Can AI Serve as Your Mentor? Practica, a startup, believes it can. The company, originally founded as a platform for one-on-one executive coaching, has now introduced an AI system that leverages the knowledge it has accumulated from working with a diverse group of human coaches over the years. This AI chatbot offers a personalized workplace mentoring and coaching experience, aimed at helping professionals improve their skills in various areas, including management, strategy, sales, personal development, growth, customer success, marketing, data, design, finance, and more.
Founded in January 2020 by Dave Whittemore, former Thinkful head of Product (which was acquired by Chegg), and former Dropbox engineering manager Andy Scheff, Practica initially tackled the challenge of continuous upskilling throughout one’s career through a traditional executive coaching marketplace.
“We both became executive coaches,” Whittemore explains. “Andy and I both immersed ourselves in it… I coached product managers, and Andy coached engineers.”
During that year, the platform expanded by adding a marketplace for other executive coaches, and it now boasts 250 human coaches with expertise in specific domains. The majority of the business, about 90%, is focused on serving businesses (B2B). This aspect of the company has also become profitable.
However, the founders recognized that the pricing structure made their services inaccessible to many people.
“The average hourly price was $200, and it varied based on the seniority of the person being coached and the coach,” Whittemore stated. “If you were coached for a full year, the average cost was about $3,000 per year. This pricing acted as a barrier, which led us to explore AI coaching.”
The idea was to combine their insights from one-on-one coaching with AI technology. Over the years, they had identified the factors that make coaching relationships successful, which guided them in creating a framework around and on top of a Language Model (LLM). Their knowledge base includes a vast collection of publicly available learning materials, such as blog posts, conference talks, videos, podcasts, books, and more, curated into hundreds of different skills across various topics.
It’s worth noting that many websites today block AI web crawlers, but Practica differentiates itself by directing users to the sources of its content, thus driving more traffic to the original publishers.
However, Practica has not formalized licensing agreements with its educational source providers, which include resources as simple as a helpful blog post from an engineer or a personal account of overcoming managerial challenges.
Regarding the AI models used, Practica aims to be vendor-agnostic and focuses on the application layer. This means it doesn’t directly compete with companies like OpenAI or Google but can collaborate with them.
Practica employs a technique called Retrieval Augmented Generation (RAG) to match the best learning resources for a learner’s specific situation. The AI coach provides explanations for the retrieved sources and why they are beneficial, much like a human coach would. It references the sources and encourages users to explore them further as “homework.”
Practica’s use of third-party content functions more as a curated search engine than machine learning model training, with AI serving as the coach. This approach combines various coaching tools, including instruction, contextual questioning, identification of current challenges in the learner’s job as learning material, aligning learning progress with career goals, and acknowledging achievements. It organizes insights from materials into an interactive list with notes for future reference.
What sets Practica apart from typical AI chatbots is its ability to remember a learner’s history, allowing it to build on their skills over time, much like a human coach would.
The system underwent private testing starting in July and is now available to individual learners for a monthly fee ranging from $10 to $20 per user. An employer-focused “teams” version is also in limited testing.
At this price point, Practica aims to make executive coaching more accessible and hopes to introduce more people to AI coaching, potentially leading them to explore the company’s premium human coaching services.
Practica secured $1.5 million in external funding across two rounds in 2021 and 2022 before transitioning into AI-based coaching. Both rounds were led by Script Capital and featured over 40 individual angel investors, many of whom were leaders in the domains where the company focuses its coaching.
