Do these numbers represent all ecommerce sites in Türkiye?

No. All sites come from the three platforms' own reference lists. Reference lists probably lean toward larger, better-maintained stores (selection bias). For now at least, the results describe these reference stores rather than all of the platforms' customers or ecommerce in Türkiye as a whole.

Does the open dataset contain every number in this post?

No. The measurements of the banner itself (position, buttons, colour, emphasis) are in the dataset. Tracking after the choice, ad signals, banner texts, GTM data and site identities are not; the dataset focuses on how the banner is used and presented. Different datasets can be derived from the collected data; the first one I published focuses on the banner.

Is it always a problem when ad and analytics signals continue after the visitor rejects?

For analytics, not always. Ad signals are a separate matter that needs its own assessment. The purpose of a request, whether the site obtains consent in another way, and how the data is processed on the server are separate questions and not part of this study.

Why are platform and CMP names coded in the open dataset?

The dataset describes the banner's structure: position, buttons, colour, emphasis. Studying that structure does not require knowing the platform or CMP name; the codes are enough to group the rows that belong to the same platform or CMP. This post is about the platforms and CMPs themselves, so the names are given here.

Cookie Banner Dataset Details: Data Collection, Labelling and Signals After the Choice

How the ecommerce cookie banner data I published as an open dataset was collected and labelled, what it shows by platform and CMP, and which ad signals are sent before a choice and after the visitor rejects.

TL;DR

About half of stores on three Turkish ecommerce platforms' reference lists show a cookie banner on the homepage; about a third of these are information-only, two-thirds offer a choice. Most banners come from a few templates, so the CMP or platform template, not the store, mostly sets their format and behaviour. About two-thirds of sites with a choice banner send an ad signal and set an ad cookie before any choice; about 85% of tracking sites keep tracking after rejection.

Create Free Account Sign In

The Rest of This Post Is for Members

You can read this post and the other members-only posts for free.

analytics consent

The Rest of This Post Is for Members

RELATED

Does a Consent Rate Measure User Preference or Banner Design?

You Added a Consent Banner, Your Traffic Dropped 30%: GDPR and KVKK's Impact on Measurement

OpenAI Ads Identity Matching and Privacy: SHA-256, oppref, and opt_out

Advertising Measurement Within Consent Limits: Google's Signal Architecture

T-Soft Cookie Law and Google Consent Mode V2 Incompatibility: Analysis and Solution

GDPR Compliance: Obligations When Your Tracking Tools Collect Personal Data

Personal Data Inventory Under Consent Mode: GA4, CAPI, Newsletter and Webhook Leakage Surfaces

Google Consent Mode v2 Implementation: gtag, GTM, Meta Pixel and Platform Integrations

Tracking Data Between iframe and Parent: postMessage, Consent and Payment Scenarios

Tracing the Source of GA4 and Ads Signals in Shopify