Diligence on a game studio does not look like diligence on a SaaS company. The revenue has not started, the asset is a build rather than a contract book, and the two documents that decide what the investor is actually buying are usually the publishing agreement and the intellectual property file. Around those sit the structural questions a Turkish studio has to answer before the money moves: whether to raise into the Turkish company or a foreign holding, how the game gets into the company’s balance sheet in the first place, where investor protections can live given the Commercial Code’s limits on what the articles of association may say, and how much of an option pool Turkish law will actually let you promise. This article works through those in the order a term sheet forces them, and assumes the reader has already dealt with the chain of title, which we covered separately in the first round and the chain of title problem.
Which company the investor buys into
Most Turkish studios incorporate as a limited şirket because it is cheaper and faster, and most institutional rounds end up converting to an anonim şirket. The reason is mechanical rather than aesthetic. Share transfers in a limited şirket require a general assembly resolution and notarised transfer documents; in an anonim şirket, registered shares move by endorsement and delivery with entry in the share ledger, and bearer or registered share certificates can be issued. A cap table that has to survive several rounds, an option plan and a secondary or two is far easier to operate in the second form. Conversion is available under the Commercial Code’s change-of-type provisions and is usually done before the round rather than during it, because an investor pricing a round does not want its closing conditioned on a registry process.
The choice matters less than the timing. Converting mid-round is what causes the delay, not converting as such. A studio that expects to raise within a year should deal with the form early, together with the share ledger, so that the diligence answer is a document rather than a plan.
Getting the game into the company
There are two honest routes, and studios often use neither. The first is contribution as capital in kind. Article 342 of the Turkish Commercial Code No. 6102 provides that assets which are free of any limited right in rem, attachment or injunction, which can be valued in money and transferred, including intellectual property rights and virtual environments, may be contributed as capital in kind, while services, personal labour, commercial reputation and receivables not yet due may not. That sentence was drafted with software in mind and it names intellectual property expressly. Article 343 then requires that contributed assets be valued by experts appointed by the commercial court of first instance at the company’s registered office, with the report explaining why the chosen valuation method was the fairest and most appropriate for the case.
Read the two together and the practical consequence appears. A game whose ownership is contested cannot be contributed as capital in kind at all, because Article 342 asks for an asset free of injunction and the dispute is the injunction risk. The expert valuation is also a public document that an investor will read. Studios that would rather not have their prototype valued by a court-appointed expert take the second route: the founders assign the economic rights to the company for a stated consideration, in writing, with the rights listed separately as Law No. 5846 requires, and the company records the asset accordingly. That is faster and less exposed, but it only works if the founders had the rights to assign, which returns to the chain of title.
The publishing agreement is a diligence document
A signed publishing deal reads like validation. In diligence it reads like an encumbrance, and the investor’s lawyer will go through it before they go through the cap table, because it defines what the company still owns. Four clauses do most of the work. The first is the grant: whether the studio has licensed the game for a term and territory or assigned the rights outright, and whether the licence covers sequels, ports, derivative works and merchandising. A studio that assigned rather than licensed has sold the thing the investor thought it was buying.
The second is recoupment. Advances recouped against the studio’s revenue share, with marketing spend also recoupable, can mean the studio sees nothing until a threshold that the financial model quietly assumed away. The third is term and reversion: what happens at the end, whether rights come back automatically or on notice, and whether the publisher keeps a tail. The fourth is change of control. Many publishing agreements let the publisher terminate, or require consent, on a change of control of the studio, and a funding round with a preferred share issue and a board seat can trip that definition. An investor who discovers a consent requirement after signing has a closing condition they did not price. The fix is prosaic: read the clause before the term sheet, and if consent is needed, obtain the waiver in parallel rather than after.
Where investor protections can actually live
Founders often assume that liquidation preference, veto rights, drag-along and anti-dilution simply get written into the articles of association. Article 340 of the Commercial Code says otherwise: the articles may deviate from the Law’s provisions on joint-stock companies only where the Law expressly permits, and supplementary provisions that other laws allow take effect only as attributed to that law. That is a real constraint. Some protections have a statutory home and belong in the articles, notably privileges attached to share groups, including privileges as to dividend, liquidation proceeds and nomination to the board. Others, above all contractual promises among shareholders such as drag-along, tag-along, transfer restrictions beyond what the Law allows, and information undertakings, live in a shareholders’ agreement.
The distinction is not cosmetic, because it decides the remedy. A right in the articles binds the company and can invalidate a resolution taken against it. A right in a shareholders’ agreement binds only the parties and sounds in damages, which is why well-drafted agreements pair the promise with a security mechanism: a call option, a pledge over shares, a penalty clause, or a board composition that makes breach impractical. When a term sheet lists fifteen investor rights, the first legal question is which of them the articles can carry, the second is what happens to the rest, and the third is whether the company’s organs can actually perform them.
What an option pool can and cannot promise
This is where Turkish law disappoints teams used to Delaware documents. The statutory route for an employee option plan is the conditional capital increase. Article 463 allows the general assembly to resolve on a conditional increase granting, to those holding newly issued bonds or similar debt instruments and to employees, the right to acquire new shares by exercising conversion or purchase rights set out in the articles, and the capital increases automatically at the moment and to the extent the right is exercised and the subscription debt is settled. Article 464 sets two limits: the total nominal value of conditionally increased capital may not exceed half of the capital, and the payment made must be at least equal to nominal value. Article 465 lists what the articles must contain, including the groups entitled to benefit and the fact and extent of the disapplication of existing shareholders’ pre-emptive rights.
Two of those constraints bite in practice. The instrument speaks of employees, which does not obviously carry advisors, independent contractors or a studio’s frequent collaborators, so a plan that promises options to a part-time composer or an external art house needs a different construction. And because payment must be at least nominal value, a zero-strike grant is not available; the exercise price has a floor. The alternative route, having the company hold its own shares for the plan, runs into Article 379, under which a company may not acquire its own shares for consideration beyond one tenth of its principal or issued capital, with the general assembly authorising the board for a period of up to five years and setting the nominal amounts and the price range. One tenth is a ceiling on the mechanism, not a target for the pool, and it is the reason many Turkish plans end up as phantom or contractual bonus schemes rather than real equity. We set out the mechanics of moving a plan onto the statutory footing separately; what matters at term-sheet stage is that the pool percentage in the investor’s model has to be deliverable under one of these routes, and the answer should be established before it is agreed.
Whether to flip before the round
Some investors will ask for a foreign holding company as a condition. That is a separate decision with its own tax and operational consequences, and we treat it on its own in the five key points before moving a Turkish startup under a US holdco, with the aftermath in running the Turkish subsidiary afterwards. Two points belong here. The first is sequencing: a flip executed while the game’s ownership is unresolved exports the problem rather than solving it, because the foreign parent acquires exactly what the Turkish company had. The second is that the incentives a studio may be using, including technopark and research-and-development regimes, attach to the Turkish entity and its activity, so the structure has to keep the development work where the incentive lives or accept the loss knowingly rather than discover it at the first corporate tax return.
What to have ready before the data room opens
The list is short and each item answers a question the investor will otherwise ask twice. A share ledger that matches the trade registry and the cap table in the model. A chain of title file: employment and contractor assignments, written and with rights listed separately, plus confirmatory assignments made after the works existed rather than only a founding-day promise. Licences for the engine, middleware and every third-party asset, in the company’s name. The publishing agreement with the grant, recoupment, term and change-of-control clauses marked, and any consent already sought. A note on the option pool setting out which statutory route will carry it. And a disclosure letter that states the imperfections plainly, because the cost of a known problem is a negotiated term, while the cost of a discovered one is the round.
