Runway launches an AI ad platform built on its own campaign workflow

Runway launched Runway Ads on September 30th, an AI ad platform that generates video and image ads, publishes approved versions to Meta, Google and TikTok, reads back performance and produces the next round based on what earns spend. Runway says its internal use of the system since July 2026 raised weekly ad volume from 77 to roughly 900, doubled return on ad spend, lifted conversions about 34% and cut cost per subscriber 41%, and the company is piloting the product with select enterprise partners. Co-CEO and co-founder Anastasis Germanidis said Runway built the system because the company needed it itself, and Runway's product page attributes more than $100 million in additional annual recurring revenue to performance marketing, though the announcement gives no pricing, absolute spend or customer-level results.

Runway launches an AI ad platform built on its own campaign workflow Runway says its internal system lifted weekly output to roughly 900 ads; it is piloting the product with select enterprise partners. By Ryan Merket https://runtimewire.com/author/ryan-merket · Published Primary source: Runway https://runway.com/news/company-news/introducing-runway-ads Why it matters Runway is packaging its own paid-marketing workflow as a product, betting that generative video can earn a role in campaign operations when it connects creative production to publishing and measured results. Runway launched Runway Ads on September 30th, describing it as a system that generates video and image ads, publishes approved versions to Meta, Google and TikTok, reads back performance and creates the next round based on what earns spend. In its announcement https://runway.com/news/company-news/introducing-runway-ads?ref=runtimewire , co-CEO and co-founder Anastasis Germanidis https://runway.com/news/new-leaders-to-support-next-phase-of-growth?ref=runtimewire said Runway built it because the company needed it itself. That origin gives Runway Ads a more concrete pitch than a general-purpose AI creative tool: it is meant to join the production and measurement steps that marketing teams often manage across separate systems. Runway says it is piloting the product with select enterprise partners. Its early-access page https://runway.com/product/ads?ref=runtimewire describes the workflow: generate, approve, publish and learn. A growth tool becomes the product Germanidis has led Runway's research and product vision; in February, Runway formalized that role by naming him co-CEO alongside Cristobal Valenzuela. The founders started Runway around creative tools, and Germanidis's explanation for Ads follows that practical tradition: Runway says it built the system to handle a problem in its own paid marketing, rather than starting with a standalone ad-tech thesis. Runway's account of its internal results is striking, and self-reported. Runway says that since July 2026 it increased weekly ad volume from 77 to roughly 900, doubled return on ad spend, raised conversions by about 34% and reduced cost per subscriber by 41%, while click-through rate held steady. Runway says spending also increased, but the announcement gives no absolute spend, subscriber count, campaign mix or methodology for those comparisons. The figures describe Runway's own program; they do not establish what a customer should expect. Runway's product page https://runway.com/product/ads?ref=runtimewire also attributes more than $100 million in additional annual recurring revenue to performance marketing. That is a company claim, and the page does not explain the measurement period or how it connects the figure to the Ads product. The launch announcement provides no product pricing or customer-level results. Runway Ads is designed to make creative volume less labor-intensive. A customer connects an ad account and supplies brand guidelines, past ads and product imagery. The system generates variants, adapts them for different placements, and can localize on-screen text and product screenshots alongside voiceover. It also runs an automated brand check before sending creative for review. Human approval is on by default; teams can later allow automatic publishing for selected campaigns or types of variants, within budget rules they set. Runway is asking marketing teams to let software produce and iterate on ads, while keeping a person responsible for what goes live. The system can change how much work a team can test, but the announcement does not show whether the additional variants improve results across customers or categories. Runway's own numbers are a useful demonstration of its workflow, not an independent test of its effectiveness. The competition is already selling the loop The product enters a category with existing claims to the same workflow. Omneky's platform https://www.omneky.com/mcp?ref=runtimewire describes generating creative, connecting ad accounts, launching campaigns across major platforms and using performance data to inform optimization. Runway's argument is that its generative video capabilities, localization, brand checks and publishing tools can sit together in one system. The announcement does not establish that combination as unique, and buyers will have to judge the quality of the creative and the effect on campaign economics. The ad platforms themselves are also building AI tools for creative and campaign work. Google https://business.google.com/us/accelerate/announcements/?ref=runtimewire has been expanding AI Max for Search campaigns, including creative capabilities, while Meta https://about.fb.com/news/2025/02/gen-ai-transparency-metas-ads-products/?ref=runtimewire has described generative AI features for advertisers. Runway depends on those platforms for distribution and performance signals, but it is also trying to own a workflow that overlaps with their own automation efforts. This launch extends Runway's move from making generative models toward products that organize how people use them. In July, RuntimeWire reported on Runway's Media Router https://runtimewire.com/article/runway-launches-the-first-ai-router-for-image-video-and-audio-models , which routes generation requests across image, video and audio models. Ads takes a related product-building approach into marketing: it ties generation to a downstream task, then uses results from that task to guide the next output. Runway's case rests on connecting generation to campaign outcomes. Runway has substantial capital behind that expansion. TechCrunch reported https://techcrunch.com/2026/02/10/ai-video-startup-runway-raises-315m-at-5-3b-valuation-eyes-more-capable-world-models/?ref=runtimewire in February that Runway raised a $315 million Series E at a $5.3 billion valuation, led by General Atlantic, with NVIDIA and other investors participating. That round was framed around developing world models and bringing them into new products and industries. Ads gives Runway another commercial application for its media-generation technology, while placing it in a market where customers can evaluate claims against campaign outcomes. For Germanidis, the near-term test is whether Runway can turn its own internal operating system into a product other teams will trust with their brands and budgets. The pilot puts the system into enterprise settings; the published performance figures remain Runway's own. Customer results will show whether the workflow improves campaign performance at scale.