The Video Games Market in 2025-26: Where the Money Sits

Global games revenue passed $200 billion for the first time in 2025. Newzoo's updated figure closed the year at $201.6 billion, with PC the fastest-growing platform at 12% year on year, mobile up 10.7% and console up 2.8% (Newzoo via GamesBeat, 2025). The same report counts 3.58 billion players, roughly 60% of everyone online, and notes that player growth is slowing as the market matures. For a founder, that last point matters more than the headline: the pie is large, but new customers are no longer arriving for free. Revenue growth now comes from spending per player, not from more players.

The United States is the biggest single buyer. The Entertainment Software Association puts 2025 US consumer spending at $60.8 billion, split into $52.4 billion on content, $5.4 billion on hardware and $2.9 billion on accessories (ESA, 2025 spending release). Its 2026 Essential Facts report counts 212.3 million Americans playing every week (ESA Essential Facts, 2026). Content is 86% of that spend. If you are writing a plan for a US lender, this is the sentence to anchor on: most dollars go to software and in-game purchases, not boxes or consoles.

The UK is smaller but unusually well organised. Consumers spent £8.76 billion on games in 2025, up 7.4% on 2024 and the highest annual figure recorded, with software at £6.03 billion (UK Gaming Market Statistics 2026, citing UKIE). The labour picture is mixed: between May 2024 and September 2025 there were 3,655 full-time job cuts across 491 companies, while 513 studios added 2,571 roles, leaving UK games employment at 27,347 against 28,516 a year earlier (same source). Layoffs at large studios have pushed experienced developers toward founding small teams, which is why competition for Steam shelf space is heavier than the revenue curve suggests.

Four businesses hiding under one keyword

"Video games business" describes four different companies, and the plan changes more between them than between, say, a bakery and a cafe. A developer turns salaries into a product and hopes for a long tail of sales. A publisher or label turns capital into marketing and takes a share of net revenue. A retailer turns inventory into margin and lives on trade-ins. A venue (lounge, arcade, esports room) turns floor space into hourly or per-visit revenue. Lenders and investors read each differently: a bank wants collateral and cash-flow cover from the retailer, an angel wants a team and a prototype from the developer, and a grant body wants UK or Ontario or Quebec spend from either. Pick one model before you write a word of the plan. Studios that blend two (develop and publish others, or run a shop and a studio) can do it, but the plan should show them as separate profit centres with separate assumptions.

The survivorship problem every plan has to confront

Case studies in this sector are all outliers. Stardew Valley, built over four years by one person under the name ConcernedApe, has passed 30 million copies; Hollow Knight was made on a Kickstarter budget of about $57,000 and has sold over 15 million units; its sequel Silksong sold more than 7 million in its first three months; Balatro, a solo project by LocalThunk released through publisher Playstack, has sold around five million copies (Stardew Valley sales commentary, VGChartz on Hollow Knight, Balatro). They are real, and they are the reason founders start. They are also the top sliver of a distribution whose middle sits far lower. Steam Page Analyzer's 2026 dataset puts the median indie game at $5,000 to $15,000 lifetime gross, roughly 500 to 2,000 copies, with about half of releases never passing $5,000 and only 10 to 20% recouping their development cost (Steam Page Analyzer, 2026). Treat those as third-party estimates rather than audited statistics, but the direction is consistent across every dataset we reviewed. A credible plan shows the base case at the median-to-good range, and the hit as an upside scenario with its own label.

Questions Founders Ask First

These come up in search before any spreadsheet opens. Short answers here; the detail sits in the sections that follow.

How much does it cost to start a video game company?

It depends almost entirely on whether you pay yourself. Cash-only costs for a solo developer run about $5,000 to $50,000 (software, contract art or music, a trailer, Steam's $100 fee), but that figure ignores one to four years of unpaid labour (Steam Page Analyzer). A paid team of four to eight for 18 to 24 months lands between roughly $450,000 and $1.4 million. A small game shop needs about $95,000 to $330,000.

How do video game companies make money?

Seven ways in practice: premium sales (one price), free-to-play with in-app purchases, subscriptions and live-service passes, advertising in casual mobile, platform or publisher advances and grants, work-for-hire and porting, and licensing or merchandise. Most first-time studios earn from the first and from advances. The plan should say which two you rely on and what share each represents in years one to three.

Do I need a business plan to publish a game on Steam?

No. Steam Direct asks for a legal entity, tax and bank details, and a $100 fee per game that you recoup after $1,000 in adjusted gross revenue (Xsolla guide to self-publishing on Steam). You need the plan for everyone else in the chain: the lender, the angel, the grant assessor, the publisher's business development team and, if you are in the UK, the person completing your Start Up Loan application.

Can I get a loan to start a studio?

Yes, with caveats. Software studios have almost no hard assets, so lenders lean on the founders' personal credit, a published or shipped title, or a signed publisher contract. In the UK, the Start Up Loans scheme lends up to £25,000 per founder at a fixed 6% with free mentoring, and SEIS can sit alongside it for equity. In the US, SBA 7(a) loans suit retail and venue models far better than pre-revenue developers. The regional section below covers where the money is.

Should I build for PC, console or mobile first?

Pick the platform your genre already lives on. Deckbuilders and management games sell on PC; narrative adventures travel to Switch and PlayStation; hyper-casual and puzzle products live on mobile. PC grew fastest in Newzoo's 2025 numbers (12%), and PC is the cheapest place to ship because you avoid console kit and certification costs, so most first titles launch on Steam and port later.

What a Video Games Business Costs to Start, by Model

Startup cost in this sector is a function of headcount and months. Everything else (engine licences, a Steam deposit, dev kits) is small by comparison. That is why plans that list "software $2,000, equipment $8,000, marketing $5,000" and stop there are not credible to anyone who has shipped a game. Below are the four models with the cost bands we use when building client plans. Treat them as planning ranges; local wages and your own scope will move them.

Model 1: solo or two-person digital developer

• Cash outlay over 12 to 24 months: $5,000–$50,000 (£4K–£40K), excluding founder wages
• Contract art, animation, music, sound: $3,000–$25,000 depending on how much you cannot do yourself
• Software and middleware: $0–$3,000 a year (Godot is free; commercial middleware varies)
• Steam deposit, capsule art, trailer edit: $100 plus $500–$4,000
• Hidden cost: the founder's living costs. Plan 12 to 24 months of runway, or a day job that funds the project

Model 2: funded studio of four to eight

Industry guides put typical indie Steam budgets between $30,000 and $500,000, but a salaried team sits at the upper end of that range or above (Cubix pricing guide, 2026). For planning, assume $450,000 to $1.4 million (£350K–£1.1M) over 18 to 24 months, built up as follows:

• Salaries and payroll costs: 65–75% of the total. A fully loaded cost of $4,000–$8,000 per person per month is a sensible regional range
• Contractors (art, audio, narrative, localisation): 8–15%
• QA, certification and platform fees: 3–6%, higher if you add console ports
• Marketing: at least 15–25% of development cost. Studios that put 3% against it are choosing to be invisible
• Legal, accounting, insurance, premises or co-working: 3–5%
• Contingency: 10%. First projects overrun, and the usual cause is scope

Model 3: publishing label

A label's startup cost is its advance pool. A small label signing three or four games a year might hold $600,000 to $1.5 million in committed capital (advances of $150,000 to $400,000, plus matched marketing), with two or three staff running business development, production and community. Devolver Digital, the Texas publisher founded in 2009 whose breakout was Hotline Miami, carried a portfolio of about 130 games at the end of 2024 (Devolver Digital); that is the scale of the long game. A new label starts with a thesis (a genre, a region, a platform) rather than a catalogue. See also our video game publisher template.

Model 4: retail store or lounge

A compact US game store of 1,000 to 1,800 square feet typically needs an initial cash plan of $95,000 to $330,000, with opening inventory the largest line at 40 to 50% of the budget (StartupFinancialProjection estimate). Vintage and retro stores are cheaper to open, $30,000 to $100,000 by the same family of sources, and an online-first reseller on eBay or Shopify can start on $2,000 to $5,000. These are third-party modelling estimates, so replace them with your own supplier quotes and lease terms before they reach a lender. A lounge adds fit-out, console and PC hardware, and licences, and is covered in our video game lounge template.

Funding routes by model

Developers usually stack: personal savings, a small prototype grant or loan, equity from angels (SEIS and EIS in the UK, SAFE notes in the US), a publisher advance and, later, a tax credit that pays out after spend. Publishers need larger equity or a fund. Retailers and venues are the natural customers for SBA 7(a) loans and conventional bank debt because inventory and fit-out give lenders something to lend against. Where a plan combines sources, show a sources-and-uses table that adds up and a month-by-month cash curve that never goes below zero. Lenders and publishers both read the cash curve first.

Platforms, Engines and Vendors You Will Actually Pay

A games business has fewer suppliers than a restaurant, but each one takes a percentage of revenue or sets a rule you must plan around. Name them in your plan. A lender who sees "Steamworks, Epic, Unreal royalty at 5% after the first $1 million" knows you have read the contracts; a lender who sees "distribution costs" does not.

Storefronts and what they take

• Steam (Valve): $100 per game through Steam Direct, recouped after $1,000 in adjusted gross revenue. Valve keeps 30% of revenue up to $10 million per title, 25% between $10 million and $50 million, and 20% above that (Xsolla, Immutable)
• Epic Games Store: a 12% revenue share, with Epic covering the Unreal Engine royalty on sales made through its store, and no store cut on a title's first $1 million in revenue under its 2025 programme (PC Gamer, PocketGamer.biz)
• Apple App Store and Google Play: both charge a standard 30%, falling to 15% on the first $1 million of annual proceeds for qualifying small developers. Apple requires enrolment in its Small Business Program; Google applies the reduced tier automatically (PricePush summary)
• Console storefronts (PlayStation, Xbox, Nintendo eShop): you apply to each platform holder as a registered company. Sony and Microsoft run programmes that supply dev kits, Microsoft through ID@Xbox and Sony through its PlayStation Partners programme, and Nintendo uses a developer portal plus paid dev hardware (Xsolla on dev kits)
• itch.io: an open-split storefront where the developer sets the share; useful for demos, game-jam builds and early community, rarely the main revenue source

Engines, tools and middleware

• Unreal Engine (Epic): free to start; a 5% royalty applies to lifetime gross revenue above $1 million per product, reduced to 3.5% for games that launch on the Epic Games Store before or alongside other stores
• Unity: the dominant engine for mobile and small PC titles; re-check its current runtime and plan terms at the time you sign, since they have changed more than once
• Godot: open source with no royalties, a growing choice for 2D and small 3D teams who want no licensing line in the model
• Blender, Aseprite, FMOD, Wwise: 3D art, pixel art and audio middleware; budget small annual licence costs where needed
• Perforce or Git, Jira or Linear, Discord: version control, task tracking and community. Large binary assets make version control a genuine line item in a 3D project
• Steamworks tools and a wishlist tracker: free, and the place where your marketing plan becomes measurable

Retail suppliers (store and lounge models)

New-release stock comes from distributors on thin margins, which the figures below make clear. Independent shops therefore build gross margin on trade-ins, used and retro stock, accessories, repairs and collectibles. The cautionary data point is the incumbent: GameStop reported 3,203 stores at 1 February 2025, shed 590 locations in fiscal 2024 and confirmed over 470 further closures (GameStop 10-K, Yahoo Finance). A neighbourhood shop is not competing with GameStop's size; it is serving the collectors and families the closing branches leave behind.

How Video Games Businesses Make Money, With Worked Maths

Four calculations carry most of this sector. Run them before you build the full model, and put the results in the plan's financial section. Every figure below is illustrative; labels say which assumptions are ours.

1. A premium PC game from a three-person studio

Assume a $19.99 list price. After launch discounts, bundles and regional pricing, the average realised price is $14.50 (our assumption; check against comparable titles). Year-one units of 12,000 and a 6% refund rate leave 11,280 net units, which is $163,560 after refunds. Steam's 30% takes $49,068, so the studio receives $114,492. Sales taxes and VAT are collected and remitted separately and sit outside this figure.

Now the cost side. Three people for 20 months at a loaded $4,200 per person-month is $252,000. Add $30,000 of contract art and audio, $6,000 of software, $14,000 of QA and localisation, $40,000 of marketing, $8,000 of legal and accounting, and the $100 Steam fee, for $350,100, plus a 10% contingency of $35,010: $385,110 in total. Net revenue per unit is $14.50 x 94% x 70% = $9.54, so the game needs about 40,400 units to break even. Year one at 12,000 units returns roughly 30% of cost, which is why the median studio keeps a second job or a second product.

The last tile applies the UK Video Games Expenditure Credit. Core spend (salaries, contractors, software and QA) is $302,000; the credit works on the lower of 80% of core expenditure or UK core expenditure, so $241,600 qualifies, and 34% of that is about $82,100. The credit is taxable and arrives after the game is certified and the claim is filed, so it improves the economics but not your opening cash. Net cost drops to about $303,000 and break-even to roughly 31,800 units.

One more pattern worth modelling is timing. Steam Page Analyzer estimates that launch week delivers roughly 22 to 38% of first-year gross revenue for the median game (Steam Page Analyzer, first-week data). A plan that spreads sales evenly over twelve months will show cash that does not exist in months four to twelve. Use a front-loaded curve with a long tail, and a lift line for each major sale.

2. A publishing deal, from both sides

Terms vary widely, so label this a common shape and not a rule. A label advances $250,000 to the developer and spends $100,000 on marketing, all recoupable from net revenue. The game earns $900,000 after platform fees. The label takes the first $350,000 to recoup, then the remaining $550,000 splits 70% developer, 30% label. The developer receives $385,000 from the split on top of its $250,000 advance, $635,000 in total. The label receives $350,000 plus $165,000, $515,000, a profit of $165,000 on $350,000 invested, or 47%. The plan insight: a label's return is dominated by the games that recoup well, so a thesis of ten small bets needs the one winner that pays for the rest. A developer reading this should model the plan with and without a publisher; independent revenue is higher per unit but marketing reach is lower.

3. A free-to-play mobile title

Mobile plans run on four numbers: installs, retention at day 1, day 7 and day 30, average revenue per daily active user (ARPDAU), and the cost to acquire each install. A simple test: if a user costs $1.80 to acquire and lifetime revenue per install is $1.20, scaling spend makes you poorer. Model the loop on a spreadsheet before you hire anyone, and be wary of any plan that grows paid installs without a payback period under 90 days. Apple and Google both take 15% on the first $1 million of annual proceeds for small developers, 30% beyond that, so net revenue per user improves materially for studios that stay under the threshold.

4. An independent game store

New AAA releases give retailers only 5 to 15% margin while used titles can yield 40 to 60%, which is why independent stores live on trade-ins (FinancialModelsLab retro store insights). Take a 1,400 square foot shop with $520,000 annual revenue and a blended gross margin of 45%, giving gross profit of $234,000. Costs: rent $42,000 (about $30 per square foot), payroll $88,000 including a $48,000 owner wage and 0.8 of a full-time employee, utilities, insurance and POS $18,000, marketing $10,000, shrink, repairs and returns $14,000, other $9,000. That totals $181,000, leaving $53,000 of operating profit, about 10% of revenue. Third-party sources quote 15 to 25% net margins for retro stores; our figure is lower because it pays the owner a wage before profit, which is how an SBA lender will underwrite it.

Where You Build Matters: Tax Credits and Funding by Region

Unlike most small businesses, a games studio can choose where to be based, and the incentive differences are large enough to change the plan. The table below compares the three jurisdictions most of our clients ask about. It is a planning summary, not tax advice; confirm eligibility with an accountant who files these claims.

Sources: Carpenter Box on VGEC, Creative Accountants on VGEC, Ontario Creates and Ryan on Canadian game funding.

UK: clusters and the funding stack

The UK has established studio clusters in Dundee (home to Abertay University's games courses), Guildford, Leamington Spa, Brighton, Sheffield and Manchester, and a trade body, Ukie, that lobbies for sector support. UKIE's 2025 campaign work pressed for a higher "games growth" credit rate of 53% for projects with budgets of £10 million or less, and the government has recognised games as a priority growth sector with £30 million committed to the UK Games Fund and UK Video Games Council (Ukie campaigns, ICAEW industry profile). Treat these as proposals and announcements rather than guaranteed rates until legislated. A typical UK stack in a plan: founder capital, a Start Up Loan (up to £25,000 per founder at 6%), SEIS or EIS equity for investors, a prototype or publisher advance, and the VGEC at the end of production.

Canada: two strong credits

Montreal and Toronto compete directly on the value of the credit. A studio doing its own IP can claim Ontario's 40% on eligible labour; fee-for-service work earns 35%. Quebec's credit is capped at 37.5% of eligible salaries but sits on a deep local labour pool. For a UK founder choosing a second base, Ontario's refundability makes cash flow modelling simpler.

United States: scale instead of subsidy

The US offers no federal games credit, but it gives you the largest paying audience, 212.3 million weekly players, and the deepest publisher and investor market. Austin (home of Devolver Digital), Seattle and Los Angeles are the main hiring and publisher hubs. For retailers and venues, SBA 7(a) loans, up to $5 million with terms of up to 25 years for real estate, are the common route, and lenders will ask for a lease, equipment quotes and a three-year projection. Our free business plan template hub carries the SBA structure, and the Bespoke plan includes lender-ready projections.

Ratings, Privacy and Licensing: The Compliance Cost Nobody Budgets

United States

• Entity and tax: an LLC or C-corp, an EIN, and sales-tax registration in states where you sell physical goods. Digital storefronts usually collect and remit tax on your behalf
• COPPA (FTC): if children under 13 may play, you need verifiable parental consent before collecting personal data. Epic Games paid a $275 million civil penalty in 2022, the largest ever for an FTC rule violation, plus $245 million in refunds over dark patterns (FTC, 2022)
• ESRB rating: a self-assigned rating through the ESRB's questionnaire, required by console storefronts and most retailers
• State privacy laws: California and an expanding list of states add consent, deletion and disclosure duties for player data
• Music and asset licences: keep a rights register for every font, sound and sample. One unlicensed asset can pull a title off a store

United Kingdom

• PEGI and the Video Recordings Act 1984: boxed games must carry a PEGI rating, and supplying 12, 16 or 18-rated physical games to underage buyers is an offence. There is no equivalent offence for digital supply (Game Developer)
• ASA loot box disclosure: the Advertising Standards Authority issued an enforcement notice requiring ads and app store listings to disclose loot boxes, with active monitoring beginning on 26 May 2026 (ASA)
• Online Safety Act: games with user-to-user chat or user-generated content fall within Ofcom's rules on illegal content and child safety
• UK GDPR and the ICO Children's Code: data protection by design for any game likely to be accessed by under-18s
• Premises (retail or venue): business rates, fire risk assessment, public liability insurance, PRS or PPL licences where music plays

European Union and elsewhere

Loot boxes are the clearest cross-border risk. Belgium treats paid loot boxes as gambling, and several EU regulators have followed with consumer-law guidance (Promise Legal, 2026). Design one monetisation system for all regions and you may be forced to remove it in one. Germany, Australia and most large markets also run their own age-rating bodies, though many storefronts now accept the IARC system, a single questionnaire that returns ratings across territories. In Canada, check provincial consumer-protection rules for virtual currency and the credit eligibility conditions described above.

Six Mistakes That Sink First-Time Studios and Game Shops

These recur in the plans we are asked to rescue. Each is a line you can fix on paper, which is far cheaper than fixing it after launch.

1. Budgeting for development and treating marketing as optional

The median indie title earns $5,000 to $15,000 lifetime because most never reach an audience, not because most are bad. Plan for a wishlist target before launch, a named festival or showcase slot (Steam Next Fest, Gamescom, PAX, EGX), a press and creator list, and a spend line equal to at least 15 to 25% of development cost.

2. Modelling gross sales instead of net receipts

Between list price and your bank account sit regional pricing, discounts, refunds, platform fees, withholding tax, publisher recoupment and currency conversion. The worked example above loses nearly half of list price before it is cash. Show that waterfall as its own table.

3. No scope control

Teams add features because they can, then discover each one adds testing and balancing time. Set a vertical slice date, a feature freeze date and a content-complete date, and attach funding tranches to them. Publishers do this through milestone payments; if you self-fund, do it to yourself.

4. Ignoring child-privacy and monetisation law

The $275 million COPPA penalty in the Epic Games case is the cautionary tale, but small studios get caught by far smaller issues: an analytics SDK that collects data from under-13 players, an unlabelled loot box in a UK app listing, a chat feature with no moderation plan. Put a compliance line, even a modest one, in the budget and name an owner.

5. Choosing a model without a unit of survival

Every games business should state how many months it survives without revenue, how many it needs to ship, and what happens at each shortfall. A studio with 14 months of cash and an 18-month schedule is a failed plan on day one, whatever the market data says.

6. For retailers: treating new-release stock as the business

New games are the traffic driver; used and trade-in stock is the profit. Plan the trade-in appraisal process, the grading and testing bench, a returns policy and the shrink allowance. Cash tied up in sealed stock at 5 to 15% margin is how shops run out of money in a good sales month.

Launch Timeline, Month by Month, for a Small Studio

This sequence assumes a three to five person team, a PC-first premium game and a Steam release. Compress or stretch to suit your scope.

• Months 0 to 3, concept and prototype: register the company, open a business bank account, build a playable prototype, record a 60-second gameplay video, and apply for a prototype grant or a Start Up Loan with the plan attached
• Months 3 to 6, vertical slice: one polished level that shows final quality. This is your investor and publisher demo. Open the Steam page as early as you have a trailer and capsule art, because wishlists need time to compound
• Months 6 to 9, funding and pitch: pitch publishers with the slice and the plan. Close angel funding under SEIS or a SAFE. Do not wait until cash is low; publishers read desperation quickly
• Months 9 to 16, production: build content, hold monthly playtests, hit milestones. Start the BFI Cultural Test paperwork if you intend to claim VGEC, since it is easier to document as you go
• Months 16 to 18, beta and certification: external QA, localisation, rating questionnaires, age ratings, and a Steam Next Fest demo if the schedule fits
• Month 18 to 20, launch: pick a date away from large releases, seed creators, and watch refund rate and reviews in the first 72 hours
• Months 20 to 24, tail and next: patches, a discount cycle, platform ports, the tax-credit claim and the plan for the second product, which is the one that matters most to your investors

Glossary of Industry Terms You Will Meet in a Pitch

• Wishlist: a Steam user's saved intent to buy. Teams use wishlist counts at launch to forecast first-week sales
• Vertical slice: a short, near-final-quality section of the game that proves the whole can be built to that standard
• Recoup: repayment of a publisher's advance and marketing spend from net revenue before the developer receives royalties
• Milestone payment: a staged payment from a publisher or funder, released when an agreed deliverable is accepted
• Certification (cert): the compliance testing a console platform holder runs before a game may ship
• ARPDAU: average revenue per daily active user, the core monetisation figure for mobile games
• Live ops: post-launch events, seasons and updates that keep revenue coming from an existing audience

Sample Business Plan Extract

Here is an extract from a video games business plan of the kind our team writes, so you can see the level of specificity you are buying.

What the Template Contains

Every Avvale template follows the structure lenders and investors expect, and the video games edition adds the pieces below that a generic template misses:

• Executive Summary: game or business concept, audience, platforms, funding ask and the single metric that proves demand
• Company Overview: legal structure, founders' shipped credits, IP ownership and any assignment agreements
• Market Analysis: genre size, comparable titles and their review counts, platform mix and regional pricing
• Product and Development Plan: engine choice, vertical slice date, milestones, risk log and technical debt policy
• Marketing and Community Plan: wishlist targets, showcase calendar, creator outreach and community channels
• Platform and Distribution Plan: storefront fees, console programme applications, certification and ports
• Operations and Compliance: ratings, privacy, loot box position and tax credit eligibility
• Management Team: founder credits, advisers and key hires

The Financial Forecast add-on, included in our $300/£250 and $1,000/£800 packages, provides a five-year Excel model with the net-revenue waterfall, sources and uses, a monthly cash curve, break-even analysis and tax-credit schedule. For related models see our indie game studio template and bespoke plan service, or read the Avvale case studies.

Frequently Asked Questions