When a marketing message looks like an invoice, who answers for it?

What happened

In October 2024, Sweden's consumer ombudsman (Konsumentverket) criticised insurance broker Söderberg & Partners for sending marketing messages through Kivra, the digital mailbox service, in a way that made them look like invoices ^[1]. The messages appeared under the category "invoice", with a due date and a false urgency that suggested the recipient had to act. The ombudsman said this violated the Marketing Practices Act, which bans misleading commercial messages ^[2].

Söderberg & Partners said it was a technical error on Kivra's side, and that it had not intended the messages to look like invoices ^[3]. Kivra, for its part, said the sender chooses the category and that it had sent repeated warnings about the rules ^[4]. The ombudsman's position was clear: the sender is responsible for how the message appears to the consumer, regardless of where the error originated ^[2].

It is a bit like putting the wrong label on a parcel. You might blame the courier for sticking it on, but if you sent the parcel, you are the one the recipient will call when it arrives at the wrong address.

Who is responsible when the format deceives?

The Marketing Practices Act says that a commercial message must not be misleading, and that the trader is responsible for what it sends ^[2]. If a message is configured to appear as an invoice when it is marketing, the trader cannot simply point at the platform and say the platform did it.

This is not new law. What is new is how easily a configuration mistake, or a poorly chosen dropdown option, can make a lawful message unlawful. Kivra and similar platforms let the sender choose how a message is categorised and presented. If the sender picks "invoice" for a newsletter, that choice is on record somewhere, and the question in a dispute becomes: who made it, and when?

The evidence question

In this case, Söderberg & Partners said Kivra made the error, and Kivra said it had warned the sender. Both sides had an interest in the question of what was configured, by whom and when. But what neither side appears to have had is a neutral, checkable record of the settings at the time the messages were sent.

Without such a record, the dispute becomes a matter of credibility. Each party points at the other, and a regulator or a court has to decide whom to believe based on emails, screenshots and statements that each side controls.

This is where a sealed record changes the conversation. If the sender's system sealed a record of each message as it was sent, including the category, the subject line and the timestamp, that record could later be checked by anyone. If the platform sealed a record of the sender's account settings and the warnings it sent, that record could be checked too. Neither party would have to trust the other's version, because both would have their own checkable evidence.

What this means for you

If you send commercial messages through a digital platform, three questions are worth asking now:

1. Do you know which category and format settings your messages use, and can you show it?
2. If a regulator or a customer questions how a message appeared, do you have your own record, or only the platform's word?
3. When something goes wrong, can you show when you were warned, what you configured and when you fixed it?

The law puts the responsibility on the sender. A sealed record gives the sender the evidence to back up what it says.

A checkable record of what was sent

Here, the thing to seal is each outbound message as it leaves your system: the recipient reference, the message content or fingerprint, the category setting, the subject line and the timestamp. Sealing takes a fingerprint of the record, not the message itself, so the content never leaves your systems.

If a message is later questioned, you can show the regulator, the platform or a court exactly what you sent and how it was configured, in a record that anyone you share it with can check. It doesn't prove that the message was lawful or that the recipient saw it, but it does show what you did on your end, unchanged since the day you sent it.

The how-to is straightforward. Your sending system calls WitniumChain through the REST API, the SDK or the MCP server, sending a fingerprint of the message record. You get a witness ID back. To check later, anyone calculates the fingerprint again and compares it with the sealed one. The first 2,000 witnesses a month are free. Read the docs.

How it would work

1. At send: your system seals a record of the message, including category and settings.
2. Later: if questioned, you share the record and the witness ID.
3. Anyone checks: they calculate the fingerprint and see that it matches, proving nothing has changed.

What to watch

The Marketing Practices Act already applies, and the consumer ombudsman enforces it. The Digital Services Act, which came into force across the EU in 2022 and applies fully to large platforms from 17 February 2024, sets out transparency duties for online platforms ^[5]. Disputes about who configured what, and when, are likely to become more common as regulators look more closely at how platforms and senders share responsibility for what consumers see.

For anyone sending messages at scale through a digital mailbox or a platform, the question is no longer whether you can be held responsible for a format that misleads. The question is whether you can show what you actually did.