Game Studio Due Diligence: What Buyers Check After Krafton in 2026
Co-founder at Peony. Former M&A at Nomura, early-stage VC at Backed VC, and growth-equity / secondaries investor at Target Global. I write about investors, fundraising, and deal advisors from the deal-side perspective I spent years in.
TL;DR: Game studio due diligence comes down to four questions, which we call Own, Keep, Prove, Stay: do you own the game, can you keep selling it, do the numbers hold, and will the team stay? The market is active: Drake Star counts 147 gaming M&A deals in the first three quarters of 2026 (Drake Star, Oct 2026), and Aream reports 18 control deals for developers and publishers above $100M this year, nearly as many as 2023 to 2025 combined (Aream & Co., Oct 2026). EA's $55B take-private closed on August 4, 2026 (EA, Aug 2026), at 7.4x sales and 21.2x EBITDA. Most studios won't see those numbers: one adviser puts big hits at 4x to 5x revenue and work-for-hire studios at 1.5x to 2.5x, with recent deals taking 10 months to a year (Naavik, Jun 2026). The biggest legal lesson of 2026 is Krafton: Delaware's Court of Chancery reinstated the CEO of Unknown Worlds and extended a $250M earn-out window by 258 days (Mayer Brown, Apr 2026), before the case settled and the CEO left. If you're a founder selling a studio, build the data room before the first buyer call. Peony's Data Room plan ($52/admin/month, billed yearly) gives you per-bidder permissions, watermarks and page-level analytics.
Last updated: November 2026
I'm Sean Yu, co-founder of Peony. Before that, I worked in M&A at Nomura and in venture and growth investing, and I run Peony, a data room company used by 8,000+ teams. This guide is for founders selling a game studio and for the corporate development and private equity teams buying one. It isn't legal advice.
Game studio due diligence is the buyer's review of a studio's intellectual property, contracts, revenue, compliance and team before an acquisition. It works like any tech deal until you reach the details: a game is thousands of creative assets, made by people who often weren't employees, sold through a few platforms that set the rules.
What are buyers really checking in a game studio sale?
Buyers are checking four things, and every request list maps to one of them. We call this the Own, Keep, Prove, Stay framework.
| Question | What the buyer wants to know | Where deals get stuck |
|---|---|---|
| Own | Does the studio hold the rights to the game? | Missing contractor assignments, engine terms, AI assets |
| Keep | Can it keep selling after the sale closes? | Change-of-control clauses, platform terms, loot boxes |
| Prove | Are the revenue and retention numbers real? | Screenshots in place of platform data |
| Stay | Will the people who made the game remain? | Earn-outs, key-person terms, control after close |
The order matters. A buyer will forgive soft KPIs in a game it believes it can fix, but it won't price a game it can't own.
What should I do first if a buyer approaches my studio?
Sign an NDA before you share numbers, and don't agree to exclusivity yet. An inbound approach feels flattering, but one interested party isn't a process, and exclusivity gives away your best leverage early.
- Reply, and buy time. Thank them and ask what they have in mind.
- Sign an NDA first. Do this before any financials or roadmap leave the building.
- Bring in an adviser and counsel. A games-focused adviser can tell you who else would pay, and whether the approach is real.
- Test the market quietly. A second party changes the conversation, even if you never sell to them.
- Start the slow work now. Your IP schedule takes the longest, and every week before an LOI is cheaper than a week after it.
How active is game studio M&A in 2026?
Game studio M&A is busy by deal count, even though most deals are small. Drake Star counted 51 deals in Q1, 51 in Q2 and 45 in Q3 2026, for 147 so far. Its $2.4B Q3 total was "driven largely by divestitures" (Drake Star, Oct 2026). Be careful with headline totals: trackers book the same deal at different times, and 2026's value is dominated by Paramount's takeover of Warner Bros. Discovery, a media deal with a games division. Aream's chart counts 155 deals in the same period because it uses different rules, so the two shouldn't be mixed.
The better signal for founders is Aream's count of 18 developer and publisher control deals above $100M. Buyers are paying for content again. Recent examples:
| Deal | Status (Nov 2026) | Why it's useful |
|---|---|---|
| EA to PIF, Silver Lake, Affinity | Closed Aug 4, 2026, about $55B | 310 days from signing to close, by our count |
| Savvy to Moonton | Agreed Mar 2026, about $6B; shown as announced in October | No closing announcement we could find after six months |
| Scopely to Loom Games | Majority stake, over $1B valuation (Q1 2026) | A young studio can sell on one hit |
| Nazara to Bluetile | 50% for $100.3M, with earn-outs of about $98.2M | Nearly half of the headline is contingent, by our math |
| NCSoft to JustPlay | 70% for $202M | $173M revenue and $19M operating profit in 2025 |
Deals in Peony's State of M&A data took about 7.9 months on average in Q3 2026, across all sectors, which sits between one gaming adviser's figures of six months and 10 to 12 months. Plan for a long process.
Do you own everything in the game?
Prove ownership with a schedule, not a statement. Revision Legal's guide names copyright chain of title across "code, art, audio, narrative, and character designs" as the core review, and calls freelancers and contractors "the critical risk" (Revision Legal, Jan 2023, historical anchor). Ownership rules vary by country. A&O Shearman advises a careful review of employment and contractor agreements for that reason (A&O Shearman, Sep 2023, historical anchor). Under Polish law, for one, transfers of economic copyright need written form, so an email thread may not be enough (Wardyński & Partners, Jul 2021, historical anchor). That matters if your art or code was outsourced to Eastern Europe. Our IP due diligence guide covers the general process, and a buyer-side view shows how the same request list looks from the other side.
Engine licences. Check that you're on the right terms for your revenue. Unreal's standard licence charges 5% on gross revenue above the first $1M per product, as described in Meta's developer docs, Sep 2026, so buyers will check your royalty reports. Read Epic's current EULA for audit and assignment terms. Unity dropped its Runtime Fee, but Enterprise is required above $25M in revenue or funding over the last 12 months (Unity, accessed Nov 2026). A studio over that line and still on Pro is a concrete finding.
Open source. Catalogue it before a buyer's scan does. Across 947 audited commercial codebases in many industries, 68% contained open source licence conflicts (Black Duck, Feb 2026).
Music and voice. A commercial track needs both a sync licence and a master licence, and a commissioned score counts as work for hire only with the right written agreement (Legal Moves, Aug 2026). If you're a SAG-AFTRA signatory, the Interactive Media Agreement requires clear written consent for digital replicas and runs through October 31, 2028 (Frankfurt Kurnit, Jan 2026).
Generative AI. Keep an asset log. The Supreme Court declined to review Thaler v. Perlmutter in March 2026, and the Copyright Office has concluded that "prompts alone are insufficient" (Mayer Brown, Mar 2026). Purely prompted art may give you nothing to assign. Steam also asks about AI content that ships with the game (Steamworks, accessed Nov 2026).
Can you keep selling the game after the deal closes?
Find out by reading three stacks of paper: publishing, platform and regulatory. Each can change the value of the game overnight.
Publishing and licensing. Publishing agreements may say that a change in ownership "requires the consent of the other party" (Wardyński & Partners, Jul 2021, historical anchor). Sheppard Mullin warns that such terms "could delay or prevent" a deal (IGDA, Mar 2021, historical anchor). No public data shows how often these clauses appear, so assume yours has one until you've checked. If your game uses licensed IP, review scope, term and who owns anything new.
Platform economics. Model the fee after the deal, not before it.
| Platform | Current terms | What to check |
|---|---|---|
| Steam | 30%, falling to 25% above $10M and 20% above $50M; developers received 76% across non-Valve games in 2025 | Your actual blended rate |
| Apple | 30% standard; 15% small business up to $1M, with associated accounts counted together | Whether an acquirer's accounts end your 15% rate |
| Google Play | From June 30, 2026 in the US, UK and EEA: service fee from 10% on the first $1M, plus 5% billing fee on Play billing | Which rate applies to new and existing installs |
| Console | Sony, Microsoft and Nintendo don't publish standard rates that we could find | Your own agreements |
Sources: Game Developer, Nov 2018 (historical anchor), GameDiscoverCo, Mar 2026, Apple and PocketGamer.biz, 2026.
The Apple point is easy to miss. Apple counts associated developer accounts together, so check whether joining a larger group ends a studio's 15% rate. Also plan how store developer accounts move to the buyer, because a live game can't afford a gap. Concentration matters at every scale: even EA says Sony and Microsoft accounted for about 39% and 16% of its FY2026 net revenue (EA 10-K, 2026).
Regulation. Loot boxes and children's data are now live deal risks. On September 30, 2026, EU consumer authorities opened coordinated action on nine video game companies (LexisNexis, Sep 2026), examining whether "variable reward systems" comply with EU consumer law (Lewis Silkin, Oct 2026). The action is at an early stage, and no breach has been announced. New York's attorney general sued Valve in February 2026, calling loot boxes illegal gambling (National Law Review, Feb 2026). Brazil has barred loot boxes in games likely to be accessed by minors (Covington, Sep 2025, historical anchor). COPPA's amended rule has required compliance since April 22, 2026 (Hunton, Apr 2026). Bring your odds disclosures, age gates and data maps to the first request.
Do your live-ops numbers hold up?
Your numbers hold up when a buyer can tie them to platform data, not to your deck. A mobile game broker told AppAgent that buyers focus on "profit, retention, downloads from Tier 1 countries" and sessions, and that checks are often done through screenshots or direct console access (AppAgent, historical anchor). Offer read-only exports from App Store Connect, Google Play Console, Steamworks and your attribution platform early. It saves weeks of questions.
Compare your retention to these 2025 benchmarks, from 16,262 mobile games and 3,582 PC games (GameAnalytics, Jan 2026):
| Metric | Mobile median | Mobile top 25% | PC median | PC top 25% |
|---|---|---|---|---|
| D1 | About 22% | Just above 30% | About 9% | 15% to 16% |
| D7 | Just under 4% | 6% to 7% | 1.7% to 1.8% | 3.5% to 3.9% |
| D30 | 0.68% to 0.79% | 1.6% to 1.8% | Not used here | Not used here |
Buyers will also want cohorts by month and channel, revenue by title and country, ad spend against return, and refunds. A quality of earnings review will test the same data, including how you recognise in-game revenue and whether you capitalise development costs, so agree those policies with your accountant first. We found no reliable public benchmark for chargebacks or LTV/CAC, so show your own trend. Pay attention to how you handle disputes: the FTC's order against Epic bars locking players out for disputing unauthorized charges, after the FTC alleged that Epic had blocked some who did. Epic settled those allegations (FTC, Jul 2025, historical anchor).
Will the team stay, and what did Krafton change?
Krafton showed that a buyer can lose by taking control too early. Krafton bought Unknown Worlds, maker of Subnautica, for $500M upfront plus up to $250M in earn-outs. After a dispute, Delaware's Court of Chancery held on March 16, 2026, that Krafton lacked cause to fire key employees and that its seizure of operational control was "independently wrongful." The court reinstated the CEO and extended the earn-out period by 258 days (Mayer Brown, Apr 2026). The court left damages for a second phase and didn't restore the founders to their roles. The parties then settled, effective June 30, 2026, and the reinstated CEO left the studio; the sources we read don't state the amount (Korea Times, Jul 2026).
Three lessons carry over to any studio deal:
- Spell out control. Unknown Worlds' key employees kept operational control during the earn-out while any of them remained. That term, with the narrow definition of cause, drove the result.
- Narrow the definition of cause. Here it was limited to felonies, intentional fraud or dishonesty, gross misconduct and wrongful disclosure of trade secrets or confidential information.
- Assume your messages are evidence. The court quoted AI chat responses as insight into the buyer's motives. Mayer Brown's note says "AI communications may be discoverable."
Founders should also take a hard look at the earn-out itself. Brogan Keane of Double Black Capital says retention earn-outs run two to five years, and that performance earn-outs are "more likely not to hit than hit" (Naavik, Jun 2026). Weigh the contingent part of any offer on that basis. Nazara's Bluetile deal shows how far this goes: Aream's EBITDA multiple went from 7.2x in its Q1 report to 10.9x in its Q3 report, a change we attribute to including the earn-out. So ask whether a headline multiple counts money you may never receive.
What is a game studio worth?
A studio is worth what its model supports, and that varies widely. Public markets show the spread: PC and console companies traded at 12.6x forward EBITDA on September 30, 2026, against 4.4x for Western mobile-first companies (Aream & Co., Oct 2026). In private deals, one adviser says "you might see a 4X, 5X potentially on a big hit game," with work-for-hire studios between 1.5x and 2.5x revenue, and adds that private equity rarely pays 20 times profit (Naavik, Jun 2026). That's a single view, as multiples below $100M are rarely disclosed.
If your revenue comes from one live game, expect a buyer to discount for decay and to push part of the price into an earn-out, so bring monthly cohorts and a funded plan for the next title. If you do contract work, expect a services multiple. If you have your own IP with several titles, argue for the higher range and bring the cohorts to support it. Private equity is a growing buyer: Drake Star expects more PE-led deals, and our private equity data room guide shows what sponsors expect.
What goes in a game studio data room?
A game studio data room should follow the four questions, so that a buyer can find an answer without asking. Use this index as a starting point:
| Folder | Contents | Release stage |
|---|---|---|
| 1. Corporate | Formation documents, cap table, board minutes, group structure | First round |
| 2. Own | IP schedule, contractor and employee assignments, engine licences, OSS list, music and voice contracts, AI asset log | First round, names redacted |
| 3. Keep | Publishing and licence agreements, platform agreements, age ratings, loot box odds, privacy policies and data maps | Second round |
| 4. Prove | Audited or reviewed accounts, revenue by title and country, cohorts, marketing spend, refunds | First round summary, then detail |
| 5. Stay | Org chart, key employee contracts, bonus plans, retention plans | Second round, names redacted |
| 6. Roadmap | Content pipeline, budget by title, unannounced projects | Final bidder only |
| 7. Technology | Architecture notes, third-party components, scan results; source code only by separate process | Final bidder only |
Three habits make this work. Stage the release, so that a competitor who drops out hasn't seen your unannounced game. Give each bidder a separate group, as covered in our multiple bidders guide. And use the analytics: a buyer who rereads the publishing agreement three times has a question coming. Our sell-side due diligence guide and M&A data room guide go deeper on preparing the room.
How does Peony fit a game studio sale?
Peony gives a studio a buyer-ready room at a published price. Riot Games and 8,000+ teams use it, and viewers are free, so a large buyer team doesn't change the cost.
- Structure: AI room generation and auto-indexing on Data Room ($52/admin/month) build and number the folders.
- Permissions: per-bidder visitor groups and per-file permissions on Data Room support staged release.
- Leak control: dynamic watermarks and signed NDAs on Data Room, with anomaly alerts on Enterprise.
- Insight: page-level analytics on every plan show which bidder read what.
- Process: Advanced Redaction and Advanced Q&A on Deal Team ($64/admin/month, 4-admin minimum).
Peony is SOC 2 Type II-ready and ISO 27001-ready, with both audits underway. If a large strategic buyer's security team needs an issued report, raise it early. Peony also isn't the right choice if your counterparty insists on a specific legacy provider.
Bottom line
- Founder of a small studio: start your IP schedule now, because it's the slowest item to fix. Set up a Peony Data Room before you speak to buyers.
- Studio with a live game and several bidders: add Deal Team for Advanced Q&A and redaction, and take advice on any earn-out.
- Corporate development or PE buyer: ask the four questions first, and read Krafton before you draft the operating covenants.
If you're raising rather than selling, see our top gaming investors and fundraising data rooms. For the rest of what studios share, see Peony for game studios and our closed playtest guide.
Frequently Asked Questions
I'm selling a 30-person game studio. What will buyers ask for first?
Buyers will ask whether you own the game, then whether the revenue is real. Expect requests for contractor IP assignments, engine and platform agreements, publishing contracts, and store-console exports of your live-ops numbers. For a 30-person studio on a $30M to $50M sale, have those ready before the first NDA. In a Peony data room on the Data Room plan ($52/admin/month), page-level analytics show which bidder is reading them, which a shared drive can't.
We used freelance artists and outsourced QA. Is that a problem for a sale?
It can be, if you can't show a signed assignment for each contributor. Copyright doesn't always move to the hiring party automatically, and the rules vary by country, so buyers read the contracts rather than assume. For a studio that used 40 freelancers, build an IP schedule that lists each person, what they made and the signed agreement. Keep it in a restricted folder in Peony with per-bidder permissions, so that only late-stage buyers see names and rates.
What happens to our publishing deal if we sell the studio?
It depends on the contract, so read it before you market the business. Publishing and license agreements may require the other party's consent to a change in ownership, which can delay or block a deal. If you're a 25-person studio with one publisher, check change-of-control, recoupment and sequel rights first. Put the agreements in your data room with Advanced Redaction on Peony's Deal Team plan ($64/admin/month, 4-admin minimum), so that early bidders can't see commercial terms.
How long does it take to sell a game studio in 2026?
Plan for six months to a year. One gaming M&A adviser said in June 2026 that about six months is the usual expectation, but recent deals have taken 10 months to a year. No public dataset tracks this. A prepared data room shortens the part you control: Peony's AI room generation and auto-indexing on Data Room ($52/admin/month) build the structure in minutes, so a small team can open diligence the day a term sheet lands.
What multiple should a game studio expect from a buyer?
It depends on the business model more than on the headline market. One adviser puts big-hit studios at 4x to 5x revenue and work-for-hire studios at 1.5x to 2.5x, while public PC and console companies traded at 12.6x forward EBITDA in September 2026 against 4.4x for Western mobile. Private small-studio multiples are rarely disclosed. To defend yours, show buyers clean KPIs in a Peony room and see which numbers they keep revisiting.
Should a founder accept a large earn-out in a studio sale?
Only if the contract protects your ability to hit it. One adviser says performance earn-outs in games have more often missed than hit in recent years. Unknown Worlds' key employees won on liability in Delaware because their deal gave them operational control and limited what counted as cause. For a founder with 40% of the price at risk, negotiate control, budget and cause definitions. Keep every draft and board paper in one Peony room with a clean audit trail.
Do buyers care if we used generative AI for game art?
Yes, because it affects what you can warrant. The US Copyright Office has concluded that prompts alone are not enough for copyright protection, and Steam asks developers to disclose AI content that ships with a game. For a mobile studio selling a live title, keep a log of which assets used AI and what humans changed. A buyer will ask for it, and a Peony folder with dynamic watermarks on Data Room lets you share it without losing track of who has it.
Should we put our source code in the data room?
Not at the start. Share architecture notes, third-party component lists and scan reports first, and keep code access for the final bidder under a separate process agreed with counsel. For a studio with a live game, source code is the crown jewel, and a failed deal with a competitor can hurt. Put the summaries in Peony, with per-file permissions and dynamic watermarks on Data Room ($52/admin/month), so that each bidder's copy is identifiable.
A publisher emailed out of the blue about buying our studio. What do we do first?
Reply politely, sign an NDA before you share financials, and don't grant exclusivity yet. Then speak to an adviser and counsel, because one bidder rarely gives you a fair price, and start your IP schedule. For a 40-person studio, a Peony data room on Data Room ($52/admin/month) with signed NDAs lets you share a first set of numbers with a watermark on every page, which a shared drive won't do.
Most of our revenue comes from one game. How will buyers treat that?
Expect a discount for concentration and a close look at how fast the game is declining. Buyers will model the decay, ask what the next title has to do, and may move price into an earn-out. For a studio with 80% of revenue in one live game, bring monthly cohorts and a funded pipeline plan. In Peony, page-level analytics show which bidder keeps returning to your cohort data, so you can prepare for that question.
Which Peony plan does a studio need for a sale?
Most studio sales fit the Data Room plan at $52/admin/month, with dynamic watermarks, signed NDAs, granular permissions and unlimited storage. Choose Deal Team at $64/admin/month with a 4-admin minimum if you expect several bidders and need Advanced Q&A or Advanced Redaction, and Enterprise if you need anomaly alerts or SAML SSO. Viewers are free, so a 50-person buyer team won't change your bill, unlike legacy data rooms that price by page or volume.

