Opens the cited record in a new tab

Coop and Meny apps: a red Accept all button and tracking on legitimate interest drew serious criticism

Denmark's data protection authority looked inside two supermarket apps. Coop's consent screen made "Accept all" red and "Accept selected" white, which the regulator called nudging. Meny tracked in-app activity without asking, and bundled several marketing purposes, including sharing with partners, into a single consent.

Regulator
Datatilsynet (Danish Data Protection Agency)
Law
GDPR Articles 4(11) and 6(1)
Stage
Decisions issued: serious criticism and criticism; no fine; both companies asked to report on corrective measures. Coop decision (published 10 November 2026)

The problem

In July 2025 Datatilsynet opened inspections of how grocery chains process personal data in their shopping apps, a planned focus area for 2025. It selected Coop Danmark's Coop app and Dagrofa's Meny app (Coop decision; Dagrofa decision).

Both apps used the phone's built-in location data to work out which store a shopper was in for scan-and-pay. Coop also used location to show content and marketing for the nearest store. Dagrofa collected users' in-app interactions through cookies and similar technologies for statistics, improvements, troubleshooting and games, relying on legitimate interest. The consent screens for marketing were also examined.

What the regulator found

CompanyFindingLevel
Coop and DagrofaLocation data for scan-and-pay had no valid legal basis. It was not necessary, because a QR code at the store entrance would do, so consent was required (Article 6(1))Serious criticism
CoopLocation data used for marketing (whether the user was within 200 meters of a Coop store, later the nearest store) could not rely on legitimate interest; consent was requiredSerious criticism
CoopMarketing cookie consent was invalid: "Accept all" was red and "Accept selected" was white, which the regulator found nudged users to accept everything (Articles 6(1)(a) and 4(11))Criticism
DagrofaOne combined consent covered several marketing purposes, including disclosure to partners such as social media; users could not choose purposes, and the first layer did not clearly say data went to partnersCriticism
DagrofaIn-app interactions collected with tracking technologies for statistics, improvements, troubleshooting and games could not rely on legitimate interest; consent was requiredSerious criticism

The regulator accepted other processing, such as membership administration, Coop's climate-footprint consent and Dagrofa's analysis. Sources: Coop decision, sections 1 and 3; Dagrofa decision, sections 1 and 3.

Technical issue

  • Consent screen design: in the Coop app, "Accept all" was highlighted in red and "Accept selected" was white. Coop updated the screen after its first response on 15 August 2025; the decision assesses the version that existed then.
  • Bundled consent: the Meny app used one checkbox for email and push marketing, and its partner sharing (target audiences plus a first-party ID such as a CDP ID or hashed email sent to ad networks and social media) was only explained through the privacy policy link.
  • Tracking without consent: the Meny app collected user behavior and app interactions, IP address and device metadata through cookies or similar technologies on legitimate interest.
  • Device location: both apps read the phone's location to identify the store, although a QR code alternative existed.

Inth's technical interpretation: give Accept and Reject or Accept selected the same visual weight, ask for each purpose separately, and say on the first screen when data goes to partners. In-app analytics SDKs that go beyond what the app needs to work should wait for consent, just like web trackers.

Illustrative pattern, not the defendant's code. Generic consent API, not a specific c15t call.
// Unequal buttons and analytics on legitimate interest
showBanner({ acceptAll: { color: 'red' }, acceptSelected: { color: 'white' } });
analytics.init(); // runs before any choice

// Equal choices, one purpose at a time, SDK gated on consent
showBanner({ acceptAll: { style: 'primary' }, rejectAll: { style: 'primary' },
  purposes: ['measurement', 'marketing', 'partner-sharing'] });
if (consent.has('measurement')) analytics.init();

Timeline

DateEvent
3 July 2025Datatilsynet tells Coop and Dagrofa they have been selected for inspection
14 and 15 August 2025First responses from Dagrofa and Coop; Coop later updates its consent screen
October 2025Meny's scan-and-pay feature is phased out; Coop changes how it uses location for marketing
3 August 2026Decision on Dagrofa (Meny app)
5 October 2026Deadline for Dagrofa to report on corrective measures
2 November 2026Decision on Coop (Coop app)
4 November 2026Deadline for Coop to report on corrective measures
10 November 2026Datatilsynet publishes a press release on both decisions

Outcome

Datatilsynet issued serious criticism and criticism and asked both companies to explain the measures they are taking to bring the processing into line with the rules. According to the press release, both companies accepted the criticism and have started specific measures (Datatilsynet press release). No fine was imposed.

What it cost

ItemAmountSource
FineNoneCoop decision; Dagrofa decision
OrdersReport on corrective measures; obtain valid consent for location data, marketing and app-interaction trackingCoop decision; Dagrofa decision

Why it matters

  • Color counts. A regulator found a red "Accept all" next to a white alternative enough to make consent invalid.
  • Apps are not exempt. In-app analytics and game tracking collected with SDKs needed consent, as with the app tracker in CNIL v. Fnac.
  • One checkbox is not consent to everything. Bundling marketing purposes and partner sharing into a single consent failed the specificity test.
  • Less intrusive options matter. Because a QR code could identify the store, using phone location without consent was not necessary.

Related pages: Unequal choices or identity checks, Notice, consent scope and assent, Consent before collection, GDPR and ePrivacy cookies.

How Inth catches this

  • Consent banner

    Can show Accept and Reject with equal weight, ask for each purpose separately, name partner sharing on the first layer, and hold analytics and marketing until the matching purpose is accepted.

  • Website audit

    Loads web pages and records which tracking requests fire before a choice and after a refusal, so you can see whether a refusal actually stops them.

Out of reach

  • Inth does not control whether an app reads device location, it does not judge banner design for you, and it cannot decide whether legitimate interest applies to a given processing purpose.

Consent records with the exact wording shown, choice, purpose, country and policy version.

Start a free scan

Checks for your own systems

Your ticks stay on this page. Inth does not save or send them.

0 of 6 done

In the repository

0 of 2 done
0 of 2 done

With the people responsible

0 of 2 done

Sources and limits

The decisions are in Danish; quotations and button labels are Inth's translations. Whether the companies sent their reports, and what they said, is not public in these sources.

Source review: 11 November 2026. This page describes the cited record and procedural stage. It does not audit the company's current systems. Inth wrote the technical checks; they cannot decide which laws apply or prove compliance.