PRINTABLE VERSION HERE

In an industry where information is plentiful but time is limited, retail professionals are seeking ways to make ideas more digestible, shareable, and memorable. This is why the International Association of Department Stores (IADS)  has decided to evolve from static formats to dynamic audio content, with the launch of IADS Retail Park, an AI-powered podcast series. Importantly, we have not abandoned our static formats; instead, we are embracing this new approach alongside them.

What started as an experiment to convert written exclusives into human-like audio stories has matured into a repeatable system that mimics the ebb and flow of honest conversations. This model bridges language gaps, compresses production timelines, and leverages AI not just for speed but for simulated connection.

This Exclusive aims at sharing the key learnings made so far with our members.

Why audio, why now?

Retail teams today are drowning in information but starving for meaning. Between dashboards, presentations, and constant communications, there's often little time to digest the why behind the what. Audio fills that gap by enabling passive yet meaningful learning, allowing retailers to learn while they do.

Unlike conventional media, podcasts provide an intimate listening experience that feels personal and relatable. As Scott Galloway notes, “When people approach me in the wild, it’s easy to discern where they’ve been exposed to my content. […] If they greet me like a friend they haven’t seen in a while, podcast. It’s a very intimate medium. You are physically in somebody’s ear, in a private setting — washing the dishes, working out, walking the dog. It’s just you and them.” When hosts interact with their listeners in a warm, familiar tone, it fosters a sense of connection and companionship, making the audience feel included in a conversation rather than just receiving information passively. This personal relationship boosts listener engagement and loyalty, encouraging audiences to return to voices that resonate with them on an emotional level. In a world full of distractions and competing sources of information, this personal touch can elevate a podcast from just another show to a vital source of insight and motivation.

The growth of podcasting is no longer speculative. With over 500 million listeners worldwide and platforms like YouTube spearheading its popularity, podcasts have become an integral part of our information landscape.

The attention deficit in retail

Retail professionals today face an overload of communication (constant emails, dashboards, decks, messages). But volume doesn't equate to understanding. The real issue is a scarcity of high-quality attention. McKinsey’s 2025 report on the “attention equation” argues that time spent engaging with media tells only part of the story. The quality of attention, measured by focus and intent, is what truly drives understanding and action.

Audio offers professionals to absorb targeted insights during “in-between” moments—on a stockroom break, during a morning commute, or while resetting a display. Rather than carving out extra time to consume content, employees absorb insights seamlessly within their routines, increasing the likelihood of information retention and action.  McKinsey’s findings show that consumers in the top quartile of attention spend twice as much as those in the bottom quartile. In an internal business context, this translates into more focused and better-informed retail employees who make smarter decisions and are more likely to act in alignment with company goals. With generative AI now able to simulate human tone and behaviour, these insights can be delivered in voices that sound intuitive and familiar.

Why podcasts work

Unlike traditional text-based communications, podcasts are intimate and emotionally resonant. They invite listeners into a shared space, creating what Galloway calls "companion media"—where the delivery feels less like a broadcast and more like a conversation. The spoken voice, especially when generated with attention to tone and cadence, can signal curiosity, authority, or empathy. It reinforces ideas not just through what is said, but also through how it's said.

Moreover, podcasts reduce cognitive friction. There’s no need to sit down and read. The message comes to the listener in a form that’s easy to consume and, often, more memorable than the written word. Podcasts are becoming the modern analogue to print magazines and newspapers—media consumed not only for entertainment, but also for education and professional development. Roughly three-quarters of podcast listeners use the same platform to stream both podcasts and music, making audio content an ambient part of their media diet. The podcast medium is also dynamic—27% of listeners consume content at accelerated speeds, particularly among Gen Z and millennials.

This emotional resonance is further enhanced by AI voice cloning, which enables podcast hosts to sound not only human but familiar, often replicating the tone and cadence of a known executive or contributor. This is increasingly relevant in retail, where leadership visibility is vital but usually limited by logistical constraints. Hearing a trusted voice (even an AI-generated one) can reinforce a sense of connection and clarity across teams. A study led by Cornell University found that students who listened to personalised AI-generated podcasts not only enjoyed the experience more but also retained information better and learned more effectively. Meanwhile, separate research shows that 80% of people perceive AI-generated voices as real, and most struggle to distinguish them from actual human speakers.

A clear example of this in action can be seen with Langham Logistics, which partnered with Stratablue to implement an AI voice agent that processes employee call-ins and immediately delivers updates to managers, ensuring consistent, human-like messaging across its workforce. Applied in internal retail briefings, this level of consistency helps forge emotional bonds between leadership and frontline teams. That’s powerful: when a familiar executive voice is mimicked believably, team members are more likely to trust the message and stay aligned unconsciously.

The result is a new paradigm for business communication. A podcast episode, briefing, or internal memo can now be drafted, voiced, and distributed in hours rather than days. And unlike conventional content formats, these audio pieces carry a personal tone that encourages engagement rather than obligation. This strategic use of audio enables retail organisations to create emotional resonance and align teams across locations, turning communication into connection and information into momentum.

Retailers already on air : Department stores and frontline voices

IADS is not the only one embracing audio as a strategic tool. Several department stores and retail groups already run their own podcasts. Galeries Lafayette’s Minuit aux Galeries shared behind-the-scenes stories after hours to celebrate their 150th anniversary. Harrods’ True Tales from Harrods brings in designers and creatives to explore what luxury means today. The Chalhoub Group runs The Podcast by Chalhoub Group (YouTube), an ongoing series that began in 2021 and remains active today, hosted by Lynn Al Khatib, VP of Communications. The show features regular conversations with internal leaders, partners, and industry figures on topics such as innovation, sustainability, and organisational culture, making it one of the few department store group podcasts consistently produced. Ámbito Cultural (YouTube), the cultural arm of El Corte Inglés, extends its in-store programming (literary events, exhibitions, and performances) through online recordings and talks, using digital content to broaden its cultural reach. Nordstrom’s The Nordy Pod, hosted by Pete Nordstrom, is a familiar example of how retailers are shifting from sellers to storytellers—using podcasts to share culture, leadership, and customer stories, and enhancing customer connection and brand visibility. These examples reflect what the industry already recognises: audio is now an established tool for connection, storytelling, and visibility

At the same time, there’s growing recognition of the value in podcasts hosted by or featuring front-line employees. Shows like Frontline Fridays, Retail Warzone, and Frontline Innovators bring unfiltered perspectives from the shop floor. They share stories about store operations, leadership, and workplace culture that rarely appear in formal communications. These voices bring balance to leadership messaging, revealing how strategy is put into practice.

Building the podcast – The IADS workflow

Since the IADS is committed to continually learning and sharing with its members and the retail community, we have developed a process from scratch and are now sharing the key learnings, with the hope that this information will be useful to retailers.

From page to prompt

The podcasting process at IADS begins with a clear editorial strategy: repurpose written content into audio dialogue that feels conversational rather than read. The first step involves using tools like Dust.tt to convert editorial pieces into a back-and-forth format, assigning roles and simulating a casual rapport between two hosts. Prompts are crafted not just to summarise content, but to inject realism—pauses, clarifications, side comments—that mimic natural speech. Human editors then refine the AI-generated draft to improve clarity and flow, preserve the core insight while removing any robotic inflexion or repetition. This is the first checkpoint where editorial sensibility meets automation.

In the early stages of development, several AI platforms were tested. Tools like Google’s NotebookLM, then still in beta, showed promise in converting written materials into podcast-ready dialogue. However, its automated scripting often lacked editorial precision, reordering ideas or inserting speculative commentary that strayed from the intended narrative. For IADS, having 100% control over what was said and how it was said was critical. This prompted a broader search and trial of platforms including Podcastle, Play.ht, and Murf.ai. While many offered high-quality voice options and intuitive features, few provided the level of script fidelity and voice customisation required to simulate truly editorialised dialogue. The takeaway: no single tool could meet every need. Instead, IADS developed a modular system that combined best-in-class AI capabilities with human oversight at each stage.

This process wasn’t just about finding the most advanced tool; it was about identifying the right fit. IADS needed a system that supported scripted conversations with editorial control, a familiar vocal presence, and the flexibility to iterate quickly. The human editing team remained central throughout, shaping scripts, correcting tone, and aligning each segment with the voice of the original piece. The result is a format that scales efficiently while preserving nuance, intention, and warmth.

Vocal authenticity through AI

Once the script is finalised, it moves to voicing. Using platforms like Speechify, cloned voice profiles for regular hosts read the scripts aloud. These voices have been trained on their real speaking styles, enabling a more personal and recognisable listening experience. Adjustments to tempo, pitch, or emotional emphasis are manually applied where needed to reflect the appropriate tone.

Particular care is taken with pronunciation, especially with brand names or geographic references such as "Printemps" or "Monoprix." Editors will often use phonetic spelling or AI-specific markup to ensure accuracy. The acronym "I.A.D.S" is also spelt out, never read as a word, to maintain consistency and clarity.

Editing, packaging, and distribution

The final audio file is edited in tools like Audacity, where sound levels are adjusted, music bumpers are added, and segments are stitched together to create a seamless flow. Intro and outro jingles—short, branded audio cues—bookend each episode.

Episodes are distributed through a multi-platform strategy. They are hosted on Substack (which doubles as a transcript archive) and repurposed for YouTube and internal platforms. This distribution model allows IADS to reach listeners where they already are, whether browsing podcast platforms or catching up on content via email.

The entire pipeline—from drafting to final publication—can be completed in under a day. This allows IADS to respond to new developments or highlight stories quickly, reinforcing its value as a real-time knowledge partner.

Human-centric AI and the sound of storytelling

AI simulating human rapport

One of the key innovations powering the IADS Retail Park is its ability to simulate the nuances of human conversation through generative AI. Traditional AI-generated content often sounds either overly scripted or unnervingly robotic. IADS addressed this challenge by developing agent-style dialogue structures that incorporate elements of natural interaction, including interruptions, clarifying questions, and expressions of curiosity. This approach builds on recent advances in large language models, which now simulate not only coherent dialogue but also context-aware personas. These agents mimic not only human speech but also interpersonal dynamics, creating a sense of shared understanding.

This aligns with research from Stanford HAI showing that generative agents can replicate real-world answers with 85% accuracy, and with Auxiliobits’ findings that “unstructured data gathered from social media interactions” supports AI systems in learning emotional context and decision heuristics. This isn’t mimicry, it can be, though, out of operational empathy. To maintain trust and coherence, every episode undergoes a three-step loop: content grounding using memory modules, dialogue naturalisation with conversational pacing, and listener persona simulation to test clarity and emotional tone. Only when each layer passes do hosts go to voice, with phonetic disambiguation added for complex names.

Cloning voices, not people

The voices featured in each episode are not chosen randomly. They are based on real people who work at IADS, like Maya or Anchita, but enhanced by cloned voice models. These profiles are created through ethical voice training, with complete transparency and approval. This cloning allows for consistency across episodes while also maintaining a personal touch. Now, contributors whose voices are cloned opt in through recorded samples and are kept informed of how their voice models are used.

It’s important to note that these voices aren’t static. Human editors adjust pitch, cadence, and pauses based on the episode’s tone. Whether it’s a light-hearted commentary or a serious policy review, the sound is tailored to fit. In doing so, IADS avoids the uncanny valley of synthetic speech and instead delivers something closer to radio journalism.

This level of sonic authenticity helps overcome a barrier in AI adoption, listener trust. When a podcast sounds warm and familiar, it becomes easier to accept that it’s AI-assisted, not AI-imposed.

Designing for emotional intimacy

The strength of podcasting lies not only in what is said, but in how it’s felt. At IADS, this emotional resonance is deliberately designed. Dialogue is scripted to feel conversational, familiar, human, and unscripted, using templates that encourage back-and-forth exchanges. Host 1 might respond to Host 2’s point before moving to the next question, while Host 2 offers brief reflections before answering. These moments create the rhythm of a real conversation. To support this tone, reusable scripting templates were developed to guide structure and phrasing. Editors asked prompts like: “Can Host 1 relate to what was just said?” or “Can we replace robotic affirmations like ‘Absolutely’ with more natural responses?” Pauses were added manually (e.g., [pause 0.25s]) to simulate human timing and improve flow.

AI-cloned voices were refined through repeated editing to improve pronunciation and tone. Tools like Perplexity and Grammarly helped rewrite scripts to sound more like natural speech. Voice outputs were regenerated up to three times per segment to fix mispronunciations or flatten robotic inflexion. Phonetic spellings were often used to ensure clarity for names and non-English words. This wasn’t just about getting the script “right.” It was about crafting something that felt warm, relatable, and thoughtful. The result isn’t artificial realism—it’s a listening experience that feels intentional and human, even when powered by AI. This approach is supported by growing research showing that emotionally resonant media improves retention and can drive behavioural change.

What this means for retailers

Retailers as publishers

The learnings made along the IADS Retail Park development journey offers some insights for retailers looking to explore audio content. At its core, the model positions the retailer not just as a merchant but as a media entity. Any brand that generates insight—be it in customer experience, sustainability, or product design—can turn those ideas into episodes. These aren’t promotional ads; they’re value-driven conversations.

Retailers can easily turn employee onboarding guides, product explainers, or executive interviews into digestible podcast episodes. These can be distributed internally for training or externally to bolster thought leadership. Furthermore, with AI tools automating most of the pipeline, the barrier to entry is significantly reduced. Retailers should ask themselves: what insights are trapped in decks or reports that could live more vibrantly in a voice?

Generative AI now enables role-specific audio content on demand. Weekly voice briefings for merchandisers or planners, tailored by region or function, are already in use. These human-like segments bridge the gap between leadership and frontline teams, delivering updates in a way that feels natural and intuitive.

Starting small, learning fast

What sets IADS apart is not just the outcome but the approach. The Retail Park podcast was born as an experiment, refined in public, and improved with feedback. Pronunciation issues, pacing oddities, and tonal misfires were addressed not with overhauls but iterations.

This agile development approach, more familiar in the tech sector than in retail, allowed IADS to refine its content engine continuously. Retailers exploring similar projects need not fear imperfection. The key is to start with a manageable scope, such as one series, one team, or one story.

Over time, these pilots can evolve into full-fledged audio programs. With AI doing the heavy lifting and editors steering tone and intent, the process becomes less about production muscle and more about editorial vision. Ultimately, retailers that adopt a test-and-learn mindset—focusing on utility, authenticity, and speed—will find that audio isn’t just a trend. It’s a new layer of brand presence.

What IADS has accomplished with Retail Park is not simply a creative experiment—it’s a signal of what’s possible when legacy institutions embrace emerging technology without losing sight of human connection. This project proves that audio can be more than entertainment or marketing filler. It can serve as an operational tool, a cultural artefact, a knowledge vehicle, and, above all, a strategic asset.

For retailers, the message is clear: you don’t need a recording studio to sound present, or a celebrity voice to sound human. You need clarity of purpose, a willingness to prototype, and the right blend of AI and editorial input to turn your everyday ideas into experiences that travel. Whether you’re onboarding seasonal staff, communicating sustainability goals, or simply reinforcing company values, audio, especially when assisted by generative AI, offers unmatched speed, reach, and relatability.

So if you’re in retail and wondering what the future of brand communication sounds like, don’t just imagine it. With summer knocking at your door, let the  IADS Retail Park join you —whether you’re poolside, in transit, or recharging between meetings. Pop in your headphones and catch up on smart, surprising retail stories that travel as well as you do.(Available now on Apple Podcast, Spotify, Substack, and Youtube)

Credits: IADS (Maya Sankoh)