Fine Print, Future Problems
Most startup founders are excellent at building things. Contracts, not so much.
That is not a criticism. It is just the reality of early-stage company life. When you are racing to close a funding round, ship a product, or onboard your first enterprise client, legal paperwork feels like background noise. So agreements get copied from the internet, clauses get pasted from templates someone used three years ago, and critical protections either end up vague or missing entirely.
The trouble is, contracts do not fail during the good times. They fail under pressure: a co-founder exit, a client dispute, a due diligence exercise, or an acquisition that suddenly stalls. And when they do fail, the clause that everyone glossed over at signing becomes the most expensive paragraph in the room.
Here are five clauses startups consistently get wrong, and what to do about them before the pressure arrives.
1. Intellectual Property Assignment
Think of your company’s IP like a property deed. Just because someone built something on your land does not mean they handed over the title. Without a signed, written assignment, the builder may still own what they created.
In India and the United States both, IP ownership must be explicitly assigned in writing. A verbal understanding, a loose “work for hire” reference, or an employment contract that assumes ownership without stating it can leave a significant gap in your chain of title. That gap shows up during investor due diligence, and it rarely ends quietly.
Investors routinely flag missing assignment deeds or improperly drafted clauses. If ownership cannot be clearly established, acquirers may insist on steep indemnities or withdraw from the deal entirely.
Three mistakes come up repeatedly:
Relying only on “work for hire” language without an express assignment clause. Forgetting that contractors, not just employees, need separate assignment agreements. And ignoring pre-incorporation IP, the code, designs, or product architecture built before the company even existed.
Under Indian law, IP assignment agreements need to confirm compliance with Section 17 of the Indian Copyright Act, which governs ownership of works made for hire. Agreements involving transfer of IP rights also require clear documentation and, depending on the jurisdiction, payment of applicable stamp duties.
What a strong IP clause looks like: It expressly assigns all present and future IP rights to the company. It covers founders, employees, consultants, and contractors. It reaches back to pre-incorporation work related to the business. It includes a waiver of moral rights where legally permissible, and it requires cooperation for any future filings or registrations. Most importantly, it survives termination.
Execute these agreements before the work begins, not after the product is already built.
2. Confidentiality and NDA Clauses
An NDA is not a force field. It does not protect everything you consider confidential. Courts will generally only enforce confidentiality obligations over information that is specifically identifiable, commercially valuable, and reasonably safeguarded. Broad, sweeping language that covers “everything we discuss” is often the first thing a court sets aside.
The other common mistake is behavioural, not just drafting. Early-stage founders share product strategies, customer lists, pricing models, and technical architecture with vendors, freelancers, and prospective hires, often with no structured confidentiality in place at all.
In technology and AI startups especially, the most valuable assets are frequently not registered patents. They are confidential datasets, prompts, internal workflows, and proprietary models, all of which can disappear quietly without a properly drafted NDA.
What an effective confidentiality clause looks like: It clearly defines what counts as confidential information. It carves out what is publicly available or independently developed. It specifies permitted disclosures, sets a clear duration, and requires return or deletion of confidential material at the end of the relationship. It includes injunctive relief language for situations where monetary damages are inadequate.
But here is the part founders often underestimate: confidentiality is not just contractual. It is operational. Restricted access controls, internal classification systems, and proper onboarding and exit procedures matter as much as the words in the agreement.
3. Change of Control and Scope Change Clauses
These two clauses look unrelated on the surface. One governs what happens when your company gets acquired. The other governs what happens when a project expands beyond what was originally agreed. But they share the same underlying failure: both describe what happens when the original deal shifts, and most startup contracts handle that shift very badly.
Change of Control
Think of a change of control clause as the contract’s answer to the question: did you sign with this company, or with whoever ends up owning it?
When a startup is acquired, its key vendor agreements, customer contracts, and partnership deals do not automatically transfer cleanly to the acquirer. Many contracts include change of control provisions that give the other party the right to terminate, renegotiate, or withhold consent when ownership shifts. That right, buried in a boilerplate clause your team signed two years ago, can surface mid-acquisition and stall or derail the entire deal.
A change of control clause generally states that if a party undergoes a change in managerial control, the other party has the right to terminate the contract. This is sometimes called the “poison pill” in acquisition contexts, because if the IP contracts or key agreements do not survive the transaction, they hold no value for the acquirer.
Startups on both sides of a deal get this wrong. As a vendor or service provider, you may not realise that your customer’s acquisition triggers a consent requirement under your agreement. As a startup being acquired, you may not have mapped which of your contracts contain these clauses until due diligence is already underway.
A well-drafted change of control clause sets out specific triggers (mergers, acquisitions, transfers of substantial assets), and specifies the rights of each party, including the ability to terminate, accelerate payment obligations, or impose restrictions on the transferee.
Scope Creep and Change Order Clauses
Scope creep is the slow, polite way a fixed-price contract becomes an open-ended one. The client asks for one small addition. Then another. Then a restructure of what was already built. Nobody signs anything. And by the time the dispute surfaces, both sides have a completely different memory of what was agreed.
The Project Management Institute reports that 52% of all projects experience some degree of scope creep, typically caused by a poorly defined initial scope, lax contract management, or a lack of agreement on what the contract was meant to accomplish. In tech services and product development agreements, the absence of a proper change control mechanism is one of the most reliable paths to a billing dispute.
A scope freeze clause locks requirements at a certain point, after which any modifications, including additions, deletions, or changes to features or functionality, require a written change order signed by both parties, specifying the scope change, cost impact, and timeline adjustment.
What strong clauses on both look like: For change of control, the clause should define the trigger events clearly (majority ownership transfer, merger, asset sale), specify notice obligations, and state whether consent is required or whether termination is automatic. For scope changes, the agreement needs a detailed Statement of Work with explicit deliverables and exclusions, a written change order requirement before any additional work begins, and a no-oral-modification provision that makes clear that email requests and verbal approvals do not constitute binding scope changes.
Courts interpret ambiguity against the drafter. If your SOW is vague, your client’s version of what was included carries significant weight.
4. Limitation of Liability
This clause is where startups often copy from enterprise contracts without understanding why those contracts were structured that way, or what they were protecting against.
The result tends toward one of two extremes: either the startup accepts unlimited liability exposure, or it inserts such an aggressively capped limitation clause that no serious enterprise client will sign it.
Think of it like insurance coverage. Too little and a single claim can be catastrophic. Too much exclusion and the other party walks. The goal is a commercially realistic middle ground that protects the business without making it commercially uninvestable as a counterparty.
For SaaS and technology startups, the stakes are real. A platform outage causes downstream losses. Customer data is compromised. An AI output triggers third-party claims. A vendor integration fails. Without properly drafted limitation language, any one of these can become existential for an early-stage company.
What a balanced limitation clause looks like: It caps liability to a commercially reasonable amount, typically tied to fees paid or a fixed figure agreed between the parties. It clearly defines excluded liabilities, the ones that should never be capped, such as confidentiality breaches, IP infringement, fraud, wilful misconduct, and data protection violations. It disclaims indirect and consequential damages. And it aligns indemnity obligations with the same liability limits.
For startups negotiating enterprise contracts, this clause often becomes the real commercial negotiation, even more than pricing. Founders should also keep in mind that courts may scrutinise unreasonable exclusions, particularly in unequal bargaining relationships.
5. Termination and Exit Clauses
Most contracts do a reasonable job of describing how a business relationship begins. They do a much worse job of describing how it ends.
Termination clauses are often vague, inconsistent, or silent on practical consequences. Notice periods are unclear. Post-termination obligations are missing. There is no transition assistance, no return of assets, and no guidance on what happens to data, work product, access credentials, or continuing confidentiality obligations.
This matters because every business relationship eventually ends. The real test of a contract is not how it operates when everything is going well. It is how it functions during separation.
For startups dependent on external developers, agencies, or cloud vendors, weak exit terms can create operational paralysis. Imagine losing access to your own codebase because the contractor’s agreement was vague about ownership of deliverables, or discovering that your vendor has no obligation to assist with migration when you want to switch platforms.
What a strong termination clause looks like: It distinguishes clearly between termination for convenience and termination for breach. It includes cure periods. It specifies transition obligations, return of company assets, and data portability. It addresses what gets deleted and what payments remain due. It explicitly lists which clauses survive termination, because some always should: confidentiality, IP assignment, limitation of liability, and dispute resolution among them.
Who retains access after termination? Who owns the work product? What must be destroyed? What is still owed? The answers to those questions matter far more than founders initially realise.
The Real Problem Is Timing, Not Drafting
Most startups do not ignore contracts because they are careless. They ignore them because legal work feels non-urgent against the pressure of product launches, fundraising, and hiring. Contracts sit in the “important but not urgent” quadrant right until they are both.
The problem is that contract problems compound quietly. A missing IP assignment today becomes a due diligence crisis two years later. A vague founder agreement becomes litigation after a stressful funding round. An uncapped liability clause becomes catastrophic after the first enterprise dispute.
Good startup contracts are not about sounding sophisticated. They are about reducing ambiguity before pressure tests the relationship.
The strongest startup agreements share three characteristics. They are commercially realistic. They are operationally practical. And they are drafted before conflict begins.
That timing makes all the difference.
Footnotes and References
[1] Section 17, Indian Copyright Act, 1957: governs ownership of works made “for hire” in the course of an author’s employment. See also Section 19, which requires written assignment of copyright.
[2] Lexology / Maheshwari & Co., IP Due Diligence in M&A Transactions (2025): flags missing assignment deeds as among the most common deal-breakers in Indian startup acquisitions. Available at maheshwariandco.com.
[3] iPleaders, Change of Control Clauses vs Assignment Clauses in Technology Contracts: covers how change of control clauses function as “poison pills” in acquisition contexts, and the distinction between assignment and change of control triggers. Available at blog.ipleaders.in.
[4] Lexology, Navigating Change of Control Clauses in IT Contracts (2024): outlines typical rights and obligations triggered on a change of control event, including termination rights, payment acceleration, and transferee restrictions. Available at lexology.com.
[5] Project Management Institute (PMI): reports that 52% of all projects experience some degree of scope creep, typically caused by a poorly defined initial scope or lack of agreement on contract deliverables.
[6] LegalClarity, What Is Scope Creep and When Does It Breach a Contract? (2026): covers no-oral-modification clauses, scope freeze provisions, and change order best practices for private-sector tech contracts. Available at legalclarity.org.