Bold & Boutique

Louise Macfadyen: Designing AI Interfaces

Louise is the author of the upcoming book Designing AI Interfaces, which distills her extensive experience from tech giants like Google and Microsoft into practical guidance for creating exceptional AI products. Whether you're a seasoned designer or just getting started with AI, this conversation offers valuable insights into the future of interface design and the unique challenges of building products in the age of artificial intelligence.

Ben Sauer: How to present high-stakes work and live to tell the tale.

Most work doesn’t speak for itself. This episode is for everyone who's sick of presenting amazing work and ideas that don't land. International keynote speaker, coach, and designer Ben Sauer breaks down how he teaches the art of strategic storytelling. Ben Sauer trained people all over the world, including at NASA, and his frameworks are used by organizations like Amazon, Google, and the BBC. In this episode, Phoebe Ohayon and Ben Sauer discuss his book, Death by Screens, which teaches designers (or anyone with something to pitch) how to present their idea without boring people to tears!

TRAILER Season 1 Bold & Boutique

Bold & Boutique isn’t just about voice, design or AI — it’s about building something meaningful by going deep, not wide. It’s a show for creatives, strategists, and entrepreneurs who believe that staying niche can be a superpower. Let’s explore what it really takes to grow a business rooted in expertise. Season 1 dives into the rise of AI characters and virtual brand ambassadors, but future seasons will open up to a wide range of topics shaping the creative and technological landscape, all with the same boutique lens: focused & intentional.

#6 The Secret Behind Scalable AI Characters

After user research and quantitative testing comes the real force multiplier: a design system that lets every team spin up on-brand AI characters without reinventing the wheel. Phoebe explains why a system is more than a style guide and how it becomes the shared language for voice, visuals, and behavior across markets. Highlights - Style guide ≠ design system – how reusable assets, research-backed rules, and governance turn a static brand book into a living toolkit that accelerates creative work instead of policing it. - Core building blocks – voice-tone tokens, expression libraries, “basic pause” timing, fallback-response prompts, error-recovery patterns, earcons, and perception-test templates. - Begin with the end in mind – anchoring the system to the company’s mission so every asset (from TTS voice profiles to UX flows) ladders up to the same north star. - Politics & practicality – picking the right contributors, defusing naming battles, and using data (not opinions) to resolve push-pull between global consistency and local nuance. - Lightweight → enterprise-grade – why a starter kit of reusable prompts is still a design system, and how it can grow into code packages, speech-data pipelines, and performance benchmarks. - Scalable guidance, not rigid rules – giving regional teams research frameworks and analysis tools to localize characters while staying unmistakably on-brand. If you’ve nailed a single AI character and are wondering, “How do we replicate this across products, languages, or markets without chaos?”—this episode shows how to turn hard-won insights into a system that scales creativity rather than stifling it.

#5 How to Use Data to Design a Personality

Now that the standout virtual-character concepts have been refined through qualitative feedback, Phoebe and Maikel jump into the numbers phase—where data, not gut feeling, decides which designs go live. Key take-aways - From shortlist to scorecard – Why you first narrow to a small set of high-potential designs, then pit them head-to-head in structured tests rather than betting the budget on a boardroom favorite. - Quant 101 for character design – Crafting perception surveys in tools like Qualtrics, defining traits (trust, energy, professionalism) in testable language, and keeping sessions under 20 minutes so attention—and data quality—stay high. - Participants that matter – Sourcing 100-300 respondents who match the target market, building demographic diversity, and spotting red-flag responses (five-minute completions, straight-line scores) before they skew results. - Hybrid methods – Using “think-aloud” moderated surveys with a sub-sample to capture the why behind the sliders, linking qualitative nuance to quantitative scale. - Pilot before you plunge – Quick internal dry-runs expose confusing wording, broken sliders, or survey fatigue before hundreds of users see it. - Making the data talk – Cleaning noisy responses, running correlation and subgroup analysis (e.g., slower speech → higher trust among 60-plus drivers), and translating findings into concrete visual, vocal, or personality tweaks. - Knowledge that compounds – Each quantified insight becomes part of a growing, cross-project playbook that lets future teams start smarter and align faster. If Episode 3 covered inspiration and Episode 4 covered refinement, Episode 5 shows how to turn user perception into hard evidence—so the AI character that finally ships is the one your audience actually wants.