S10E219: Why 88% of Schedules Fail, and What AI Still Can't Fix with Michael Pink
About this episode
The episode discusses the failures of project schedules and the limitations of AI in providing accurate insights based on flawed data.
88% of project schedules don't follow best practice. So why are we so keen to feed that data into AI and trust the output? In this episode, Dale and Val are joined by Michael Pink, Founder of SmartPM, for a sharp conversation on the forensic side of delay and the state of AI in project controls. Mike has analysed hundreds of thousands of schedules and built a career on working out who delayed what, and why nobody could ever agree. His core message: clean data is not the same as honest data. A technically tidy schedule can tell you a story that never happened, and if you point AI at it, you don't get insight, you get a confident wrong answer at scale. What we cover: Mike's route into construction, from a study-abroad debt to KPMG's construction consulting group 25 years of change in the people and process side of scheduling The shift from reactive forensic work to proactive project controls Data honesty vs data quality, and why integrity is the combination of both (framing credited to Isaac Dyer and Carlos Sanchez) The 88% problem: missing logic, inflated float and moving dates The compression factor, and the exact point a schedule becomes infeasible Why a weekly update beats a…
More episodes of The Project Chatter Podcast
Explore listener stats, chart rankings, contacts and more on the The Project Chatter Podcast podcast page.