AI-Powered Messaging Gap Analysis in Two Hours at Docebo

This case study was created from an episode of the Product Marketing Adventure Podcast. It has been adapted into Dusted's case study format to preserve and showcase the guest's real-world experience. Written to preserve the voice of the true author, Emily Pick.

I had only been at Docebo for about two and a half months—my first enterprise public company and my first time in the learning and development space. Docebo is a learning platform serving enterprise audiences for both internal use cases (employee training, compliance, onboarding) and external ones (customer education, partner and franchise training).

Our CRO brought in a former mentor to run a full end-to-end pipeline analysis across lead sources, progression, and backend lead-gen systems. One clear finding stood out: a specific segment was closing at a rate six to seven percent lower than our benchmark.

The drop-off was happening after the demo stage. Our hypothesis was straightforward: we were not telling the market the right message and were not meeting buyer expectations once they reached a demo.

I needed to investigate what this segment expected when they came in, why we won or lost opportunities, whether outcomes changed by persona, company size, or industry, and how our past product reality and future roadmap should shape what we leaned into or avoided. The request landed around noon on a Friday. Pre-AI, gathering win-loss data, NPS trends, call insights, and industry signals alone would have taken days or weeks before any real analysis began.