A fractional CTO contract should cover scope and deliverables, hours and availability, fees, IP assignment to your company, confidentiality, conflicts with other clients, non-solicitation, company ownership of every account, security duties, liability, notice period, handover obligations and, if offered, equity with vesting. Get those right and the relationship is easy to run and easy to end.
The sections below take each clause in turn. Contract law, employment rules and tax treatment vary by country, so treat this as a map for the conversation with your lawyer rather than wording to copy. This is general information, not legal or tax advice: have a qualified lawyer review your agreement before anyone signs.
How a fractional CTO contract is usually structured
Most engagements use two documents. A consulting or services agreement holds the legal terms that rarely change: IP, confidentiality, liability and termination. A statement of work (sometimes called a schedule) holds the commercial details: scope, hours, fees and deliverables. When the engagement shifts, say from building to advising after launch, you update the statement of work without reopening everything else.
Three structural points are worth settling early:
- Who you are contracting with. If you hire through a firm, the contract is with the firm. Add a named-person clause so the CTO you interviewed is the one who does the work, and say what happens if they leave.
- Contractor, not employee. A fractional CTO is normally an independent contractor. In many countries, how the relationship works in practice (fixed hours, exclusivity, your equipment, your control) can matter more than the label, so check the local rules.
- Discovery first. If you start with a paid discovery, give it a short agreement of its own, with the IP and confidentiality terms already in place.
The contract usually comes at the end of the hiring process, after references and a discovery. Its job is to turn everything you learned into terms you can both be held to.
Rule of thumb: if the contract is silent on something, expect to argue about it on the worst possible day.
Scope, time and fees: the commercial clauses
Scope and deliverables
Describe the role and the first results. The role is the ongoing responsibilities: roadmap, architecture, vendor oversight, hiring, and hands-on development if that is part of the deal. The results are three to five deliverables for the first 90 days, written so either side could check them. Also list what is out of scope, such as round-the-clock on-call cover or design work, and how changes get approved: in writing, with any fee change attached.
Time and availability
Fix the commitment in days or hours per week or month, the overlap hours you can rely on and the response times for normal messages and for production incidents. Agree on how time off is covered and whether unused hours roll over. Many retainers don't roll them over, or cap the rollover; either is fine as long as it is written down.
Fees and payment
State the model (monthly retainer, day rate or hourly), when invoices go out, the payment terms and how rates are reviewed. Retainers are commonly invoiced monthly in advance. Expenses should need your approval first, and cloud bills and software subscriptions should be paid by the company on company accounts, never passed through the CTO. If you work across borders, name the currency and who pays transfer fees. For what the numbers usually look like, see what fractional CTO retainers usually cost.
IP, confidentiality and conflicts: the protective clauses
IP assignment
This is the clause that matters most. In many countries, a contractor can own what they create unless a written agreement transfers it, so the contract should assign to the company all rights in the work created for it: code, designs, documents, architecture and data models. Assignment, not just a license. Some contractors ask for ownership to pass as each invoice is paid, which is a common compromise.
The CTO will usually keep their pre-existing tools, templates and code libraries. That is fair, as long as the company gets a permanent, royalty-free license to anything of theirs that ends up in your product, and they flag any open-source code with restrictive license terms. There is more on this in who owns the code when you outsource.
Confidentiality
Make it mutual, cover business plans, customer data and code, and have it survive the end of the contract. If the CTO will handle personal data, data protection law may require extra terms on how it is processed and stored, depending on where you and your users are.
Conflicts with other clients
A fractional CTO works for several companies, so the contract should deal with that openly. A sensible clause says they won't work for a direct competitor during the engagement without your written consent, will tell you if a conflict arises, and will disclose any payment, commission or stake they hold in a vendor they recommend.
Non-solicit
A non-solicitation clause stops either side from poaching the other's staff, and stops the CTO from pursuing your customers, for a period after the engagement ends; 6 to 12 months is a typical range. Broad non-competes are a different matter. They are restricted or unenforceable for contractors in many places, and they don't fit a model built on serving several companies. A narrow conflicts clause plus confidentiality usually does the job better.
Accounts, credentials and security duties
Spell out that every account belongs to the company, is paid for by the company and has the founder or another officer as owner. The CTO gets their own named login with the access the work needs. No shared passwords, and nothing registered in the CTO's name or their company's name, however convenient that is on day one.
| Asset | Registered to | CTO's access |
|---|---|---|
| Code repository | The company | Own login with admin rights |
| Cloud hosting | The company, billed to a company card | Own login with admin rights |
| Domain and DNS | The company | Delegated access only |
| App store developer accounts | The company | Team member role |
| Email, payments and analytics | The company | Individual user accounts |
| Passwords and secrets | A company password manager | Access to shared vaults |
Then add security duties that fit the role: two-factor authentication on every account, access limited to what each person needs, no production data on personal devices unless agreed, and notice of any suspected breach within a set number of hours. If the CTO runs your infrastructure, list what they will set up, such as backups, monitoring and regular access reviews. A good CTO does these things anyway; writing them down makes them visible.
Liability, term and notice
Liability and insurance
Contractors commonly cap their liability, often at the fees paid over a set period such as the previous 6 or 12 months. That is normal; you won't get unlimited liability from an individual. What you can reasonably ask is that the cap does not apply to breaches of confidentiality, IP infringement, data protection failures or deliberate wrongdoing, and that the CTO carries professional liability insurance if the role justifies it. The contract should also make the CTO responsible for their own taxes.
Term and notice
Fractional engagements often run month to month, sometimes after an initial period of around three months during a build. A notice period of around 30 days on both sides is common, with immediate termination for serious breaches. You should be able to reduce hours with notice, not just end the engagement. Be wary of long lock-ins and fees for leaving early. If you and the CTO are in different countries, name the governing law and where disputes would be heard.
Handover obligations: write the exit in on day one
Nobody wants to discuss the ending at the start, which is exactly why it belongs in the contract. When either side gives notice, the CTO should:
- Finish or document all work in progress, with its status.
- Hand over every credential and transfer anything still registered to them.
- Bring the documentation up to date: architecture overview, decision log, deployment steps and an inventory of vendors and accounts.
- Run knowledge-transfer sessions with whoever takes over, for an agreed number of hours.
- Delete company data from their own devices and confirm it in writing.
- Stay reachable for questions for a short period afterward, such as 30 days, at the normal rate.
If the handover is to a full-time hire, the outgoing CTO often helps write the role and interview candidates too. That transition has its own playbook: moving from a fractional to a full-time CTO.
Equity and vesting in a fractional CTO contract
If equity is part of the deal, the services agreement should only refer to a separate grant. That grant should be properly approved by the company and set out the amount, the vesting schedule, any cliff, what happens to unvested equity when the engagement ends and whether anything accelerates on a sale. Never promise equity in an email or a chat message; it creates an expectation without any of the protections. Whether to offer equity at all, and how much, is a separate decision covered in the guide to fractional CTO equity.
The bottom line
A good fractional CTO contract does two jobs: it makes the work measurable and it makes leaving painless. If you only have the energy to get three clauses right, make them IP assignment, account ownership and handover, because those decide whether you still control your product on the day the relationship ends. A candidate who pushes back hard on any of those three is showing you one of the classic fractional CTO red flags.
Frequently asked questions
Should a fractional CTO sign an NDA before the first call?
Usually not for an introductory call. Share the idea at a high level, the way you would with an investor. Once the CTO needs access to code, customer data or detailed financials, typically at the paid discovery stage, put confidentiality terms in place, either as a short NDA or inside the discovery agreement. An experienced fractional CTO will expect this and should not object.
Can a fractional CTO contract include a non-compete?
It can, but broad non-competes are often restricted or unenforceable for contractors, depending on the country, and they clash with a model built on working for several companies. A narrower clause usually works better: no work for a direct competitor during the engagement without consent, plus strong confidentiality and non-solicitation terms. Ask a local lawyer what is enforceable where you and the CTO are based.
Who usually drafts the fractional CTO contract?
Either side can. Many fractional CTOs have a standard agreement they use with every client, which saves time and legal fees. Treat it as a starting point, not a final draft: check that IP is assigned rather than licensed, that every account stays in the company's name, and that notice and handover terms are clear. Having your own lawyer review it is money well spent.
Do I need a separate contract for a paid discovery?
It is common to sign a short discovery agreement first, or a statement of work under the main agreement. It should fix the fee, the length and the deliverables, and confirm that the company owns everything the discovery produces even if you decide not to continue. It also needs confidentiality terms, since the CTO will see your code, contracts and numbers during the work.