A strict test of sector runway, leadership, moat, balance-sheet strength, and free cash flow. Evidence through September 23, 2026.
Matthew Bromberg canceled the Runtime Fee, reset the portfolio, and turned Unity from a cash-burning story into a compounding one: six straight quarters of adjusted EBITDA margin expansion, Vector past a $1 billion run rate, free cash flow strongly positive, and the balance sheet flipped to net cash. The one partial is the moat, where Unreal and AppLovin keep Unity honest.
The moat is the one partial, where Unreal and AppLovin keep Unity honest. The other four pass: sunrise sector, leadership (Bromberg's turnaround record), balance sheet (flipped to net cash), and free cash flow (strongly positive). The moat upgrades if the Vector data moat deepens and Unity 7 adoption broadens.
The sunrise sector downgrades on a sustained collapse in mobile game demand or a developer exodus toward AI-native creation platforms. Leadership downgrades on a guidance cut, renewed trust damage with developers, or a stall in margin expansion. The moat upgrades on Vector deepening and broad Unity 7 adoption.
The five-pillar engine asks five questions. Each pillar passes, partially passes, or fails on evidence, not narrative. A partial grade means the evidence is real but incomplete. A fail means the current state does not meet the standard.
| Pillar | The question | Grade |
|---|---|---|
| 1. Sunrise sector | Is the industry growing for a decade or more? | Pass |
| 2. Leadership | Has the leader scaled before, and done what they said? | Pass |
| 3. Moat | Can competitors take the customers? | Partial |
| 4. Iron-fortress balance sheet | Can the company survive a bad two years without help? | Pass |
| 5. Free cash flow | Does the business generate real cash, repeatedly? | Pass |
Pass
The question: is the industry growing for a decade or more?
Real-time 3D and interactive content is a multi-decade runway. Gaming is the largest entertainment medium in the world, the mobile gaming ad market is recovering, and XR plus industrial digital twins are long-runway optionality. Unity's engine is deeply embedded in mobile game creation, and the demand signal is no longer theoretical: Vector, its AI-powered ad platform, grew 23% quarter over quarter in Q2 2026 against management's own 12-13% expectation and passed a $1 billion annual run rate two quarters ahead of plan. A billion-dollar ad product built on the engine's runtime reach shows the market is real, not aspirational.
The September 23, 2026 Meta Connect keynote reads modestly positive to neutral for Unity. Meta announced Meta VR Glasses at $1,299.99, available spring 2027: a glasses-form-factor VR device five times lighter than Quest 3 with a separate compute puck, micro-OLED lenses, and compatibility with the existing Quest game catalog. A lighter, higher-resolution headset that runs the existing catalog expands the addressable hardware base for Unity-built XR applications, which supports the Create line. Meta's Horizon-native creation layer competes with Unity, but only for lower-complexity creators: it does not replace Unity for full native Quest games and cross-platform applications. No new Quest-branded headset was announced, no Unity-specific partnership was disclosed, and the September 24 developer sessions were still pending at the time of this research. It is not a standalone Unity catalyst.
What would change the grade: already at the ceiling. Downgrade only on a sustained collapse in mobile game demand or the mobile ad market, or a developer exodus away from real-time 3D tooling toward AI-native creation platforms. None is visible today.
Pass
The question: has the leader scaled before, and done what they said?
Matthew Bromberg has been President and CEO since May 15, 2024, succeeding interim CEO Jim Whitehurst. He came in with turnaround credentials: COO of Zynga from August 2016 to April 2022, where he played a key role in that company's turnaround and ran game studios globally plus product development, design, technology, data, and analytics; before that, Senior Vice President of Strategy and Operations for EA Mobile and Group General Manager of all BioWare studios worldwide; earlier, President and CEO of Major League Gaming, founder and CEO of I'mOK Inc, and SVP/General Manager at AOL from 1999 to 2005. He holds a B.A. in English from Cornell and a J.D. from Harvard Law School, and joined Unity on an equity-heavy package aligned to the share price.
The record since arrival is the evidence. Bromberg canceled the Runtime Fee on September 12, 2024, ending the pricing debacle that had broken developer trust, and raised subscription prices instead (Pro up 8%, Enterprise up 25%). He reset the portfolio: the ironSource Ads Network was sunset on April 30, 2026, and Supersonic was sold to Tripledot Studios in a deal that closed August 4, 2026. Adjusted EBITDA margin expanded for six consecutive quarters, from 19% in Q1 2025 to 29% in Q2 2026, guided to about 33% in Q3 2026. Unity beat its own guidance two quarters running: Q1 2026 actual revenue of $508.238 million against $480-490 million of guidance, and Q2 2026 revenue of $546.468 million with adjusted EBITDA of $160.191 million against $505-515 million and $130-135 million of guidance. He pulled GAAP profitability forward to Q3 2026, one quarter ahead of the prior Q4 target, and flipped the balance sheet to net cash, with the 2026 converts to be repaid in November 2026.
The history must be stated factually. The per-install Runtime Fee announced September 12, 2023 triggered broad developer backlash, and John Riccitiello retired in October 2023. That episode is exactly why the turnaround had to start with trust, and it did. Bromberg called Q2 2026 "arguably the best quarter in Unity's history as a public company."
What would change the grade: upgrade path is a full downcycle managed cleanly; he has about 28 months in the role and no full downcycle yet. Downgrade on a guidance cut, renewed trust damage with developers, or a stall in the margin expansion.
Partial
The question: can competitors take the customers?
The moat is real in mobile game engines. Switching engines mid-development is prohibitively expensive, so the installed base is sticky. Unity has the largest mobile developer ecosystem, a mature asset store, and runtime data from roughly 3 billion monthly consumers that began feeding Vector's live models late in Q2 2026. Management described that runtime data as a deep, sustainable competitive advantage, and only Unity can collect it at that scale. A third-party SDK-presence study found Unity Ads in 75% of sampled Android apps globally and 98% of sampled Android gaming apps: SDK presence is not revenue share, but it shows the breadth of the runtime footprint.
But the moat has visible gaps. Unreal dominates high-end and PC-console development, and Unity does not win there. In ad mediation, AppLovin MAX dwarfs LevelPlay: a 2025 GameBiz sample of 175 top-downloaded ad-monetized games found MAX in 73.1% and LevelPlay in 5.7%, a sample-specific but directionally clear read on the pecking order. AppLovin is the key Grow competitor, doing $1.924 billion of revenue in Q2 2026 at an 84% adjusted EBITDA margin. And AI-native creation tools are emerging as a new vector of disruption: if creation moves up the stack, the engine layer has to prove it still owns the workflow.
What would change the grade: upgrade on the Vector data moat deepening (more runtime coverage feeding the models) and broad Unity 7 adoption (beta December 2026, full release targeted Q1 2027, with no breaking changes from Unity 6). Downgrade on developer churn toward Unreal, Godot, Roblox-native tools, or AI-native creation platforms, or further share loss to AppLovin in mediation.
Pass
The question: can the company survive a bad two years without help?
Yes. Unrestricted cash was $2.351987 billion at June 30, 2026, against a convertible-note carrying value of $2.237402 billion: about $115 million of net cash, and the CFO said on the Q2 2026 call that Unity had "flipped from a net debt position to a net cash position this quarter." Note the arithmetic carefully: $2.36 billion is gross cash, not net cash. Net cash on management's stated basis is about $115 million. The balance sheet also shows the discipline of the turnaround: the company issued $690 million of 2030 converts in the first half of 2025, bought $44.436 million of capped calls, and repaid $641.691 million of notes, actively managing maturities rather than ignoring them.
Watch item: the convert stack still matters. The 2027 notes are $1.0 billion at a 2% coupon with a $48.89 conversion price; the 2030 notes are $690 million at a 0% coupon with a $36.15 conversion price. If the stock runs through those prices, dilution follows. The 2026 notes (about $558 million at 0%, conversion price $308.72, effectively a cash repayment) are due in November 2026: the repayment must happen from cash, not from new financing.
What would change the grade: downgrade on a financing-dependent November 2026 repayment, a return to debt-funded operations, or renewed balance-sheet leverage. A clean November repayment with the cash pile intact moves the pillar toward the ceiling.
Pass
The question: does the business generate real cash, repeatedly?
Yes. Trailing-twelve-month free cash flow is about $533 million, and the trend is the point. Q2 2026 free cash flow was $201.959 million, up 59% year over year from $126.650 million, on operating cash flow of $205.622 million. Full-year 2025 free cash flow was just over $400 million, up 41%, with free-cash-flow conversion of about 99% of adjusted EBITDA. The first half of 2026 produced $268.416 million of free cash flow on $1.054706 billion of revenue. This is no longer a burn story: it is a cash-compounding story, driven by the mix shift toward strategic revenue (up 38% in Q2 2026 to $486.411 million) and adjusted S&M and G&A falling in absolute dollar terms while revenue grew 24%.
The GAAP picture is still catching up: Q2 2026 net loss was $22.672 million, and the first half carried a $279 million non-cash impairment tied to the ironSource Ads Network shutdown and the Supersonic divestiture. But the cash is what the pillar grades, and the cash is real and repeated. Stock-based compensation, the old dilution engine, fell from 32.9% of revenue in FY2024 to 20.8% in FY2025 to about 14.6% in Q2 2026, with stock comp expense down 25% year over year, its lowest level ever.
What would change the grade: downgrade on free cash flow turning negative again, a stall in Vector and strategic Grow, or opex reaccelerating. GAAP profitability landing in Q3 2026, one quarter early, moves the pillar toward the ceiling.
| Pillar | Current | Upgrade path | Downgrade trigger |
|---|---|---|---|
| Sunrise sector | Pass | Already at ceiling. | A sustained collapse in mobile game demand or the mobile ad market; a developer exodus toward AI-native creation platforms. |
| Leadership | Pass | A full downcycle managed cleanly moves it toward the ceiling. | A guidance cut, renewed trust damage with developers, or a stall in the margin expansion. |
| Moat | Partial | Vector data moat deepening and broad Unity 7 adoption. | Developer churn toward Unreal, Godot, Roblox-native tools, or AI-native platforms; further mediation share loss to AppLovin. |
| Iron fortress | Pass | A clean November 2026 convert repayment from cash moves it toward the ceiling. | A financing-dependent November repayment, renewed balance-sheet leverage, or debt-funded operations. |
| Free cash flow | Pass | GAAP profitability landing in Q3 2026, one quarter early, moves it toward the ceiling. | FCF turning negative again; Vector or strategic Grow stalling; opex reaccelerating. |
How the group is priced. All figures at the September 23, 2026 close. Unity at $44.10 carries about a $19.4 billion market cap on about $2.028 billion of trailing revenue (P/S 9.6x) and about $532 million of trailing adjusted EBITDA, for an EV/EBITDA of about 36.3x on trailing numbers. The 36x multiple prices in the margin expansion: Q3 2026 guidance implies about a 33% adjusted EBITDA margin, and the bridge points toward 35-39% in the outer years.
| Company | Price | Market cap | TTM revenue | P/S | TTM EBITDA | EV/EBITDA |
|---|---|---|---|---|---|---|
| Unity | $44.10 | ~$19.4B | ~$2.028B | 9.6x | ~$532M adj | ~36.3x |
| AppLovin | $315.17 | ~$110B | ~$6.831B | 16.1x | ~$5.725B adj | ~19.3x |
| Roblox | $48.90 | ~$34.9B | ~$5.69B | 6.1x | negative | n/a |
| Autodesk | $217.41 | ~$45.39B | ~$7.79B | 5.8x | ~$2.0-2.3B | ~20-23x |
| PTC | $140.84 | ~$14.7B | ~$2.96B | 5.0x | ~$1.25B | ~12.7x |
AppLovin is the premium comp: higher multiples on structurally higher margins (84% Q2 2026 adjusted EBITDA margin). Autodesk and PTC are the mature creator-software comps, growing slower at lower multiples. Roblox trades on revenue growth with negative EBITDA. Unity sits between the two worlds: a growth-priced multiple on a margin-expansion story that is still proving itself.
Context. $44.10 sits 78% below Unity's all-time high of $198.69, with year-to-date total return of about -0.2%. Over six months the stock ran 96% to 136% depending on the vendor window, outperforming the S&P 500 by about 93.65% and trading about 29% above its 200-day moving average. After the Q2 beat, BofA and Deutsche Bank upgraded to Buy with $50 targets, Macquarie, Morgan Stanley, and Jefferies set $50, Citigroup $48, HSBC $47.10, and Piper Sandler $55, with the average target near $49. The market has repriced the turnaround; the comps show it is now priced on execution, not on hope.
The Meta Connect read-through, in full. Meta's September 23 opening keynote announced Meta VR Glasses ($1,299.99, spring 2027, Quest catalog compatible), camera-free Ray-Ban Meta Audio at $349, and Ray-Ban Meta Gen 3, with no new Quest-branded headset. A Meta session for September 24 covers "Building with the latest SDK updates for game engines: Your guide to building for VR with Unity, Unreal, or native engines." No Unity-specific partnership, new Unity workflow, enterprise XR push, or digital-twin initiative was disclosed in the keynote coverage, and detailed September 24 developer-session announcements were still pending at the time of this research. Net assessment: modestly positive to neutral. New hardware that runs the Quest catalog is a tailwind for Unity-built XR apps; Meta's Horizon-native tooling is a competitor only for low-complexity creators. Not a standalone catalyst.
Method. Each pillar is graded on evidence through September 23, 2026; financials come from company releases and SEC filings; figures marked as estimates in the research are labeled as estimates.
| Claim | Source |
|---|---|
| Q2 2026: $546.468M revenue (+24%); strategic $486.411M (+38%); adj EBITDA $160.191M (29%); FCF $201.959M (+59%); net loss $22.672M; adj EPS $0.28 vs ($0.11). Q3 guidance: strategic $540-550M (+44-47%); adj EBITDA $185-190M (+69-74%, ~33%) | Unity Q2 2026 earnings release, August 6, 2026 (Business Wire) |
| Unrestricted cash $2.351987B; convert carrying value $2.237402B ($557.173M current + $1.680229B long term); "flipped from a net debt position to a net cash position"; 2026 converts to be repaid November 2026 | Unity Q2 2026 release; CFO Jarrod Yahes per MarketBeat transcript, August 6, 2026 |
| Convert principal: 2026 notes $557.724M (0%, $308.72); 2027 notes $1.0B (2%, $48.89); 2030 notes $690M (0%, $36.15); capped-call caps $343.02 / $47.74 | Unity Q1 2026 10-Q, March 31, 2026 |
| GAAP profitability expected Q3 2026, pulled forward from Q4 target. Beat guidance twice: Q1 $508.238M vs $480-490M; Q2 $546.468M / $160.191M vs $505-515M / $130-135M | Motley Fool call summary, August 2026; Unity Q1 and Q2 2026 releases |
| Matthew Bromberg CEO since May 15, 2024, succeeding interim CEO Jim Whitehurst; Zynga COO 2016-2022; EA SVP Strategy/Operations EA Mobile and GM BioWare; MLG CEO; I'mOK founder/CEO; AOL SVP/GM 1999-2005; Cornell B.A., Harvard J.D. | Unity leadership page, unity.com; WN Hub, May 2024 |
| Runtime Fee announced September 12, 2023; developer backlash; John Riccitiello retired October 2023. Fee canceled September 12, 2024; Pro up 8%, Enterprise up 25% | Polygon, October 10, 2023; Game World Observer, September 2024 |
| ironSource Ads Network sunset April 30, 2026; Supersonic sold to Tripledot Studios, closed August 4, 2026; Q1 2026 $279M non-cash impairments | Unity Q2 2026 and Q1 2026 releases; MarketBeat transcript, August 6, 2026 |
| Vector: +23% QoQ vs 12-13% expectation; $1B run rate two quarters ahead of plan; Day 28 spend ~3x from Q1; Day 28 ROAS adoption >25% of ad base; ~3B monthly consumers' runtime data feeding live models late Q2 | Motley Fool Q2 2026 call takeaways, August 2026 |
| GameBiz 2025 sample: MAX 73.1% vs LevelPlay 5.7% in 175 top-downloaded ad-monetized games (sample-specific). Unity Ads in 75% of sampled Android apps, 98% of sampled Android gaming apps (SDK presence, not revenue share) | GameBiz Newsletter, May 2025; Pixalate Q3 2025 study via GlobeNewswire, Nov 6, 2025 |
| Unity 7 announced Unite Seoul, July 21, 2026; beta December 2026, full release targeted Q1 2027; no breaking changes from Unity 6 | GamesBeat, July 2026; Unity 7 roadmap PDF (q4cdn) |
| FY2025: revenue $1.849648B, gross margin 74.2%; adj EBITDA margin 22% (~$407M); FCF just over $400M (+41%), ~99% of adj EBITDA. SBC: FY2023 $649M (29.7%); FY2024 $596.249M (32.9%); FY2025 $385.214M (20.8%); Q2 2026 $79.948M (14.6%) | Unity FY2025 release, February 11, 2026; Unity 10-K filings; Q2 2026 release |
| Meta Connect 2026: Sep 23-24, Menlo Park; keynote Sep 23 4pm PT; Meta VR Glasses $1,299.99, spring 2027, Quest catalog compatible; Ray-Ban Meta Audio $349; no new Quest headset in keynote | TechCrunch, September 23, 2026; Gizmodo live blog, September 23, 2026 |
| Analysts: BofA, Deutsche Bank upgraded to Buy ($50); Macquarie, Morgan Stanley, Jefferies $50; Citi $48; HSBC $47.10; Piper Sandler $55; average near $49. $44.10 is 78% below the $198.69 all-time high; YTD about -0.2%; 6-month +96% to +136%. Reference: $44.10; market cap about $19.4B; 439.987M shares at June 30, 2026 | Tickeron and MarketBeat, Sep 2026; financecharts.com and upstox, Sep 23, 2026; Unity Q2 2026 release |
Disclaimer: independent research for informational purposes only, not investment advice. Investing involves risk. This analysis discusses the ticker only and contains no portfolio or position information.