Licensing is the moment the abstract question of franchise architecture becomes concrete and expensive. Someone else wants to make something in your world, and everything you never wrote down is about to become a series of phone calls.
It is also the point at which a world's documentation stops being an internal convenience and becomes the product. A licensee is not buying your enthusiasm for the world. They are buying a specification precise enough to build from and permissive enough to leave them room, and the gap between those two is where most licensing relationships go wrong.
This guide covers what to prepare before you license anything, what a licensee genuinely needs, how approval should work, and what should never leave your hands. It sits under how to build a story franchise, which covers the canon architecture this depends on.
You are selling a scope, not a world
The instinct is to think of licensing as granting access to the world. It is more useful, and considerably safer, to think of it as carving out a specific piece and defining its edges. Which properties, which medium, which era, which characters, for how long, in which territories.
Scope is where the value and the risk both live. A license that grants a licensee the right to make a game set in your world is enormously broader than one granting the right to make a game set in one named city, in a stated decade, without dramatizing your central characters. The second is more work to write and far easier to live with.
The most useful test when drafting scope is to ask what the licensee could do that would make you unhappy while remaining entirely within the agreement. Whatever comes to mind is the thing to write a constraint about — and the exercise is worth doing before the negotiation, when you can still think clearly about it.
What a licensee actually needs from you
A canon record they can work from unaccompanied. This is the deliverable people forget is a deliverable. If your world's rules live in your head, you have licensed your own attention as surely as if it were written into the contract — and you will feel it within a month.
An explicit fixed-versus-open statement. A licensee is trying to make something good without breaking anything, and without that line they will do one of two things: build timidly, adding nothing, or invent freely and hit a wall in review. Both are avoidable with one page written in advance.
A named contact with a response commitment. Not a person who exists on an org chart — a person who answers, in a stated number of days. The most common cause of licensing friction is not disagreement; it is latency, which quietly teaches a licensee to make decisions without you.
A defined approval process with defined moments. Which decisions need sign-off, and at which points the work is reviewed. This protects the licensee as much as you: a creator who knows exactly when their work will be checked can plan around it, while one who might be overruled at any moment will hedge everything.
The requirements for a record that survives being handed off are set out in full in the story bible as an IP asset.
Approval without becoming a bottleneck
The two failure modes are opposite and equally common. Approve everything and the license moves at the speed of your inbox. Approve nothing and you find out what was made when it ships.
The workable middle is to approve categories in advance rather than instances on demand. Decide, once, that new minor characters in open regions need no approval, that anything touching a named character or a world rule does, and that dating events relative to existing canon always does. That single page removes the large majority of individual requests, which are mostly people seeking permission for things you were never going to refuse.
Then schedule the reviews rather than reacting to submissions. Two or three defined checkpoints in a project's life catch structural problems while they are still cheap, which continuous oversight does not do any better and at much higher cost to everyone's patience.
Write down what you approve and why. A licensing relationship that runs for years will outlast several of the people involved, and the reasoning behind a decision is the part that lets a successor apply it consistently to a situation nobody anticipated.
What should never leave your hands
Ownership of the world itself, obviously, but the more consequential items are subtler and easier to give away without noticing.
The right to say what is canon. A licensee may make something excellent; whether it becomes part of the world's official record is a separate decision and should be yours explicitly. Plenty of healthy franchises run licensed work outside canon, and saying so up front prevents the most painful version of this argument.
The unreleased parts of the world. A licensee needs what is relevant to their scope, not everything you have. Handing over the whole bible because it is easier than excerpting it is how unpublished plans end up in someone else's development notes. Keeping the unreleased material sealed while sharing the relevant part is a practical problem with a practical solution — CanonBoard's Locked Vault, on Core and above, exists for exactly this separation.
Final say over changes to fixed canon. If a licensee can alter load-bearing facts, every other licensee's work is downstream of a decision you did not make.
The audience relationship, where one exists. It is yours, it is the most durable asset in the whole arrangement, and it does not transfer with a license.
Before you sign anything
Get the canon into a form a stranger can use. Not complete — usable. Structured, searchable, with the fixed and open parts marked. If you do one thing on this list, do this one, because every other problem in licensing gets worse in proportion to how badly it is done.
Decide your fixed list and keep it short. Everything on it is a constraint on a person you are paying, or who is paying you, to be creative.
Name the approver and the response time, and mean both.
Write the scope narrowly, then widen it deliberately where you actually want to. Widening later is a pleasant conversation; narrowing later is not a conversation you get to have.
And check the licensed work against your canon at the checkpoints you defined, with a method rather than a memory. At the point where several properties exist at once, nobody involved has read everything, and contradictions arrive in good faith — the operational answer to that is in franchise continuity management.
Frequently asked questions
- What does it mean to license a story world?
- Licensing grants another party the right to make something specific in your world, for a defined period, under defined constraints — while ownership of the world stays with you. The word covers arrangements of wildly different sizes, from a game studio adapting your setting to a single writer producing one tie-in novel. What they have in common is that you are not selling the world; you are renting a scoped piece of it, and the scope is the entire negotiation.
- What does a licensee need from you?
- Four things, and only the first is legal. The permission itself, scoped precisely. A canon record they can work from without asking you questions daily. An explicit statement of what is fixed and what they may invent. And a named person who answers approval requests within a stated time. The fourth is the one that most often fails: a licensing deal where approvals take three weeks will produce either a delayed project or a licensee who stops asking, and the second outcome is considerably worse than the first.
- Should you license before your canon is documented?
- No, and this is the most common expensive mistake in early franchise work. If your canon lives in your head, a licensing deal converts it into a stream of interruptions — every question routed to you, every ambiguity resolved in a phone call you will not remember. The documentation is not paperwork you do after signing; it is the thing being licensed. A world without a usable record can still be licensed, but what you are actually selling is your own ongoing availability.
- How do you keep a licensee from breaking your canon?
- By making the constraints legible rather than by policing the output. Mark what is fixed, state what is open, and define which decisions require approval — then check the work at defined points rather than continuously. Continuous oversight is expensive for both parties and tends to produce resentment plus a licensee who routes around you. A clear line and two scheduled checks catch more than an anxious presence in every meeting.
CanonBoard scans your whole world and tells you where it disagrees with itself.
Start free