Contrary to the hype during its IPO, Reddit's AI data licensing revenue is a tiny fraction of its total business. The company's advertising segment is growing much faster, suggesting the AI narrative was overplayed to attract investors, while the core business remains traditional digital advertising.
Related Insights
Despite the hype around large language models, they represent a minority of AI compute usage at a tech giant like Meta. The vast majority of AI capital expenditure is dedicated to other tasks like content recommendation and ad placement, highlighting the continued importance of diverse, non-LLM AI systems in large-scale operations.
While the market seeks revenue from novel AI products, the first significant financial impact has come from using AI to enhance existing digital advertising engines. This has driven unexpected growth for companies like Meta and Google, proving AI's immediate value beyond generative applications.
Contrary to popular belief, advertising is the smallest part of Stack Overflow's business (20% of revenue). The company's financial stability comes from its enterprise SaaS product for internal knowledge management and a burgeoning data licensing business selling its curated Q&A data to AI labs.
Reddit's Average Revenue Per User (ARPU) is significantly lower than its social media peers. While this indicates a failure to capitalize on its massive, engaged user base, it also represents the company's single largest opportunity for future growth if it can successfully close this monetization gap.
Apple is shifting its podcast product by introducing an advertising platform. This move mirrors the strategies of Amazon and OpenAI, indicating that even for hardware and software giants, high-margin advertising revenue is becoming the most critical and dependable lever for future growth when primary product innovation slows.
While the market awaits new AI-native products from Meta, its real AI success is in its core business. A 9% CPM increase in a weak economy indicates its ad-serving algorithm's effectiveness improved by double digits in a single quarter, a massive financial win.
AI conversations capture high-intent moments, allowing ads to target active decision-making rather than passive attention-grabbing like social media. This fundamental difference could lead to significantly higher average revenue per user (ARPU), making social media's ad performance a floor, not a ceiling for AI platforms.
The long-term monetization model for consumer LLMs is unlikely to be paid subscriptions. Instead, the market will probably shift toward free, ad- and commerce-supported models. OpenAI's challenge is to build these complex new revenue streams before its current subscription growth inevitably slows.
Platforms with real human-generated content have a dual revenue opportunity in the AI era. They can serve ads to their human user base while also selling high-value data licenses to companies like Google that need authentic, up-to-date information to train their large language models.
While AI labs could build competing enterprise apps, the required effort (sales teams, customizations) is massive. For a multi-billion dollar company, the resulting revenue is a rounding error, making it an illogical distraction from their core model-building business.
