Designing consent.

What the screen shows is part of the proof as well.

October 20266 min read

With a handwritten signature, consent rested on an unmistakable gesture: picking up a pen and writing one's name at the bottom of a sheet of paper lying right in front of you. In the digital world, that gesture is replaced by a screen, some text and a button. And each of those elements can strengthen or weaken the proof that the person truly intended to be bound.

The screen is part of the evidence

If someone challenges an agreement, the question will not only be whether they signed, but what they saw when they signed. Were they shown the full document or a summary? Did the button say "Sign" or "Continue"? Were the additional terms in plain view or behind a link? What is shown to the signer, and how it is shown, becomes as important as the signature itself.

What you see is what you sign. If you cannot prove what was seen, you cannot prove what was accepted.

Six design principles

Show the complete document. The signer must be able to read exactly what they are signing, without the interface summarizing or replacing it.
Name the action clearly. The button says what it does: "Sign the contract," not "Next." An action that creates an obligation should not look like an intermediate step.
Separate the consents. Signing a contract, agreeing to the processing of personal data and agreeing to receive electronic communications are different decisions, and each deserves its own acceptance.
Avoid deceptive shortcuts. Checkboxes ticked in advance, hidden decline buttons or artificial deadlines weaken consent and may invalidate it.
Make it accessible. A person who uses a screen reader, or who reads in another language, must be able to understand what they are signing. The W3C Web Content Accessibility Guidelines offer concrete criteria.
Allow time and a way out. The signer must be able to pause, download the document, consult someone or decline, and those options must be as visible as the option to sign.

What the rules say

Different legal frameworks point in the same direction. The European Union's General Data Protection Regulation requires consent to be freely given, specific, informed and unambiguous. The U.S. federal ESIGN Act of 2000 requires that, before paper is replaced by electronic records, consumers be informed and give their consent in a way that demonstrates they can access those records. Consumer protection laws in many countries condemn hidden clauses and misleading practices. None of these rules is specific to signatures, but all of them bear on the screen where the signing happens.

Two screens, the same contract

On the first, the signer sees a three line summary, a box already checked that reads "I accept the terms," and a "Continue" button. On the second, they see the entire contract, an empty checkbox for data processing, another for communications, and a "Sign the contract" button next to one of the same size that reads "Decline." Both produce a technically identical signature. Only the second produces consent that can be defended.

What must be recorded

The consent record

The exact version of the document displayed, identified by its cryptographic hash.
The wording of each acceptance and each button, exactly as it appeared.
The order of the steps and the time of each one.
The options the signer had and the ones they chose.
The language and the channel in which everything was presented.

Key points

In digital signing, consent is designed. Showing the complete document, naming the action clearly, separating acceptances, avoiding deception, ensuring accessibility and providing a way out are principles that strengthen the proof. And what was shown must be recorded with the same precision as the signature.

Sources

← All notes