Before their AI was functional, Bespoke's founder and team manually responded to user chats, pretending to be a bot. This "Wizard of Oz" method proved market demand and provided invaluable data on user behavior, which investors initially doubted.
Bespoke's chatbot didn't just help travelers; it solved a major pain point for Narita Airport: high employee turnover at information desks caused by stressful interactions with frustrated, non-Japanese speaking passengers. This business-critical solution led to a major enterprise contract.
To understand tourists, Bespoke's founder conducted relentless in-person interviews at hubs like Shibuya Crossing, leading to her getting banned. This scrappy approach highlights the extreme lengths required for effective customer discovery, including using dating apps to source interviewees.
When Bespoke's chatbot broke during a holiday, the founder became the bot for a week. This revealed that less efficient, more conversational interactions significantly increased user engagement. This insight, contradicting the goal of pure efficiency, became a key product differentiator.
Rather than continuously raising venture capital, Bespoke used its contracts with government entities as collateral to secure bank loans in Japan. This provides a faster, non-dilutive funding alternative for profitable startups with stable, long-term government revenue, preserving founder equity.
A unique consequence of Japan's aging population is that many profitable businesses, like factories, are shutting down simply because owners retire without a successor. This creates a massive, overlooked opportunity for entrepreneurs to acquire and modernize these cash-flowing but 'orphaned' companies.
To train non-Japanese speaking workers in skilled trades, Bespoke has Japanese experts wear smart glasses to record their process. The resulting first-person video is translated into multiple languages, creating a scalable training library that bypasses the need for bilingual trainers on-site.
When an engineer resigned, Bespoke didn't automatically backfill the position. Instead, they invested in AI coding assistants for the entire remaining team. They decided the collective productivity boost from AI tooling was a better, more scalable solution than hiring a new person.