CONTROL is the ENISA privacy-design strategy most directly represented by a cookie control banner. The banner gives a visitor agency over whether optional personal-data processing occurs, and whether preferences can later be changed or withdrawn. Its explanatory text also applies the INFORM strategy, because it should explain what is collected, why, by which means and with which third parties.
These labels describe design approaches, not a conclusion that a particular banner complies with every law. A banner is one interface in a larger processing system; privacy by design and by default also covers data flows, safeguards, defaults, retention and access controls.
The direct answer: CONTROL, supported by INFORM
ENISA’s eight-strategy taxonomy identifies CONTROL as the closest answer. ENISA defines it as providing data subjects with agency over the processing of their personal data. In a cookie interface, that agency appears in actions such as accepting or refusing optional cookies, selecting purposes, changing a choice and withdrawing consent.
INFORM is the companion strategy. A banner, preference centre or layered notice informs people about the proposed processing: which cookie or identifier category is involved, the purpose, the parties receiving data and the technologies used. The same screen can therefore implement both strategies, but its controls primarily perform CONTROL while its explanations perform INFORM. ENISA’s taxonomy is set out in Privacy and Data Protection by Design – From Policy to Engineering (December 2014).
Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minuteWindows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallWhat CONTROL means in a cookie banner
Agency at the moment of choice
A visitor should be able to make a meaningful decision about optional processing. Typical controls include separate switches for analytics, advertising or personalisation, a clearly labelled refusal option, and a way to save a selected combination. A control that merely displays a notice without permitting a decision is informative, but it does not by itself provide the agency ENISA associates with CONTROL.
Agency after the first choice
Control is continuous rather than a one-time click. A persistent “Privacy settings” link, footer control or account setting should let a visitor revisit the choice. Withdrawal should be as easy to find as acceptance, and the system should stop the relevant optional processing after withdrawal, subject to the limits of the applicable legal basis and technical architecture.
Controls must map to real processing
Labels such as “marketing” or “better experience” are not enough if they conceal several unrelated purposes. Each switch should correspond to an actual processing group, and the implementation should honour the saved state across pages and sessions. Necessary functions should not be bundled into an optional switch merely to make the interface appear simpler.
What INFORM adds
INFORM concerns the quality and timing of information supplied when personal data is processed. A useful first layer states, in plain language:
Recommended Free Tools
- what data or identifiers are stored or read;
- why each category is used;
- which vendors or other recipients may receive data;
- how long the data or cookie remains available; and
- where to find fuller notices and contact information.
Long legal text can be placed behind a “Manage preferences” or “Learn more” link, but the first view still needs enough information for a person to understand the choice. Separating the informational and control functions helps diagnose a weak design: a banner may explain every vendor yet offer no refusal path, or provide switches whose purposes are impossible to understand.
CONTROL is not the same as legal compliance
Calling an interface an example of CONTROL is a design classification. It does not establish that the site’s consent, legitimate-interest assessment, transparency notice, security or retention practices satisfy a particular jurisdiction’s law. The banner can be technically correct while the underlying system still collects data before consent, ignores a refusal, or retains data longer than necessary.
The European Data Protection Board’s February 2026 summary describes data protection by design and by default (DPbDD) as a mandatory, continuous GDPR duty for every organisation. It says privacy protections should be built into systems from the beginning and that defaults should be as privacy-friendly as possible. See the EDPB summary (February 2026).
The European Commission’s Obligations guidance similarly places design measures at the earliest stages of processing. Its examples include processing only data necessary for a purpose, retaining it for the shortest necessary time and restricting access. The EDPB’s detailed Guidelines 4/2019 on Article 25 have a final version dated 20 October 2020; check the Board’s site for later updates when making a current-law determination.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
How to evaluate a banner’s design
Use the following questions as a practical review. They are comparison dimensions derived from ENISA and official regulator guidance, not an official scoring scheme.
Information
- Can a visitor understand what processing is proposed and for what purpose before choosing?
- Are vendor names, categories and retention details available at an appropriate layer?
- Do labels describe the consequence of each option rather than using vague marketing language?
Control
- Can the visitor refuse optional purposes without navigating an unnecessarily difficult path?
- Can separate purposes be selected independently?
- Can the visitor later find, change or withdraw the choice?
Defaults
- Are optional cookies and trackers off until the relevant choice is made?
- Does the system limit processing to what the stated purpose requires?
- Are scripts prevented from firing merely because the banner is visible?
Choice presentation
- Are accept, refuse and settings choices visually and verbally clear?
- Does the design avoid misleading emphasis, repeated prompts or other harmful patterns?
- Are consent choices for different purposes kept separate?
Scope
Record which country or region, audience, processing purpose and regulator guidance apply. Rules and regulator expectations differ; do not turn advice written for one scenario into a universal legal test.
UK consent-or-pay guidance as a scoped example
The UK Information Commissioner’s Office discusses meaningful choice in its privacy-by-design guidance for consent-or-pay models. It recommends concise, clear, plain-language explanations, separate consent for different purposes, avoidance of harmful design practices and an easy route to withdrawal. This page addresses UK consent-or-pay scenarios, so treat it as relevant UK regulator guidance rather than a universal rule for every cookie banner or jurisdiction.
Designing and testing a CONTROL-oriented banner
1. Map processing before drawing the interface
Inventory cookies, local-storage keys, pixels, SDKs and server-side calls. Group them by genuine purpose and identify recipients, retention and the event that starts each process. Mark strictly necessary functions separately from optional purposes.
Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Clear out junk files and repair common Windows errorsFree Scan →2. Define privacy-friendly defaults
Before a visitor chooses, load only what is necessary for the requested page. Delay analytics, advertising and personalisation calls. Store a versioned preference record so a policy change can trigger a fresh choice where appropriate.
3. Build two layers
Use a concise first layer for the immediate choice and a preference centre for purpose-level details. Keep refusal and acceptance available without forcing a person through unrelated screens. Add a persistent route back to settings.
Rank #3
4. Verify behaviour, not just appearance
Test a clean browser profile with JavaScript and network logs enabled. Confirm that optional requests do not occur before consent, that each switch changes the correct requests, that a refusal persists after reload, and that withdrawal stops future optional calls. Check keyboard navigation, focus order, contrast and screen-reader labels.
5. Capture evidence without collecting visitor data
For a review record, capture the initial banner, each preference state and the post-withdrawal view. Redact identifiers and avoid sending real user data to a testing service. A screenshot proves what the interface displayed at one moment; network traces and configuration records are needed to prove processing behaviour.
Common design failures and fixes
The banner informs but does not control
Symptom: it describes cookies and offers only an “OK” button. Fix: add a meaningful refusal or purpose-level choice and a later withdrawal route.
Control exists but is not informed
Symptom: switches say “improve experience” with no explanation. Fix: connect each control to a plain-language purpose, data category and recipient explanation.
Optional scripts run before a choice
Symptom: analytics or advertising requests appear during the initial page load. Fix: gate script insertion and server calls on the stored preference, not on banner visibility.
Refusal is harder than acceptance
Symptom: acceptance is one click but refusal requires several screens or repeated prompts. Fix: provide comparable paths and remove unnecessary friction, while still allowing granular choices.
Free tools Windows power users keep installed
One-click scans. No signup required.
Withdrawal changes the button but not the system
Symptom: the interface reports “off” while trackers continue firing. Fix: revoke vendor signals, stop future calls, clear or limit optional identifiers where appropriate, and test after reload.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Where a screenshot API fits
A screenshot service can document the visible states of a banner in CI or a compliance review, but it cannot decide whether a site’s processing is lawful. Choose a capture that can wait for the banner, set a viewport, load the full page when needed and avoid recording unnecessary personal data.
ScreenshotNeo is a website screenshot API and MCP server. It removes cookie or consent banners, newsletter popups and chat widgets before capture when you enable those steps, so it can produce either a clean page shot or a deliberately captured banner state. Only clean shots are billed: bot checks, CAPTCHAs, blank pages, timeouts, failed loads and cache hits cost nothing, and response headers report the page verdict and billing status. Its options include selector waits, delays or network-idle waits, custom JavaScript and CSS, device and viewport settings, full-page lazy-image loading, element capture, request blocking, cookies and headers, caching, signed links, async jobs and bulk capture.
Or skip the browser setup
One GET request can capture a page after you configure the desired wait and cleanup options. The examples below use the documented endpoint; see the ScreenshotNeo API documentation for parameters and response headers.
cURL
curl -G "https://api.screenshotneo.com/v1/shot" -d access_key=YOUR_API_KEY --data-urlencode url=https://stripe.com -o shot.webpPython
import requests
r = requests.get("https://api.screenshotneo.com/v1/shot", params={"access_key": "YOUR_API_KEY", "url": "https://stripe.com"}, timeout=90)
open("shot.webp", "wb").write(r.content)Node.js
const q = new URLSearchParams({ access_key: 'YOUR_API_KEY', url: 'https://stripe.com' });
const res = await fetch(`https://api.screenshotneo.com/v1/shot?${q}`);For a banner audit, combine a selector wait with custom JavaScript that opens the preference centre, or capture separate URLs and states in an isolated test account. For a clean report image, enable removal of consent UI, popups and chat widgets; for evidence of the control itself, leave those removals off and capture the first-load state. An MCP server provides the tools take_screenshot, get_page_info and capture_pdf to Claude, Cursor and other MCP clients, so an AI agent can perform the same checks.
ScreenshotNeo’s Free plan includes 1,000 screenshots each month with no card. Paid plans start at $5 for 3,000 shots; yearly billing gives two months free, and every feature is available on every plan. Sign up free to capture your first 1,000 screenshots without a card.
FAQ
Is a cookie banner always an example of CONTROL?
No. It is an example of CONTROL when it gives meaningful agency over optional processing. A notice with no real choice mainly performs INFORM.
Best Value
Does CONTROL require a particular button layout?
No. ENISA names a strategy, not a mandated visual arrangement. The adequacy of a layout depends on the purposes, audience, jurisdiction and implementation.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Fix the driver behind crashes, sound loss and screen glitches3Repair Windows errors before they cause bigger problemsCan necessary cookies be refused?
That depends on what the cookie does and the applicable law. Classify a function as necessary based on its actual purpose, not on a desire to avoid offering a choice.
Does a screenshot demonstrate GDPR compliance?
No. It documents the interface at capture time. Compliance also requires evidence about requests, defaults, legal basis, retention, security and governance.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
