cookies or tracking pixels are affected.
Chrome Privacy Sandbox: While Chrome hasn't fully eliminated third-party cookies, the Privacy Sandbox APIs are replacing traditional cookie-based tracking with new mechanisms. Tools built on cookie architectures face ongoing adaptation.
Consent rates are declining
As privacy awareness grows, cookie consent rates in Europe are dropping — some sites report below 50% acceptance. Cookie-based A/B tests lose half their data before the experiment even starts. See our GDPR compliance guide for the full picture.
Three approaches to cookieless A/B testing
1. True cookieless (Varify's approach)
No cookies at all. Variant assignment via localStorage. No consent banner needed. 100% of visitors tested. Trade-off: localStorage doesn't persist across browsers or in incognito — statistically negligible for most tests.
2. First-party cookie fallbacks (Convert, VWO)
Some tools offer "cookieless modes" that actually use first-party cookies with various workarounds (server-side cookie setting, reduced scope). These still trigger ePrivacy consent requirements and are still subject to ITP restrictions. It's better than third-party cookies but not truly cookieless.
3. Server-side identification (GrowthBook, custom)
Assign variants server-side using hashed user IDs or session tokens. Truly cookieless and powerful, but requires developer resources for implementation. Not suitable for teams that need visual editors and marketer-friendly workflows.
Cookieless capabilities compared
| Feature | Varify.io | Convert | PostHog | VWO |
|---|---|---|---|---|
| Cookieless mode | Default (always) | Optional | Configurable | Not available |
| Technology | localStorage | 1st-party cookies | Configurable | Cookies only |
| Consent needed | No | Yes | Configurable | Yes |
| ITP affected | No | Yes (7-day cap) | Depends | Yes |
| Visual editor | ||||
| Price | from €149/mo | from $99/mo | Free (self-host) | from $299/mo |
Source: Claude Research, May 1, 2026
Cookieless A/B testing. Not a workaround — an architecture.
Varify never sets cookies. No consent issues, no ITP blocking, no tracking gaps. From €149/month.
The future is cookieless — prepare now
The trajectory is clear: browser vendors are restricting cookies further, regulators are tightening consent requirements, and users are becoming more privacy-aware. A/B testing tools built on cookie architectures will face increasing friction — shorter attribution windows, smaller sample sizes, more compliance overhead.
Tools built on cookieless architectures (like Varify) are already prepared. No adaptation needed when the next browser update drops, no scramble when ePrivacy Regulation passes, no consent rate to worry about.
For teams evaluating A/B testing tools today, choosing a cookieless-native tool isn't just a privacy decision — it's a future-proofing decision. Compare your options in our European SMB tools guide.
