AppLovin Corporation (APP) Earnings

Management highlights

### Core Business Performance - Q2 2026 delivered $1.92 billion in total revenue (up 53% year-over-year) and $1.61 billion in adjusted EBITDA (up 58% year-over-year), with adjusted EBITDA margins expanding 300 basis points year-over-year to ~83%. Results fell just below the midpoint of prior guidance, which management attributes purely to timing of model improvements, not weakening demand or a changed competitive landscape. - Free cash flow for the quarter was $863 million, with lower-than-normal conversion due to timing of international cash tax and interest payments; conversion is expected to improve in Q3 and normalize to ~75% of adjusted EBITDA for full year 2026. - The SEC inquiry that was previously disclosed has concluded with no recommended action. ### Strategic Priorities - Near-term focus: Continued improvement of core ad targeting models (the primary driver of growth), advancing platform architecture to support more complex, compute-heavy models, improving creative tools and ad formats, and onboarding new high-quality advertisers via strategic partnerships. - Long-term strategy: Maintain the core gaming business while scaling the consumer vertical to add long-term revenue runway, with a target of 30% annual compound growth for the overall business. - The self-service platform Applovin Ads Manager launched publicly to the market in Q2; the company is prioritizing mid-market advertisers first, with plans to expand to long-tail advertisers gradually as data scale and product capabilities improve. ### Capital Structure - Ended Q2 with $3.05 billion cash and $3.7 billion total debt, for a net leverage ratio of ~0.1x trailing 12-month adjusted EBITDA, well below the long-term target of ~1x. - Repurchased 1.14 million shares for $551 million in Q2, with ~$1.8 billion remaining in the repurchase authorization. The slower buyback pace in Q2 reflects temporary lower free cash flow, not a change in strategy.

Guidance

- **Q3 2026 Revenue Guidance**: Management expects revenue of $2.055 billion to $2.085 billion, representing 46% to 48% year-over-year growth and 7% to 8% sequential growth. The guidance already incorporates the model improvements that launched just after Q2 end. - **Q3 2026 Adjusted EBITDA Guidance**: Management expects adjusted EBITDA of $1.71 billion to $1.74 billion, representing 48% to 50% year-over-year growth, with an adjusted EBITDA margin of ~83%. The guidance accounts for higher model training and inference compute costs tied to the new model launch. - Guidance does not include any assumptions for additional model improvements that have not yet been deployed. - Long-term, management expects adjusted EBITDA margins to remain in the low 80% range, with short-term variability allowed as the company invests in revenue-driving compute spend. The long-term target for incremental compute spend remains 10 cents per incremental dollar of revenue, with no expected change to this cadence.

Segment performance

1. **Gaming Segment**: Gaming remains the majority of the company's total revenue. Q2 2026 total company revenue was $1.92 billion (up 53% year-over-year, 4% sequentially), with the core gaming business as a primary driver of this growth. Max publisher earnings grew double digits quarter-over-quarter, and App Loving's share of publisher waterfalls remained consistent. The segment missed Q2 guidance due to slower-than-expected model improvement, with the scheduled model upgrade launching just after quarter end. 2. **Consumer (non-gaming, primarily e-commerce) Segment**: The consumer vertical continued scaling rapidly in Q2 2026. Advertiser spend hit a new record, coming in 28% above the prior seasonal peak (Q4 2025), an unusually strong result for a seasonally slow quarter. While the consumer business is not yet large enough to offset quarterly fluctuations from the gaming segment, it is growing quickly and adds long-term revenue runway for the company.

Risks & headwinds

- Quarterly results can be volatile due to the timing of model development and deployment: testing of model improvements does not guarantee uplift, and delayed material improvements can cause quarterly results to miss guidance, even if the long-term growth trajectory remains intact. - Scaling model improvements requires increasing compute investment, which can create short-term margin pressure before the associated revenue upside is realized. - Expanding into the consumer advertiser segment requires overcoming existing brand awareness (most advertisers currently associate App Loving only with gaming) and product capability gaps (most notably automated high-quality creative generation) that will take time to resolve. - Long-term growth of the consumer segment depends on gradual compounding of customer acquisition, data scale, and product improvement, which will unfold over years rather than months, with no guarantee of hitting projected growth targets.

Analyst Q&A

• Q: Why did App Loving miss Q2 guidance, and what is the current post-quarter performance of model improvements? How will margin profile change as compute costs scale? /

  A: Q2 guidance was missed because the quarter saw less material model uplift than the company’s typical cadence, with the next large improvement launching just after quarter end. Model development relies on iterative A/B testing for uplift, so periods of lower upside are unavoidable. Q3 guidance already incorporates the launched improvements and higher compute costs. Short-term margin fluctuations are expected, but long-term margins will remain in the low 80% range. All compute investment is only approved if it is tied to clear incremental revenue upside, so short-term cost increases will drive longer-term growth. (379 characters)

• Q: What is the strategy for partnerships to acquire new consumer advertisers, and what do these partnerships look like? /

  A: The company is partnering with established third-party companies (like leading e-commerce analytics providers such as Triple Whale) that already work with the mid-market e-commerce advertisers App Loving wants to onboard. This targeted approach brings higher-value, higher-spend advertisers to the platform compared to broad marketing that would attract mostly small low-spend long-tail advertisers. The partnerships create win-win-win outcomes: partners gain an additional high-quality ad platform for their customers, customers get access to a new high-performing acquisition channel, and App Loving gets valuable targeted new customers. (457 characters)

• Q: How is the public launch of the self-service Applovin Ads Manager performing so far, and what is the long-term customer scaling trajectory? /

  A: The launch is meeting the company’s expectations, and growth is tracking as planned. Currently, the business is concentrated with a smaller number of larger mid-market advertisers rather than a large number of small long-tail advertisers, which matches the company’s deliberate go-to-market strategy. It will take years of compounding customer growth, product improvement, and data collection to build widespread adoption similar to large established ad platforms, just as it took over a decade to reach full penetration in the mobile gaming category. As more mid-market advertisers onboard and see success, data improves, and the platform will gradually expand to broader groups of advertisers over time. (483 characters)

• Q: What is the current status of creative generation tools for self-service advertisers, and what are the key product gaps that still need resolution? /

  A: Creative generation remains the biggest hurdle for advertiser onboarding, as App Loving’s most effective ad units require a 30-60 second video plus an interactive end card. While the platform can now auto-generate interactive end cards at high efficiency, it can not yet consistently produce high-quality long-form video for advertisers out of the box. Until this capability is built or alternative ad templates that do not require video are developed, one-click campaign creation for new self-service advertisers will not be possible, limiting onboarding conversion especially for small long-tail advertisers that do not have existing creative resources. (462 characters)