← Breach analysis by the Dual Vulnerabilities framework

BrewDog Mobile App PII Exposure (Unattributed, 2020)

2020 · United Kingdom · C - Manufacturing · Personal data exposure via insecure mobile app API

Analysed 2026-11-07 by Hermes

Open this case in the tool →

What happened

Personal data exposure via insecure mobile app API

Dual vulnerability type

Expertise to Execution

Evidence coverage: 3/6 stages, weight 20/60

Decisive stage: entry (clearest evidence) · Alternative: Strategy to Leadership

Why this classification

(1) BrewDog's Equity for Punks app fronted GDPR-class personal data (names, dates of birth, addresses, phone numbers, shareholdings, discount IDs) of over 200,000 shareholders and many more customers, and the relevant expectations were specialist-standard ones: API bearer tokens are issued dynamically after a successful authentication, ordinary functional API testing and security review should validate an API design, and the GDPR/ICO framework governs notification when such data is at risk. (2) In practice the company hard-coded the same static bearer token into every app install from version 2.5.5 (March 2020), which made request authorisation useless and let anyone who appended another customer's (predictable, non-random) ID pull that person's whole record; the exposure lasted about 18 months, surfaced only through external researchers who inspected the published app, and after notification four fix attempts still failed before version 2.5.13 (27 September 2021). The company then chose not to notify its 200,000+ shareholders and customers, asked the finders not to name it, and told its shareholders forum-style that no ICO notification was required. (3) The classification rests on the entry/exposure stage, the only load-bearing gap: the knowledge that was not acted on sat with technical specialists (dynamic OAuth2 tokens, routine API testing - explicitly named by the finding firm and an independent API-security expert as what BrewDog did not do), and the chain broke in execution - a control needing no leadership decision was simply never realised in BrewDog's own code and its release verification repeatedly failed to show that it worked. The later non-disclosure is a documented doctrinal-origin, leadership-break gap (a secondary Strategy to Leadership pattern), but it is contributing rather than decisive, so no tie-break applies. No attacker entry is established or load-bearing; this is a self-created exposure. RM29 (Expertise to Execution lens): BrewDog's own developers did not apply the known secure API practice - a bearer token hard-coded in the app source plus predictable customer IDs (secrets in code, broken access control on exposed APIs) - so the expertise existed but was not executed in the company's own development work (causal role: Enabling). RM34 (Expertise to Execution lens): testing, review and monitoring that the organisation's own specialists know how to stand up were absent or ineffective for about 18 months, so the company never detected its own exposure and learned of it from an outside party (causal role: Obscuring). RM12 (Expertise to Execution lens): the company's own fix verification materially misrepresented control effectiveness - three releases shipped as fixes while a static key still allowed bulk retrieval of discount codes, and only the external tester's six build checks showed when the control actually worked (causal role: Perpetuating). RM39 (Expertise to Execution lens): because every request carried the same valid-looking token, experts could not establish the scope of any abuse or who had been affected, so the company's no-evidence-of-breach disclosure position rested on a scope no forensic capability could have verified at the time (causal role: Constraining).

Evidence by stage

Risk markers

Sources

Notes

Row describes the BrewDog mobile-app API bearer-token exposure reported on 8 October 2021; the description in the case file matches the stated victim. Year set to 2020 because the exposure began with app version 2.5.5 (March 2020) and the method takes the first year; public disclosure was October 2021. The source_url in the case file carries a stray trailing ' /'; the live TechRadar article opened without it. BrewDog's claim of no evidence of exploitation is contested by Pen Test Partners as unverifiable (every request used a valid-looking identical token) - recorded as an egress non-gap with that caveat rather than as an egress gap, since no exfiltration is confirmed. No ICO enforcement action against BrewDog over this incident was found in the reporting opened for this row. A separate, later BrewDog data-handling matter (2026 ICO complaints about founder James Watt emailing former shareholders) is a different incident and was not used for this classification.