The Right Problem

2026-05-24 · Source: Artificial Intelligence on Medium · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Entrepreneurship & Start-ups · Depth: Novice, medium

Summary

The author's 2012 project, thebooq.com, was a human-curated search response system designed to replace Google's link lists with single-page summaries structured by user intent. Submitted to Y Combinator in 2016, it was passed over, yet the author maintained it for a decade, believing it addressed a critical problem. The core idea stemmed from analyzing top search queries, such as "dog" (2 million Americans monthly), and grouping them by actual user intent (e.g., 36% wanted to obtain a dog, 21% wanted products), proposing a tree-like knowledge structure instead of generic definitions. Despite initial challenges like lack of funding and co-founders, the market has recently validated this problem space, with companies like Perplexity, Glean (raised \$200 million), and Microsoft's \$10 billion investment in OpenAI addressing similar knowledge retrieval and organization issues. The author notes that while AI has advanced, fundamental problems like user-provider conflict of interest and the vast majority (90-99%) of human knowledge remaining unstructured and off the web persist, suggesting a future shift towards planned knowledge production.

Key takeaway

For AI Product Managers evaluating new knowledge management solutions, recognize that current AI tools, while advanced, still operate on a fraction of existing human knowledge. Focus your strategy on solutions that not only retrieve but also actively structure and produce knowledge, especially within enterprise silos. Consider how your product can facilitate the planned creation of missing information, moving beyond mere aggregation to true knowledge production.

Key insights

Solving the "right problem" too early, alone, and invisibly can precede significant market validation.

Principles

• User search intent often differs from search engine results.
• Knowledge should be structured by human intent, not scientific classification.
• Most human knowledge (90-99%) remains off the web.

Method

Analyze top search queries, group them by user intent, and structure knowledge as a navigable tree of choices, moving from broad to specific.

In practice

• Group search queries by user intent to identify true needs.
• Design knowledge structures as decision trees for user navigation.
• Prioritize structuring internal organizational knowledge gaps.

Topics

• Knowledge Management
• Search Intent
• AI Applications
• Information Retrieval
• Knowledge Graphs
• User Experience

Best for: Product Manager, Investor, Entrepreneur, AI Product Manager, Director of AI/ML

Related on AIssential

• Knowledge Engineering for Search and Content: A Practical Guide — Structured understanding of content and user intent is crucial for effective LLM-powered search and content experiences.
• The Context Gap — "Context graphs" are a market response to the self-inflicted wound of lost process knowledge due to decades of outsourcing.
• Systems for Organizing — Effective AI requires structured knowledge organization systems mirroring human cognition, not just statistical approximations.
• How to Build an Efficient Knowledge Base for AI Models — A systematic, iterative approach to knowledge base construction is essential for AI model accuracy and scalability.
• Building Knowledge Graphs As An Agentic Operating System — An agentic operating system for business strategy aligns people and orchestrates transformations using causal thinking and knowledge graphs.
• Why Early Knowledge Graph Adopters Will Win the AI Race — Early adoption of knowledge graphs and ontologies provides a competitive edge in enterprise AI.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Artificial Intelligence on Medium.