← All insights Technology

Sharing your technology under an NDA: what the signature actually secures

An investor, a prospective partner or a contractor asks to see how the thing works. The instinct is to get a confidentiality agreement signed and then talk freely. The agreement is worth having. What it secures is a claim after the event, and a claim has to be proved by the party bringing it.

A confidentiality agreement is a promise about behaviour. It does not stop information moving, and it cannot recall it. What it does is give the disclosing party a right to complain, in England and Wales, where information covered by it is used or passed on in a way the agreement does not permit. Everything after that depends on whether the complaint can be made out.

Making it out means proving things that are difficult to assemble afterwards: that the particular information was confidential, that this particular person received it under this particular agreement, that what they then did with it fell outside what was allowed, and that the business suffered something a court can put a value on. Signing the document supplies none of that. The material that would supply it is created at the moment of disclosure, by whoever happens to be handing the information across, usually with no thought of a dispute at all.

Which is why the worth of an agreement is largely settled before it is signed, by what is put on the other side of it and in what form.

What the agreement has to be able to identify

Definitions do the real work. There are broadly two approaches, and each fails in its own way. A definition drawn by category, covering anything relating to the business and its technology, is easy to sign and hard to rely on, because a court asked to restrain somebody's use of information wants to know with precision what information is meant. A definition drawn by reference to what was marked or confirmed in writing is precise on documents and weak on everything else.

The second failure is the one that bites in technology. The valuable disclosure is frequently oral: the demonstration, the answer to a sharp question, the architectural explanation given at a whiteboard because the alternative was a document nobody wanted to create. Whether the agreement reaches that, and what has to happen for it to reach it, is a drafting question decided long before the meeting.

Then there are the carve-outs, which every such agreement contains and which are read by almost nobody: information already public, information the recipient already had, information obtained from a third party without restriction, information developed independently, and disclosure compelled by law or by a regulator. Independent development is the one that does the damage here. A counterparty worth talking to often has adjacent work of its own under way, and that carve-out is the mechanism through which they will later say so.

The recipient is rarely a single person

The agreement binds the entity that signed it. The information travels further than that entity almost immediately, and how far it may properly travel is the substance of the negotiation.

An investor needs to show material to its own advisers, to people within the fund, and sometimes to specialists who sit outside it. A corporate counterparty has group companies, and the person you met may not be the person who decides. A contractor has subcontractors and, frequently, a team in another country. Who is permitted to receive the information, on what terms they are told about it, and who answers for them if it is misused, are the clauses that get skimmed because they look administrative.

Two commercial observations sit alongside that. The first is that established investors commonly decline to sign anything at a first meeting, and that refusal is generally a function of how many similar businesses cross their desk rather than a signal about them. A first conversation can usually be held without disclosing the part that needs protecting. The second is less comfortable: a counterparty who signs whatever is put in front of them, without a single comment, has usually not read it and will not have circulated it internally either. That is precisely the situation in which the material later reaches somebody inside their organisation who never knew an obligation attached to it.

Loss is the part nobody plans for

What a business in this position actually wants is for the use to stop, and to stop quickly. In England and Wales interim relief of that kind is discretionary. The court considers how promptly the applicant moved once it knew, and how the applicant itself has behaved. Time spent internally deciding whether to make a fuss is not neutral.

Money is harder still. Putting a figure on what a misused disclosure cost is genuinely difficult where the product was never sold, where the counterparty's own work is entangled with what they received, or where the advantage taken was a head start rather than a copy. Agreements often recite that damages would be an inadequate remedy and that injunctive relief is appropriate. That recital is useful, and it does not decide the question, because the court forms its own view.

There is a commercial layer beneath the legal one. Suing an investor, or a partner in a small sector where everyone eventually meets everyone, is a decision about more than the merits. The consequence founders tend not to anticipate is having to explain to the next funding round why there is live litigation with a name the room recognises. None of that is a reason to accept misuse. It is a reason to be realistic at the point of disclosure about what enforcement would involve.

Deciding what goes across

The cheaper control lies in what is shown, and in what form it is shown.

In technology the concept is usually not the protected thing. What has value is the particular solution to a particular problem: the approach that made something work at a cost or a speed nobody else has reached. Those two can very often be separated, and a disclosure that establishes credibility does not have to be a disclosure that confers the ability to reproduce. Where the line falls is specific to the technology and to the counterparty, and it is a judgement worth taking seriously rather than a formality.

Form matters as much as content. Access that can be withdrawn is a different proposition from a copy that cannot be recovered, and a demonstration is a different proposition from a delivery. What happens at the end of the relationship, whether material must be returned or destroyed, and whether backups and archived copies are quietly excluded from that obligation, is settled by drafting that is usually left as it came out of the template.

Disclosure has consequences beyond the contract

Confidentiality is not only a matter of what the counterparty may do. It bears on what the technology itself remains capable of being.

An invention made available to the public before an application is filed will generally no longer be new, and novelty is a requirement for a patent in the United Kingdom and in most systems abroad. A disclosure made under an obligation of confidence is usually treated differently from one made without. Whether a given conversation was in fact confidential can therefore matter for reasons that have nothing whatever to do with the counterparty's honesty, and the order in which a business talks and files is a real decision rather than an administrative one.

The same point runs through the protection of commercially sensitive information generally. Information is protected as a trade secret only while it retains the character of a secret, and where reasonable steps have been taken to keep it that way. Disclosing widely, or under arrangements that are weak or inconsistently applied, can affect the character of the information itself, quite apart from any claim against the person who received it.

What is not worth doing

Insisting on an agreement before every conversation is a habit that costs more than it protects. It slows down introductions, it reads as inexperience to the people you most want to impress, and at a stage where nothing of substance is being said it secures nothing at all.

Negotiating a document for weeks is rarely worth it either, when the same protection could be obtained by trimming what is shown. Where a counterparty is immovable on their own form, the honest question is whether the disclosure can be reshaped to fit the protection actually available.

Nor is every breach worth pursuing. Where the information has already lost its commercial value, where the counterparty is outside the jurisdiction and the sums at stake are modest, a claim can consume more than the information was ever worth. That assessment is worth making while the choice is still genuinely open, and it usually takes far less time than the negotiation that preceded the disclosure.

One thing is always worth checking, and costs nothing: that the person signing has authority to bind the organisation that matters. An agreement signed by an individual who cannot commit their employer protects against the wrong party.

This guide is general information about how these matters usually run. It is not advice, and nothing becomes advice until terms are agreed in writing. Brandleys Legal Ltd delivers reserved legal activities alongside regulated partners.