PureTech uses AI to accelerate the initial steps of its process: identifying promising discontinued drugs and pinpointing what held them back. However, the crucial step of devising the scientific solution to fix the drug remains a human-driven, creative insight process, blending AI's scale with human ingenuity.

Innovation fails when treated as a sporadic event. Walmart established a formal, stage-gated pipeline (intake, evaluation, POC, MVP) that operates outside normal planning cycles. This systematic process provides a clear path for ideas to be validated and funded, increasing their success rate.

Progress in drug development often hides inside failures. A therapy that fails in one clinical trial can provide critical scientific learnings. One company leveraged insights from a failed study to redesign a subsequent trial, which was successful and led to the drug's approval.

Conventional innovation starts with a well-defined problem. Afeyan argues this is limiting. A more powerful approach is to search for new value pools by exploring problems and potential solutions in parallel, allowing for unexpected discoveries that problem-first thinking would miss.

To combat bias, the team contractually agrees on strict, predefined success metrics for major milestones *before* any data is generated. A program either meets the criteria or it doesn't, removing ambiguity from go/no-go decisions. This discipline is applied both internally and at the board level for spun-out companies.

Successful MedTech innovation starts by identifying a pressing, real-world clinical problem and then developing a solution. This 'problem-first' approach is more effective than creating a technology and searching for an application, a common pitfall for founders with academic backgrounds.

For small biotechs, the playbook for success extends beyond scientific discovery. It requires creativity and innovation in the operational process itself—finding efficient paths through regulatory checkpoints, securing non-traditional funding, and leveraging external resources to advance development with limited capital.

The Innovative Genomics Institute is tackling rare diseases by creating a standardized platform. By keeping elements like the delivery vehicle and enzyme constant and only changing the guide RNA, they aim to create a repeatable 'bucket trial' process for developing hundreds of cures, not just one-offs.

For net-new products, begin with deep problem discovery. Once a product is introduced, shift to rapid, solution-based iteration and feedback. As the product matures, revert back to problem discovery to find the next growth engine while optimizing the current product.