Startups & Founders

Yes — and before the company matters, not after. A founders' agreement is an enforceable contract under the Contract Act 1872 that records equity split, vesting, roles, IP assignment and what happens when someone leaves, which are exactly the questions that become unanswerable once a dispute exists. Every messy founder breakup we see traces back to terms that were discussed, agreed in spirit, and never written.

The load-bearing clauses: who owns what percentage, vesting and what happens to unvested equity on exit, assignment of each founder's IP to the company, roles and decision-making, confidentiality, and a mechanism for deadlock and departure. Good agreements also cover full-time commitment and side projects, because divided attention is the most common early dispute. It should be consistent with the company's articles once incorporated — contradictions between the two are a diligence red flag.

The law does not prescribe a split — equal, unequal and heavily skewed splits are all valid — so the legal job is making whatever you agree real: shares actually allotted and entered in the register of members, the split recorded in the founders' agreement, and vesting attached so the split stays fair if someone leaves. The commercial advice is to have the uncomfortable conversation early; a split renegotiated after incorporation means share transfers, documents and sometimes tax, not just a handshake.

Vesting means a founder earns full rights to their equity over time — commonly across several years — instead of owning it all outright on day one. Investors insist on it because the alternative is funding a company where a departed co-founder keeps a large passive stake while the remaining team does the work. It protects the founders from each other for the same reason, which is why we recommend it even before any investor asks.

Under the Companies Act 2017 a share, once allotted and registered, belongs to the holder — so vesting is built contractually around that fact, typically as reverse vesting: the founder holds the shares, and the founders' agreement gives the company or continuing founders a right to acquire the unvested portion at a pre-agreed price on departure. Because a company's ability to purchase its own shares is tightly restricted, the transfer usually runs to the other founders or their nominee, as of mid-2026. Mirroring the transfer mechanics in the articles is what makes the promise enforceable in practice.

A cliff is the initial period — conventionally the first year — during which no equity vests at all; if the founder or employee leaves within it, they leave with nothing. At the cliff date the first tranche vests in one step, and vesting then continues periodically. It exists to stop very early departures from carrying equity out of the company, and it is standard in both founder vesting and ESOP grants.

Whatever the documents say — which is why the documents matter. A well-drafted founders' agreement distinguishes good leavers from bad leavers, sends unvested shares back at a nominal or pre-agreed price, and sets the mechanism and valuation for any vested shares the continuing founders may buy. Without those terms, a departed co-founder simply remains a shareholder with full rights, and Pakistani company law gives the remaining founders no general power to remove them.

Yes, formally and in writing. Code, designs, brand assets and inventions created before incorporation belong to the individuals who made them — incorporation does not sweep them in — so each founder should sign an IP assignment transferring the pre-incorporation work to the company. Investors check this chain of title in every serious round, and a missing founder assignment, especially from a founder who has since left, is one of the most expensive gaps to fix late.

Not reliably. For employees, work made in the course of employment generally belongs to the employer under the Copyright Ordinance 1962, but the boundaries of course of employment are exactly where disputes live — so contracts should say it expressly. For freelancers and contractors there is no automatic transfer at all: without a written assignment, the contractor owns what they made and your startup holds, at best, a licence of uncertain scope. Every contributor agreement should carry an IP assignment clause.

It can try, if your employment contract has IP, moonlighting or exclusivity clauses and you built the startup on the employer's time, equipment or confidential information. The risk is fact-driven: what your contract says, what you built, when and with what resources. Before incorporating, read your contract, keep the startup work demonstrably separate, and where the exposure is real, resolve it — a waiver or a clean departure — rather than importing a dormant ownership claim into your cap table.

Register them as trademarks with the Intellectual Property Organization of Pakistan under the Trade Marks Ordinance 2001 — company name registration at SECP and domain ownership do not, by themselves, give trademark rights. File in the classes covering your actual goods and services, in the company's name rather than a founder's. Registration is slow enough that filing early is the practical rule, as of mid-2026 [CURRENT REGISTRY TIMELINES — TO BE VERIFIED BY REVIEWING LAWYER].

Partially. An NDA is an enforceable contract under the Contract Act 1872 and is worth signing with employees, contractors and commercial counterparties who will see genuinely confidential material. Its limits are practical: enforcement means litigation, proof of breach is hard, and an NDA protects information, not the abstract idea — anyone can build a competing product they did not learn from you. Most institutional investors will not sign NDAs to hear a pitch, and asking is generally read as inexperience.

Restraints during a contract's life are generally enforceable; restraints after it ends run into section 27 of the Contract Act 1872, which voids agreements in restraint of trade subject to narrow exceptions, and courts have applied post-employment non-competes restrictively [CASE LAW POSITION — TO BE VERIFIED BY REVIEWING LAWYER]. The practical drafting response, as of mid-2026, is to rely on what does hold: confidentiality obligations, IP assignment, and carefully scoped non-solicitation of staff and customers. A non-compete should never be the only wall around the business.

Yes. A private company builds its ESOP contractually: board-approved scheme rules covering the pool, vesting, leavers and exercise, grant letters to each employee, and authorised capital headroom so the promised shares can actually be issued. The codified public-company ESOP regime does not apply, which gives private companies flexibility and also removes the guardrails — undocumented option promises in offer letters are a recurring diligence problem. Tax treatment for employees should be checked against the Income Tax Ordinance 2001 at design time.

Pool size is a commercial term, not a legal one — it is typically negotiated with investors at each round, and where it sits in the pre-money or post-money math directly affects founder dilution. What the law requires is that the pool be real: scheme rules adopted, authorised capital sufficient to honour every grant, and the pool reflected consistently in the cap table and investment documents. A pool that exists only in a pitch deck converts into a dispute at exactly the moment it matters.

The Pakistan Software Export Board registers IT and IT-enabled services companies and freelancers, and that registration is the gateway to the sector's benefits — including the concessional tax treatment of IT export proceeds, which as of mid-2026 is tied to PSEB registration under the income tax framework [CURRENT RATES AND CONDITIONS — TO BE VERIFIED BY REVIEWING LAWYER]. If your startup exports software or IT services, registering is usually clearly worth it. It is a sector registration, not a substitute for SECP incorporation or tax registration.

Before value accumulates in the wrong place: once there is meaningful IP being written, revenue contracts to sign, people to hire, or a fundraise on the horizon, the company should exist so all of that lands inside it. Incorporating late means assigning back-work IP, novating contracts signed personally, and explaining the gap in diligence. Incorporating absurdly early has costs too — compliance obligations start at the certificate — so the trigger is real activity, not the idea.

A short, real list: founders' agreement with vesting and IP assignment, incorporation documents and thought-through articles, employment and contractor agreements with IP and confidentiality clauses, terms of service and a privacy policy if you have users, and clean records of any money that has come in and any equity promised. That set covers the questions every investor's diligence list asks first. Most of it is cheaper to do once, early, than to reconstruct under a term sheet deadline.

Nothing good, unless the documents planned for it — with equal shareholding and an even board there is no statutory tie-breaker, and the Companies Act 2017 remedies for oppression and mismanagement are blunt, slow instruments for what is really a partnership breakdown. The founders' agreement or shareholders' agreement should build the exits in advance: reserved matters, escalation and mediation, and a buy-sell mechanism for a true impasse. Agreeing those terms is easy at the start and close to impossible mid-dispute.

They do different jobs. The founders' agreement governs the founding team — split, vesting, roles, IP — and is often signed before the company exists; a shareholders' agreement governs everyone who holds shares, and becomes the central document once investors arrive with their own rights and protections. Early on, one well-drafted document can serve both purposes, but by the first priced round expect a proper shareholders' agreement, with the articles updated to match it.