A-Portfolio Daily - LFTO: Cortex is turning ad data into earnings
Date: July 28, 2026
Ticker: NASDAQ: LFTO
Company: Liftoff Mobile, Inc.
Verdict: A-PORTFOLIO CANDIDATE - 83/100
Reference price: $22.81, July 27 close [5]
Position framework: No position before the August 12 report. If the report validates the thesis, cap the first tranche at 0.5% of portfolio value and total size at 1.0% until two clean public quarters are filed.
Bias / Risk / Action
- Bias: Constructive. Q1 revenue grew 37.1%, GAAP operating margin reached 41.8%, and operating cash flow rose 69.4% year over year while the stock sits 0.8% below its $23 IPO price [1][5].
- Risk: Liftoff still has an estimated $1.15 billion of post-IPO net debt, Blackstone retains roughly half the vote, Apple and Google can change the rules under the business, and a financial-control weakness remains open [1][2].
- Action: Stage diligence, not a blind buy. Read the August 12 earnings release and 10-Q first. Demand durable growth, clean post-IPO debt disclosure, and proof that Cortex-driven gains survive outside the prospectus window [4].
Why this matters, in plain English
Liftoff helps app makers find paying users and helps them sell advertising inside their apps. Its new Cortex system uses artificial intelligence to place better ads, and the early numbers show faster growth and sharply better profit. If those gains hold, Liftoff can take a bigger cut of the money flowing through mobile apps while its existing software footprint becomes harder to replace.
Why this name now
Liftoff listed on June 4 after selling 21.85 million shares at $23, including the underwriters' full option [2]. The stock closed July 27 at $22.81, below the offer price and 19.8% below its first-day $28.45 close [5][9]. The hype premium is gone before the first public earnings report.
The operating inflection is not cosmetic. Q1 2026 GAAP revenue was $205.6 million, up 37.1%. GAAP operating income was $85.9 million, up 91.0%, and GAAP net income rose to $49.3 million from $3.9 million [1]. Core Advertising revenue, the current platform rather than retired products, reached $204.7 million, up 37.4% [1].
Cortex is the proposed cause. Liftoff says the neural-network system runs four times more tests and processes roughly 21 times more data than its prior models [1]. Customer behavior supports the claim: trailing Core Advertising net-dollar retention was 130%, meaning the same customer group spent 30% more year over year; top-100 advertiser retention was 97%, and top-100 publisher retention was 100% [1].
The August 12 Q2 report is the first clean test outside the IPO prospectus. Management has announced the date but has not issued public 2026 guidance [4]. William Blair estimates $487.3 million of 2026 EBITDA and $588.1 million in 2027; those are analyst forecasts, not management targets or GAAP results [6].
Score: 83/100
| Factor | Score | Kill-level read |
|---|---|---|
| Category leadership / moat | 17/20 | One platform connects advertiser demand with ad inventory across more than 167,000 apps and 1.4 billion daily users. AppLovin, Google, Meta, and Unity are larger threats [1][8]. |
| Growth inflection | 20/20 | Q1 revenue +37.1%, Core Advertising +37.4%, GAAP operating income +91.0%, and ten consecutive quarters of Core Advertising growth [1]. |
| Financial quality / cash durability | 15/20 | Q1 GAAP profitability and $87.2M of operating cash are strong. Debt, capitalized software, and one quarter of proof cap the score [1]. |
| Valuation asymmetry | 12/15 | Roughly 10.3x William Blair's 2026 EBITDA estimate and 18.4x annualized Q1 pro forma diluted EPS. The base case is thin, but the bull case is meaningful [1][6]. |
| Management / execution | 7/10 | Cortex migration and margin expansion are excellent. Blackstone control and an unremediated IT-control weakness earn a haircut [1]. |
| Catalysts | 8/10 | August 12 results, the first 10-Q, post-IPO deleveraging, non-gaming growth, and eventual lockup supply [1][4]. |
| Entry quality | 4/5 | Price is below the IPO and far below the first-day close, but the first public quarter can reset the stock violently [2][5]. |
| Total | 83/100 | A-PORTFOLIO CANDIDATE - EVIDENCE FIRST |
Numbers that matter
GAAP results
- 2025 revenue: $685.7 million, up 32.1% from $519.3 million [1].
- 2025 operating income: $166.8 million, a 24.3% operating margin, versus $92.9 million in 2024 [1].
- 2025 net loss: $23.1 million, improved from a $48.2 million loss. Interest expense was $127.0 million [1].
- 2025 operating cash flow: $163.4 million, up 65.4% [1].
- Q1 2026 revenue: $205.6 million, up 37.1% [1].
- Q1 2026 operating income: $85.9 million, a 41.8% margin, versus $45.0 million [1].
- Q1 2026 net income: $49.3 million, a 24.0% margin, versus $3.9 million [1]. The quarter included net favorable revaluation items; do not annualize the headline without adjustment.
- Q1 2026 operating cash flow: $87.2 million, up 69.4% from $51.5 million [1].
Management-defined metrics
- 2025 adjusted EBITDA: $374.4 million, up 46.2%, with a 55% margin [1]. This is non-GAAP and excludes interest, taxes, depreciation and amortization, stock compensation, transaction costs, and other items.
- Q1 adjusted EBITDA: $120.1 million, up 54.2%, with a 58% margin [1].
- Core Advertising net-dollar retention: 130% for the trailing year ended March 31 [1]. This is an operating metric, not booked future revenue.
- Large customers: 384 customers generated more than $100,000 of trailing Core Advertising revenue, up from 328 a year earlier; this group supplied 98% of Core Advertising revenue [1].
- Non-gaming mix: Slightly more than half of Q1 advertiser revenue came from outside gaming [1].
Balance sheet and valuation at $22.81
The underwriters exercised their full option, bringing post-IPO shares to approximately 169.33 million [1][2]. The following post-IPO cash and debt figures are estimates based on the prospectus use-of-proceeds bridge, not a filed June 30 balance sheet.
- Equity value: approximately $3.86 billion.
- Estimated debt after IPO repayment: approximately $1.40 billion.
- Estimated cash after the offering: approximately $248.5 million.
- Estimated net debt: approximately $1.15 billion.
- Estimated enterprise value: approximately $5.02 billion.
- EV / 2025 revenue: 7.31x.
- EV / 2025 adjusted EBITDA: 13.40x.
- EV / William Blair 2026 EBITDA estimate: 10.29x [6].
- Price / annualized Q1 pro forma diluted EPS: 18.40x. This is a rough run-rate, not guidance.
- The IPO repayment reduced prospectus debt by an estimated 22.8%, but leverage remains material.
Bull / base / bear case
These are valuation markers, not forecasts. Each applies an enterprise-value multiple to William Blair's $487.3 million 2026 EBITDA estimate, subtracts estimated post-IPO net debt of $1.154 billion, and divides by 169.33 million shares [1][2][6]. The EBITDA input is an analyst estimate and non-GAAP.
- Bull - $33.48: Cortex keeps growth above 25%, Q2 validates retention and margins, and the market pays 14x estimated 2026 EBITDA. That implies 46.8% upside.
- Base - $24.84: Growth decelerates but stays healthy, margins hold, and the stock receives 11x. That implies 8.9% upside.
- Bear - $16.21: Growth falls toward the market rate, privacy changes or competition weaken ad performance, and the multiple compresses to 8x. That implies 28.9% downside.
The base case is not rich enough to buy before Q2. Candidate status means the company deserves staged attention, not that today's quote deserves automatic capital.
What the market may be missing
This is an earnings inflection, not an AI label. Liftoff's GAAP operating margin nearly doubled year over year in Q1 while revenue accelerated. The critical question is whether Cortex created durable ad-performance gains or whether Q1 was an unusually favorable budget and cost window [1].
The supply-and-demand loop may be real. Liftoff operates both the advertiser-buying system and the publisher-selling system. More advertiser spending attracts publishers; more publisher inventory gives Cortex more data and places to buy ads. That can compound, but advertisers can move budgets quickly when returns weaken [1][8].
Non-gaming data is the wedge. AppLovin is the stronger incumbent in mobile gaming. Liftoff's pitch is broader: finance, travel, dating, e-commerce, social, and other apps supply different purchase signals. If that mix lets Cortex outperform outside games, Liftoff can grow without taking AppLovin head-on in its strongest lane [7][8].
The balance sheet is better, not clean. Full exercise of the IPO option likely cut debt to roughly $1.40 billion, but 2025 interest expense was still $127.0 million and Q1 interest was $35.0 million [1][2]. Q2 must show the actual post-IPO debt, cash, rate, and interest run-rate.
Social evidence did not clear the bar. The 30-day sweep found no usable direct X discussion and mostly off-topic Reddit and short-video results [10]. Primary filings and the first public earnings report outrank the crowd here.
Kill criteria
Kill the thesis if any one major criterion or any two operating criteria hit:
1. Major: Core Advertising revenue growth falls below 15% for two consecutive quarters without a deliberate exit from low-return business.
2. Major: A material Apple or Google policy change measurably impairs targeting, attribution, or access to app inventory, and Liftoff cannot offset it within two quarters.
3. Core Advertising net-dollar retention falls below 110% or top-100 advertiser retention falls below 90%.
4. GAAP operating margin falls below 25% for two consecutive quarters while revenue is still growing.
5. The count of customers generating more than $100,000 declines for two consecutive quarters.
6. Net debt fails to fall below $1.0 billion by year-end 2027 despite positive operating cash flow.
7. Interest expense remains above 30% of GAAP operating income after the IPO repayment is fully reflected.
8. The disclosed IT-control material weakness remains unremediated through the 2027 Form 10-K or causes a restatement.
9. Blackstone-directed transactions or secondary sales damage minority-holder economics without a clear company benefit.
10. Non-gaming growth fails to outpace the rest of the platform, invalidating the claimed differentiation from gaming-heavy rivals.
Entry discipline
- Before August 12: No position. The first public print matters more than a 0.8% discount to the IPO price.
- Valid starter after a clean report: At or below $23.25, maximum 0.5% of portfolio value, only if revenue growth remains above 25%, Core Advertising retention remains above 120%, and post-IPO net debt is no worse than the prospectus bridge.
- $23.25-$27: Watch. The base-case return is too thin without higher earnings proof.
- Above $27: No chase. Require either a higher earnings base or a faster debt decline.
- Below $20: Re-underwrite first. A lower price is not an edge if caused by retention decay, platform policy damage, or an accounting-control problem.
- Maximum total size before two clean public quarters: 1.0%.
The next proof points
1. August 12 Q2 GAAP revenue, operating income, net income, and operating cash flow [4].
2. Core Advertising revenue growth and net-dollar retention.
3. Customer count above $100,000 and top-100 advertiser/publisher retention.
4. Actual June 30 cash, debt, net leverage, and interest rate after the IPO repayment.
5. GAAP-to-adjusted-EBITDA reconciliation after becoming public.
6. Non-gaming revenue mix and whether e-commerce is a measured growth driver.
7. Progress remediating the IT general-control weakness.
8. Lockup-expiration terms and any Blackstone or General Atlantic secondary supply.
Sources - as of July 28, 2026
1. Liftoff final IPO prospectus, June 3, 2026: https://www.sec.gov/Archives/edgar/data/1850351/000119312526257410/iron_424b4_round_2.htm
2. Liftoff IPO closing release and full underwriter-option exercise, June 5, 2026: https://www.globenewswire.com/news-release/2026/06/05/3307603/0/en/Liftoff-Announces-Closing-of-Initial-Public-Offering-and-Full-Exercise-of-Underwriters-Option-to-Purchase-Additional-Shares.html
3. Liftoff investor relations homepage and current company positioning: https://investors.liftoff.ai/
4. Liftoff Q2 2026 earnings-date announcement, July 22, 2026: https://www.globenewswire.com/news-release/2026/07/22/3331715/0/en/Liftoff-to-Announce-Second-Quarter-2026-Results.html
5. Exa Markets LFTO July 27, 2026 historical quote: https://exa.ai/library/markets/stock/LFTO?date=2026-07-27
6. William Blair LFTO research initiation and 2026-2027 EBITDA estimates: https://www.williamblair.com/News/Liftoff-Mobile-Inc-Research-Initiation
7. Digiday CEO interview on Cortex, data breadth, and AppLovin comparisons: https://digiday.com/media-buying/liftoff-ceo-on-the-ipo-rebound-applovin-comparisons-and-why-mobile-apps-remain-an-ai-growth-story/
8. AdExchanger IPO-day interview on Liftoff's competitive position: https://www.adexchanger.com/mobile/ipo-day-qa-liftoffs-coo-on-catching-a-better-ipo-window/
9. Reuters IPO pricing report: https://finance.yahoo.com/markets/stocks/articles/blackstone-backed-liftoff-mobile-raises-004754567.html
10. Last-30-days community pulse archive: C:\Users\derek\Documents\Last30Days\liftoff-mobile-lfto-stock-ipo-raw-a-portfolio.md
11. Renaissance Capital 2026 recent IPO screen used for candidate comparison: https://www.renaissancecapital.com/IPO-Center/Pricings
12. INNIO final prospectus, screened but not selected: https://www.sec.gov/Archives/edgar/data/2109150/000119312526256992/liberty_-_424b4.htm
13. Jersey Mike's preliminary prospectus, screened but not selected because it was not yet listed as of the research cutoff: https://www.sec.gov/Archives/edgar/data/2127043/000119312526293830/ck0002127043-20260702.htm
Calculation note: All calculations used decimal arithmetic. Equity value = $22.81 times 169.330527 million shares. Estimated full-option post-IPO debt = $1.816083 billion prospectus debt less $413.8 million of estimated repayment. Estimated cash adds the approximately $5.1 million residual benefit from the full option to the $243.405 million no-option pro forma cash figure. Enterprise value = equity value plus estimated net debt. Scenario equity value = selected multiple times the $487.3 million William Blair 2026 EBITDA estimate, less estimated net debt, divided by shares.
Research only. Derek makes the allocation decision.