Data Licensing Agreement: Template and How to Draft One (2026)

What a data licensing agreement is A data licensing agreement is a contract where one party, the licensor, grants another, the licensee, the right to use a dataset on agreed terms. It defines what data is covered, how the licensee may use it, whether anyone else can use it at the same time, how long the right lasts, what it costs, and what happens when it ends. Because data is easy to copy and hard to claw back, these agreements carry risks that ordinary supply contracts do not. Datasets get licensed for many reasons: market and financial data sold to firms, research data shared between institutions, location or demographic data bought by marketers, and increasingly data licensed to train machine learning models. Two questions dominate every one of these deals. First, is there personal data in the set, because if there is, data protection law applies on top of the contract. Second, what may the licensee create from the data, because derivative data, insights and trained models can outlive the licence and are the hardest points to unwind. This is general information, not legal advice. Data licensing sits squarely on data protection and intellectual property law, which differ sharply by country, so have an important data licensing agreement reviewed by a qualified professional. ## A data licensing agreement template you can copy ```

DATA LICENSING AGREEMENT Date: [date] Licensor: [legal name and address] Licensee: [legal name and address] The Data: [describe the dataset, its fields, format, volume and source] 1. LICENCE The Licensor grants the Licensee a [non-exclusive / exclusive], [non-transferable] licence to use the Data for [permitted purpose], in the territory of [territory], during the term. 2. PERMITTED AND PROHIBITED USE The Licensee may use the Data only for the permitted purpose. The Licensee may not: resell or sublicense the Data except [as stated]; combine it with other data to re-identify individuals; or use it beyond the permitted purpose. [State whether creating derivative data and training machine learning models are permitted, and on what terms.] 3. DATA PROTECTION The parties will comply with applicable data protection law. If the Data contains personal data, the parties' roles are: [controller / processor / joint], and the data protection terms in Schedule 1 (or a separate data processing agreement) apply. The Licensor confirms it has a lawful basis to license the Data for the permitted purpose. 4. DATA QUALITY AND SOURCE The Licensor warrants it has the right to license the Data and that the Data was collected lawfully. The Data is provided [as is / to the quality standard in Schedule 2]. The Licensor [does / does not] warrant accuracy or completeness. 5. SECURITY The Licensee will protect the Data with [stated security measures] and notify the Licensor of any data breach affecting the Data without undue delay. 6. DERIVATIVE DATA AND OWNERSHIP The Licensor retains ownership of the Data. Ownership of derivative data, insights and any trained models is: [assign clearly]. Any licence back to the Licensor is: [state]. 7. FEES The Licensee pays [fee / usage-based fee] per Schedule 3. [Payment terms]. 8. TERM AND END OF TERM This agreement runs for [period]. On termination the Licensee will, at the Licensor's choice, delete or return the Data and certify deletion, subject to backups and legal retention. The treatment of derivative data and trained models on termination is: [state]. 9. WARRANTIES, INDEMNITY AND LIABILITY The Licensor indemnifies the Licensee against third party claims that the licensed Data infringes their rights or was collected unlawfully. Each party's liability is capped at [amount], except for data protection breach, confidentiality breach and IP infringement. 10. GOVERNING LAW This agreement is governed by the law of [place]; disputes are resolved in [place]. SIGNED Licensor ____________ Date ______ Licensee ____________ Date ______

- No source warranty. If the licensor will not warrant it had the right to collect and license the data, the licensee inherits that risk.
- Open ended derivative rights. Vague wording on derivatives and model training lets value leak out of the licensor's control forever.
- No deletion on termination. Without it, licensed data stays in the licensee's systems indefinitely.
- Weak security and breach terms. Licensed data in a system with no security standard or breach notice is a liability for both sides. ## How jurisdiction changes things Data protection law shapes these deals more than any other. The EU and UK GDPR set strict rules on lawful basis, roles, security and international transfers, with heavy penalties. In the United States, sectoral and state laws such as the California privacy regime apply, and there is no single federal statute. In India, the Digital Personal Data Protection Act, 2023 governs personal data, with rules commencing in phases. <!-- [VERIFY: DPDP Act 2023 rules and phased commencement; confirm current status] --> Intellectual property in data is uneven: facts are generally not protected by copyright, but a structured database may attract database or copyright protection in some regions, such as the EU database right. Confirm both the data protection and the IP position for every territory the data touches before relying on the licence. ## Keeping data agreements and their safeguards together A data licence rarely stands alone. It comes with a data processing agreement, a security schedule, a quality standard, and breach and deletion records that build up over the term. The value of the record is in keeping them linked and current, so the purpose allowed, the roles agreed, and the safeguards promised all match what both sides signed. That is the same discipline any organisation needs across its contracts. Drafting from a clear template, catching a clause where data protection or derivative rights are missing or one-sided, editing wording in plain language, and holding each version with its date can live in one AI workspace tuned to how your company handles data deals. [Adira](https://adiralaw.com) is built around that single workspace idea. For a single data licensing agreement the template above is a strong start; the point worth borrowing is that data deals stay safe only when the licence and its safeguards live in one place and in step.