A gym moves its sign-up to an online form. It asks for date of birth, employer, marital status and “How did you hear about us?”, all of them mandatory. On top of that, the web server writes every request into a log, including the IP address and the email address from the URL, and the log stays for three years. For the contract, the gym needs a name, an address, bank details and confirmation that the member is of age. A.1.4.2 requires the collection of personal data to be limited to what the identified purposes really call for, including where data only accrues on the side.
Every item that is never collected needs no protection, no retention, no access request handling and no deletion. That makes this control the cheapest privacy measure of all: it works at the source, before effort and risk arise.
What does the standard require?
A.1.4.2 opens group A.1.4 of the standard. The shared aim of that group is to design processes and systems so that collection and every later use, from disclosure to disposal, stay confined to what is needed.
- Measure collection against the purpose. The yardstick is the set of purposes from A.1.2.2. For each single item it must be possible to explain that it fits the purpose, stands in reasonable proportion to it and is actually used. Something that “might come in handy one day” fails this test.
- Indirect collection counts. The standard explicitly includes data that arises without a form, for example in web server logs and system logs. Their scope has to be limited too.
- Privacy-friendly defaults. Where collection or processing offers choices, each choice starts switched off. It is only switched on by an explicit decision of the data subject. Pre-ticked boxes and extra features that are active out of the box do not fit this requirement.
In practice
Inventory the collection points. Web forms, app sign-up, the applicant portal, call notes in the CRM, visitor lists at reception, interfaces to partners: every point at which personal data enters the organization goes on a list, each with a reference to the related processing activity.
Justify each field. Map every field to the purpose it serves. Mandatory fields need a justification that names the contract, a legal obligation or the specific service. Typical findings are the full date of birth where an age confirmation is enough, the private mobile number where an email address will do, and salutation or academic title as a required entry.
Include the logs. Together with IT and development, check what web servers, applications, load balancers and analytics tools record. Common adjustments: shorten IP addresses or remove them after a few days, keep email addresses and identifiers out of URL parameters, switch off verbose debug logging in production and set a storage period for logs.
Check the defaults. Newsletters, profile visibility, location sharing, personalization, analytics cookies: anything beyond the core of the service starts switched off. Write the default configuration down so that it does not flip unnoticed during an update.
Anchor it in development. New forms and systems go through a short privacy check before launch. A checklist with three questions works well: which items are collected, what is each one for, and what is preset?
Typical audit evidence
For A.1.4.2, auditors typically expect:
- Inventory of collection points mapped to processing activities, with a justification for each mandatory field
- Logging concept stating the scope and storage period of logs (→ A.8.15 Logging)
- Documented default settings of the applications, for instance as a configuration extract or screenshot
- Privacy check in the development process, for example as an acceptance criterion (→ Secure Software Development Policy)
- Results of the latest review with removed fields or fields made optional (→ Change Log)
KPI
% of data collection points verified for PII collection minimization
Verified means: every field and every recorded item is mapped to a purpose, unjustified items have been removed and the defaults are documented. The target is 100% of the collection points in the inventory. The figure is only as meaningful as the inventory itself: if the logs are missing there, it looks better than the actual situation.
Supporting KPIs:
- Number of mandatory fields per form compared with the previous year
- Share of systems with a defined storage period for logs
- Share of optional features that are switched off by default
Link to the GDPR
A.1.4.2 puts the principle of data minimization from Article 5(1)(c) GDPR into practice: data should be adequate and relevant for the purpose and confined to what is necessary. The yardstick comes from purpose limitation under Article 5(1)(b). The default settings correspond to Article 25(2), which requires, among other things for the amount of data collected, that by default only what each specific purpose needs is processed. For cookies and similar access to end devices, Germany adds Section 25 TDDDG, formerly TTDSG (→ TTDSG Telecommunications Telemedia Data Protection Act). The wider context is on GDPR General Data Protection Regulation.
Related controls
A.1.4.2 is the first filter, and every further limitation builds on it:
- A.1.2.2 Identify and document purpose: provides the purposes against which every collection is measured.
- A.1.4.3 Limit processing: continues the limitation after collection, for access, disclosure and storage period.
- A.1.4.5 PII minimization objectives: sets per purpose how far the type and amount of data are limited.
- A.8.15 Logging: governs what systems record and for how long.
- A.5.34 Privacy and Protection of PII: anchors privacy in the ISMS.
Sources
- ISO/IEC 27701:2025, Annex A, Table A.1, A.1.4.2, and Annex B, B.1.4.2: Limit collection
- Regulation (EU) 2016/679 (GDPR), Articles 5 and 25
- ISO/IEC 27701: Privacy Information Management System: structure of the standard and roles