What happened: A single audio ad script for Wella’s Nioxin Pro Clinical line became 100 distinct versions through AI production platform AudioStack, tailored across two audience segments and 24 UK locations for a campaign run by media agency Digitas and ad tech firm Azerion. A brand lift study measured by On Device Research against a control group found 70% of exposed listeners took or intended to take action, plus a 21-point recall lift and 22-point awareness lift among listeners who had personally experienced hair thinning.
Why it matters: The campaign is a working example of what generative AI changes in media production economics, not just creative novelty. “One team gave us one script, and our platform turned it into a hundred versions that actually speak to the person hearing them,” said Alexis van de Wyer, CEO of AudioStack. Producing 100 studio-quality variations manually, across that many locations and segments, would typically cost more and take longer than most brands could justify for a single line extension. AI production removes that constraint, which changes which campaigns get personalized treatment at all, not just how they get made.
The original insight: Julia Drucker-Allister, EMEA media director at Wella, framed the shift as moving “into personalisation at scale,” and the results suggest the audio channel specifically benefits from AI production faster than video does, since a script-to-audio pipeline has fewer production steps to automate than a script-to-video one. That makes audio a likely proving ground for AI-generated ad variation ahead of harder formats, echoing the broader push toward AI-native advertising infrastructure and running alongside gains already showing up in agentic buying tools for streaming audio.
Source: ExchangeWire
