Zero-knowledge
What Zero-Knowledge Encryption Protects, and What It Doesn't
Zero-knowledge encryption protects the content of your data from the provider: they store it but can't read it. It doesn't hide metadata, rescue a hacked device, or make you compliant on its own. It's a strong tool for one job, not a force field. Here's the honest line.
Zero-knowledge encryption protects the content of your data from the provider. They store it, but they can't read it. What it does not do, on its own, is hide metadata, protect a device that's already compromised, or make you compliant with the law. It's a strong tool for one specific job, not a force field.
That distinction matters, because the gap between what people assume it covers and what it actually covers is where the nasty surprises live.
What it protects
The core promise is real. Your data is encrypted before it reaches the provider, and the key never lands on their servers, so:
- The provider can't read your content, and neither can anyone who breaches their storage. A stolen database of ciphertext is useless without the keys.
- A subpoena turns up nothing readable. The provider can only hand over what it has, and what it has is scrambled. This is a property of the design, not a legal favour.
- It's the strongest form of collecting less. The provider never holds the plaintext, so there's nothing there to log, mine, sell, or lose.
What it doesn't (the part the marketing skips)
- Metadata. Who contacted whom, when, how often, from which IP address. Zero-knowledge protects the contents of a message, not the record that the message happened. That trail usually still exists.
- A compromised device. If your laptop has malware, or the app's own code is tampered with, the data is readable right where you work with it. Zero-knowledge moves trust to the endpoints; it doesn't remove it.
- Your legal obligations. Encryption isn't a lawful basis for collecting data, and it isn't consent or transparency. You still owe your users the GDPR basics.
- Anything after decryption. Once the recipient unlocks the data, it lives on their device, outside the provider's reach entirely.
Protects vs doesn't, at a glance
| Zero-knowledge encryption protects | It does not, by itself |
|---|---|
| Content from the provider | Metadata: who, when, how often, IP |
| Content in a server breach or subpoena | A hacked or malware-infected device |
| You from the provider mining your data | Your GDPR lawful basis and duties |
| — | Data after the recipient decrypts it |
How to use it well
Use zero-knowledge for what it's genuinely good at, keeping content out of the provider's hands, and cover the rest deliberately. Collect less metadata where you can. Keep your own devices patched. Do your GDPR homework regardless. And set retention so decrypted copies don't accumulate. Treat it as one strong layer, not the whole wall.
Where doconvoy fits
doconvoy is honest about this line. Your content is encrypted so doconvoy can't read it, while some metadata, such as timestamps, a verified email, and an IP address when available, is stored to run the access controls and the audit trail. Every request also carries an expiry, so data doesn't linger after its purpose. You get the strong part of zero-knowledge without being told it does things it can't.
Content we can't read, with the metadata and expiry controls a real workflow needs.
Strong where it counts, honest about the restRelated: Zero-knowledge encryption, explained without the jargon · Your SaaS shouldn't know your secrets · The audit trail · GDPR-sensitive data workflows
Common questions
Does zero-knowledge encryption hide metadata?
No, not on its own. It protects the content of your data from the provider, but metadata like who sent what, when, how often, and from which IP can still be recorded unless it's protected separately. Zero-knowledge protects the message, not necessarily the fact that a message happened.
Does zero-knowledge encryption make me GDPR compliant?
No. It's a strong technical measure, but it doesn't give you a lawful basis, purpose limitation, or transparency. Those are your responsibilities as the organisation collecting the data. Encryption helps you meet the security principle; it doesn't cover the rest.
Can zero-knowledge encryption protect a hacked device?
No. If the sender's or recipient's device is compromised, or the app's code is tampered with, the data is exposed where it's readable. Zero-knowledge keeps content off the provider's servers; it can't defend an endpoint that's already been breached.