When a corporate decides to build a venture around its own technology, the most consequential document is often not the shareholders' agreement or the business plan. It is the licence. The licence decides what the venture can sell, to whom, where, for how long, at what cost and what happens when the parent is sold, changes strategy or simply loses interest. A generous shareholding with a weak licence is worth very little; a modest shareholding with a strong licence can be worth a great deal.
Corporate IP licensing to a new venture is different from the licensing deals most corporate legal teams are used to. Those deals are written between two established businesses that each expect the other to survive, and they optimise for protecting the licensor's revenue. A venture licence has a different job. It has to let a company with no revenue raise money, hire people and win customers on the strength of rights it does not own outright. Terms that are ordinary between incumbents, such as short terms, broad termination rights or royalty floors from year one, can make a venture unfundable.
This article sets out the main terms of a corporate-to-venture licence, explains which choices help the venture succeed and which protect the parent, covers the Indian points that arise when royalties cross borders or move between related parties, and offers a term sheet guide that both sides can use to negotiate. It complements a separate PanScience Research piece on research commercialisation and university licensing, which this article does not repeat. It is not legal or tax advice.
Why the licence matters more than the cap table
Investors in a corporate-born venture ask one question early: does this company own, or reliably control, the thing it sells? If the core technology stays in the parent, the answer lives entirely in the licence. A licence that the parent can terminate for convenience, that expires before the venture can reach scale, or that lapses on the parent's change of control, gives investors a reason to walk away or to discount heavily.
The licence also shapes the relationship long after signing. If royalties are too high, the venture's margins never reach the level its investors need, and the parent eventually faces a choice between renegotiating and watching the venture fail. If the field of use is drawn too narrowly, the venture cannot pivot into the adjacent market where its real opportunity turns out to be. If improvements default to the parent, the venture's own engineers have little reason to invent.
A further reason the licence matters is that university-style licensing economics are shifting. NYU's entrepreneurship institute, summarising AUTM's 2024 Licensing Activity Survey, reports that income from running royalties dropped by more than 40% and gross licensing income declined by 24%, while revenue from equity exits rose by 25% over the same period. That is a university dataset, not a corporate one, but it points in a direction corporate parents should note: for ventures, the value of a licence is increasingly realised through equity rather than through running royalties.
The core terms of a corporate IP licensing deal
WIPO's technology transfer training material, in a presentation by IP consultant Yumiko Hamano titled "Licensing: Licensor and Licensee Perspectives" (2018), frames licence negotiation as a series of questions. The questions are a useful checklist because they apply as much to a corporate-to-venture licence as to any other.
Field of use. WIPO's material asks "which purpose or technical field?", with examples such as research-only, medical-only or electronics- only use, and notes that a technology may be licensed for one industry but not another. For a venture, the field is the market it is allowed to address. Drawn too wide, the parent fears the venture will compete with its core business. Drawn too narrow, the venture cannot follow demand.
Exclusivity. The same material notes that whether to grant an exclusive licence depends on the licensee's capacity, experience and market share, on territory and patent coverage, on the technology and the licence duration, and on its effect on the royalty rate. It observes that where substantial development is still needed, an exclusive licence may be needed "to justify the investment required". That is precisely the venture's position. Almost every corporate-born venture needs exclusivity within its field; the negotiation is over how wide the field is and what the venture must do to keep it.
Territory. WIPO's checklist asks in which countries the licence applies and whether conditions differ by territory. A licence can be exclusive in some territories and non-exclusive in others. For an Indian venture with global ambitions, a licence limited to India is often a hidden ceiling on valuation.
Financial terms. The WIPO material lists royalties in a range of 3 to 15%, lump sums and minimum royalty guarantees, along with the frequency and timing of payments. For a venture, the structure matters more than the headline rate. A royalty that starts only after first commercial sale, steps up as revenue grows and is capped, or that can be partly converted into equity, preserves cash when the venture has none.
Improvements and grant-backs. WIPO's checklist asks whether the licensee gets rights to improvements made by the licensor, and whether the licensor gets rights to improvements made by the licensee. The second question describes a grant-back. A broad grant-back that assigns all venture improvements to the parent turns the venture into an unpaid R&D department. A narrow one, such as a non-exclusive licence back to the parent for use outside the venture's field, protects the parent without hollowing out the venture.
Sublicensing. The material asks whether the licensee may license the technology to a third party, giving outsourced manufacturing as an example. Ventures need sublicensing rights to work with contract manufacturers, channel partners and, eventually, acquirers.
Three further terms are less often discussed but decide whether a venture licence survives contact with reality.
Term and termination. A venture licence should run at least for the life of the licensed patents or a long fixed term for know-how, and should be terminable by the parent only for material breach or failure to meet agreed diligence milestones, never for convenience.
Change of control, in both directions. The licence should survive a change of control of the parent, so that a group restructuring does not strand the venture. It should also be assignable with the venture on an acquisition, because a licence that cannot travel with the company makes the company unsellable.
Diligence obligations. In exchange for exclusivity, the parent can fairly ask the venture to hit milestones, such as a first product, a first customer or a minimum spend on development, and can narrow or end exclusivity if it does not. This is the parent's legitimate protection against a venture that sits on the technology.
Indian considerations: cross-border royalties and related parties
Two Indian issues arise often enough to deserve a section of their own.
Remitting royalties abroad. Where the parent is outside India and the venture is an Indian company, royalty payments leave the country. Obhan & Associates, writing in Bar & Bench in May 2022, explain that "remittance of royalty payments is considered as current account transactions that are permissible under the automatic route", that earlier caps on royalty remittances were "removed retrospectively from December 2009", and that all payments must still comply with the current account rules. They recommend keeping valuation reports and the executed licence agreement to support the amount paid.
Withholding tax is the second cross-border cost. BDO's summary of the Finance Act, 2023 reports that India's domestic withholding rate on royalties paid to non-residents doubled from 10% to 20% (21.84% after surcharge and cess) with effect from 1 April 2023. Because the domestic rate is now higher than most treaty rates, most non-resident licensors will rely on the applicable tax treaty, which requires a valid tax residency certificate, Form 10F where the certificate lacks prescribed details, an Indian PAN and an Indian tax return. A venture that pays royalties to a foreign parent should budget for this compliance from the first payment, not discover it at the first audit. Section references to the Income-tax Act, 1961 should be mapped to the Income-tax Act, 2025, which applies from 1 April 2026.
Transfer pricing between related parties. Where the parent holds a controlling stake in the venture and the licence is cross-border, the royalty is a transaction between associated enterprises and must be priced at arm's length. Obhan & Associates note that "determining arm's length royalty payment for transfer of intangible property can be a challenging exercise", and that tax authorities often ask taxpayers to show which intangibles are involved, what benefit is received, whether a royalty is already embedded in the price of imports or sales, and who owns the intangibles.
The OECD Transfer Pricing Guidelines add the principle that matters most for ventures. Paragraph 6.50 of the 2022 Guidelines states that "each member of the MNE group should receive arm's length compensation for the functions it performs", and applies that to the development, enhancement, maintenance, protection and exploitation of intangibles. A venture that does most of the development and commercialisation work on a technology should, over time, earn a larger share of the return from it, and a royalty that sends most of that return back to the parent can become hard to defend. The practical implication is that a licence written for a venture's first year may need to be revisited as the venture's own contribution to the technology grows.
Neither issue arises in the same form when both the parent and the venture are Indian and the venture is a minority-owned associate, but the documentation habit is worth adopting anyway. Investors will diligence the licence with the same questions a tax officer would ask.
A term sheet guide for corporate IP licensing
The table below is this article's contribution. For each major term, it sets out the default that tends to help a venture succeed, the default that protects the parent, and a middle position that experienced negotiators often reach. It is an analytical framework built from the WIPO checklist and the Indian considerations above, not a survey of actual Indian deals, and the middle positions are illustrative starting points rather than market standards.
Three patterns run through the table. First, the parent's legitimate interest is protected best by conditionality, such as diligence milestones and field limits, rather than by blunt instruments such as termination for convenience. Second, most of the parent's economic return is better taken as equity than as royalty, because royalty extracts cash when the venture has least and equity rewards the parent when the venture has most. Third, every term should be tested against the investor question in the final column, because a licence that fails that test will be renegotiated at the first funding round anyway, on worse terms for everyone.
How the licence and the equity should be designed together
A common mistake is to negotiate the licence and the shareholding in separate rooms. The two are substitutes. A parent that takes a large equity stake has less reason to insist on high royalties, because it already shares in the venture's success. A parent that takes a small stake, or none, has a stronger case for running royalties, milestone payments or a buy-out price.
A useful discipline is to model the parent's total return from both instruments under three scenarios: the venture fails, the venture reaches modest scale, and the venture succeeds. If the parent's return in the failure case is driven mainly by upfront cash, the venture is being asked to fund the parent. If the parent's return in the success case is capped by a royalty buy-out at a low price, the parent is giving away too much. The right design is one in which the parent and the venture's other investors want the same outcome.
PanScience reports that it originates, co-founds, builds and scales its ventures with shared in-house functions. Whatever the model, the venture-builder's experience points the same way: the parent's long-term interest is in the venture's equity value, and licence terms are best judged by whether they increase it.
Conclusion
Corporate IP licensing to a new venture is a test of whether the parent wants the venture to succeed or merely wants to keep the option of it. A licence that is exclusive in a sensible field, long enough to build on, conditional on diligence rather than terminable for convenience, light on early cash and portable on a sale, gives the venture a real chance and gives the parent a stake worth having. A licence that does none of these things will either be renegotiated by investors or will leave the venture stranded.
For Indian ventures licensing from foreign parents, the deal also has to work through the current account rules for remittances, a 20% domestic withholding rate on royalties since April 2023 unless treaty relief applies, and arm's length pricing that should evolve as the venture does more of the work. None of these is a reason not to license. Each is a reason to write the licence carefully, document it well and revisit it when the venture changes.
Methodology and sources
This article combines published licensing guidance and Indian regulatory commentary with an original term sheet framework. The licence term checklist and royalty range of 3 to 15% come from Yumiko Hamano's presentation "Licensing: Licensor and Licensee Perspectives" for a WIPO technology transfer programme (2018, https://wipo.int/edocs/mdocs/aspac/en/wipo_ttos_kul_18/wipo_ttos_kul_18_p19.pdf). The AUTM 2024 Licensing Activity Survey figures are as summarised by Frank Rimalovski for NYU's entrepreneurship institute (April 2026, https://entrepreneur.nyu.edu/blog/2026/04/03/from-lab-to-vc-the-university-spinout-playbook/); the AUTM report itself was not opened. FEMA treatment of royalty remittances and the transfer pricing questions come from Ashima Obhan and Akanksha Dua, "Cross-border royalty payments: A legal check", Bar & Bench, 18 May 2022 (https://www.barandbench.com/law-firms/view-point/cross-border-royalty- payments-a-legal-check). The withholding rate change comes from BDO's summary (https://www.bdo.global/en-gb/insights/tax/world-wide- tax/india-withholding-tax-rate-on-royalties-and-fees-for-technical-services-doubles). The arm's length principle for intangibles is from paragraph 6.50 of the OECD Transfer Pricing Guidelines 2022, read via TPcases (https://tpcases.com/tpg2022-chapter-vi-paragraph-6-50/). The commencement of the Income-tax Act, 2025 on 1 April 2026 is from LawSikho (https://lawsikho.com/blog/?p=6741). The PanScience operating model is company-reported from panscience.xyz. The term sheet table and the joint licence-and-equity modelling discipline are original and illustrative; the middle positions are not drawn from a dataset of actual transactions.