A LinkedIn connection is not consent: What an outreach agent must clarify before the first contact

An outreach agent is tempting because it is so easy to describe: it finds suitable contacts, connects with them on LinkedIn and, once the connection request is accepted, writes a friendly, personal message or email. Sales saves hours and the pipeline fills up. Anyone working this way often assumes that the connection is a kind of invitation. After all, you are now "connected".

This is exactly the assumption that a case challenges, which made the rounds in the press and in specialist articles in July 2026. A German local court (Amtsgericht) has reportedly ruled that automated advertising emails to LinkedIn contacts are unlawful, and awarded the claimant an injunction and reimbursement of costs. On this line of reasoning, a connection alone is not consent to advertising.

This article sets out what is documented, what we were unable to verify, and what follows for companies that are planning or already using an outreach agent. It addresses a timeless topic, because the legal position applies regardless of any individual tool. It is expressly not legal advice. For your specific case, you need a lawyer who specializes in competition law.

What was reported and what we checked

We rely on two sources, and we state openly what each of them provides.

The first is a report by Börse Express dated 5 July 2026, titled "LinkedIn-Werbung: Amtsgericht verbietet automatische Mails an Kontakte" (LinkedIn advertising: local court prohibits automatic emails to contacts). It cites case number 23 C 120/25 and states that the decision found in favor of a claimant, who receives an injunction and reimbursement of costs. The report places the case within the strict German rules on advertising emails under § 7 of the German Unfair Competition Act (UWG) and notes that modern cold-email tools change nothing about this, "no matter how well they personalize or research addresses". The report is short. It names neither the court in detail, nor the date of the judgment, nor any reasoning. It also points out that parts of it were created with AI and reviewed editorially.

The second source is a specialist article by a law firm titled "Zulässigkeit von Werbenachrichten via Social Media" (Permissibility of advertising messages via social media), last updated on 16 January 2026. Among the decisions on direct messages in social networks, it lists the judgment of the Amtsgericht Düsseldorf of 20 November 2025 (Az. 23 C 120/25). The case number matches the one in the report. From this combination it follows that it is very likely the same judgment. We have not seen the judgment itself, and we do not reproduce anything that might appear only there.

"A connection in a social network is not consent to receiving email advertising." (Specialist article on the permissibility of advertising messages via social media, updated 16 January 2026)

This is the statement that matters. It comes from a law firm's article, not from the explanatory memorandum to a statute. We therefore present it as the source's legal opinion and not as settled law for every individual case.

The legal classification in three sentences

According to the specialist article, the following applies:

• Direct messages are "electronic mail". Under § 7 (2) UWG, advertising by electronic mail without the recipient's prior express consent is an unreasonable nuisance. The article refers to a judgment of the Court of Justice of the European Union of 25 November 2021 (Az. C-102/20) and a decision of the German Federal Court of Justice (Bundesgerichtshof) of 30 January 2020 (Az. I ZR 25/19, "Inbox-Werbung I"), under which the term is interpreted broadly. According to these, LinkedIn messages are included as well.
• Consent must be express and given in advance. According to the article, a "like" or a connection is not sufficient. The double opt-in procedure is the usual way to document consent.
• The burden of proof lies with the advertiser. Whoever advertises must be able to prove that consent existed. The recipient may simply deny it. If the proof fails, then according to the article even a single unwanted message may have to be stopped.

In addition, there is the exception for existing customers in § 7 (3) UWG. The article describes it as narrow, tied to several requirements that must all be met at the same time, and generally not applicable to other companies in the network. Knowing someone only from the network does not make them an existing customer.

Availability: a legal boundary, not a product

This is not a new feature that you can activate or order. It is a legal boundary under German competition law, which applies directly to outreach conducted in Germany. Austria and Switzerland have their own rules on unsolicited advertising. We have not checked them for this article and make no statements about them. If you are active in those countries, have the legal position there clarified separately.

It is also important what the specialist article leaves open: it does not clearly answer whether a plain connection request without text counts as advertising. A connection request with advertising content, on the other hand, it treats as unlawful advertising. And it points out that, in practice, enforcement works mainly against providers in Germany or the EU. Nobody should read this as an invitation to sidestep the rules in cases of doubt. It merely shows that the legal boundary applies even when the risk sometimes stays hidden in everyday practice.

Why this hits an outreach agent in particular

A person who writes ten messages a day considers for each one whether it fits. An agent, if you let it, writes hundreds, and it does not ask that question by itself. Three characteristics make automated outreach legally sensitive:

• Scale: A flaw in the process repeats with every contact. If the process has a gap, it is present many times over before anyone notices.
• Speed: The connection is accepted, and minutes later the message goes out. There is no room for a check in the process unless you have built one in.
• Provability: The burden of proof lies with the advertiser. An agent that does not record where a contact came from and what consent existed leaves you nothing to argue with in a dispute.

On our website, we describe the Outreach Agent as an agent that writes personalized cold emails and LinkedIn messages, researches prospects and plans sequences. Kai is the outreach lead on the team, setting strategy and sequences. Michael researches contacts from public sources, directories and LinkedIn. For Kai, the site states that he respects opt-outs immediately and permanently and adheres to LinkedIn limits, and that his sequences can be paused at any time. Those are sensible foundations. A consent check for the first contact is expressly not described there, and we do not claim one. It is an approval step that you must define before sending.

What an agent with limits looks like

The difference between an outreach agent without and with an approval step can be shown with a simple process. We describe it as a procedure, not as a result at any particular customer.

Without an approval step

1. The agent finds a contact and sends a connection request.
2. The contact accepts.
3. The agent automatically sends an advertising message or email.

Nowhere in this process does anyone ask whether express consent exists. The risk grows with every message, and in a dispute the burden of proof lies with you.

With an approval step

1. The agent finds a contact and documents the source and reason.
2. Before the first advertising contact, a person checks whether documented consent exists, for example through a double opt-in. If none exists, it stays a draft or the contact is handled differently.
3. The agent prepares the message as a draft, and a person approves it.
4. The agent logs sending, consent and opt-outs.

That emails are created as drafts and not sent directly corresponds to what we describe on our website about the approval steps of our agents. This is not a guarantee against legal violations. But it gives you the moment in which a person can ask the question.

Which questions to clarify before the first contact

If you are planning or already using an outreach agent, these questions help as a checklist. They do not replace a legal review, but they show where a conversation with your lawyer is needed.

• Which channel? According to the specialist article, email, LinkedIn direct messages and connection requests with text are all to be classified as electronic mail.
• Which consent? Is there express, previously given and documented consent for this contact? What does the proof look like?
• Which content? Does the first message recognizably contain advertising? Then the strict rules apply. Whether and when a message is not advertising is a question for legal advice, not for a marketing gut feeling.
• Which existing relationship? Is the contact an existing customer within the meaning of the law? For that, several requirements must be met at the same time. A network contact is not one.
• Which opt-out? How quickly and how permanently is an objection implemented, and is it technically anchored in the agent?
• Which proof? In a dispute, can you demonstrate where the contact came from, when they consented and what was sent to them?

Research is not consent: the role of lead sourcing

A common mistake sits one stage earlier. A research agent like Michael finds contacts in public sources and enriches them with data. That is research. It says nothing about whether the contact wants to receive advertising. According to the specialist article, the fact that an address can be found publicly, or that someone maintains a profile in a network, does not replace consent.

In practice, this means that sourcing and outreach are separate steps with separate rules. The sourcing step may research, sort and record reasons for contacting someone. The outreach step needs its own check. If the two merge into one continuous process without a pause, the very point at which a person could intervene is missing. It is therefore worth building a visible handover point into the process: the list of contacts and reasons is ready, and only after a check does it become outreach. This is not extra effort to be shunned. It is the point at which the agent delivers its work and the person decides.

Proof also includes recording the origin of every contact: from which source, on which date, for what reason. Anyone who has to show in a dispute where an address came from will be grateful for a log that the agent can keep anyway. Managing objections belongs here too. A contact who no longer wants to be contacted must be blocked in all parts of the process, not only in the sequence in which they last got in touch.

What this case does not mean

It would be wrong to conclude from the case that outreach with agents is inadmissible in principle. None of our sources says that. The issue is the requirements for a first contact with advertising content. Contact at the recipient's request, replies to inquiries, outreach based on documented consent and channels subject to other rules are something different. The specialist article names, for example, paid ads and company pages as ways in which advertising in social networks is possible without consent.

It would be equally wrong to read the topic as evidence of a particular success rate or of experience with customers. We cite no figures on this because we have no verified ones that we could substantiate here. The message of this article is narrower and therefore more robust: if an agent makes the first contact, someone must first define on what basis it is allowed to do so.

What you should check this week

1. Write down the process: Describe in five sentences how your outreach works today, from finding contacts to the first message. Mark whether a person checks at any point.
2. Clarify the consent status: For a sample of your contacts, check whether documented consent exists and where it is stored.
3. Stop automatic first sends: Where an agent today automatically sends an advertising message after a connection, switch to draft until the legal position for your case has been clarified with a lawyer.

The difference: an agent with limits and an approval step

An outreach agent remains a valuable tool when it has a limit at the right point. An agent without an approval step sends what the process tells it to, and it does so quickly and reliably, even when the process is flawed. An agent with limits and an approval step stops at the point where a decision is needed and hands over to a person. That costs you a brief moment per contact and, in case of doubt, saves you a cease-and-desist letter, a dispute and the reputational damage among people you actually wanted to win over.

Sources

• Börse Express: "LinkedIn-Werbung: Amtsgericht verbietet automatische Mails an Kontakte", 5 July 2026: boerse-express.com
• Law firm (ra-plutte.de): "Zulässigkeit von Werbenachrichten via Social Media", updated 16 January 2026: ra-plutte.de
• Amazing Agents: description of the Outreach Agent, Kai and Michael agents on amazing-agents.com