Joe Way opened episode 334 of the Higher Ed AV Podcast by saying he was widening the show beyond the AV silo. He got more than he bargained for: Peter Murphy Lewis spent forty-five minutes arguing that higher education’s biggest communication problem is not budget, politics or AI — it is that the sector keeps making its case with data when only stories move people.
The line Joe stopped on came from Peter’s own notes: “You can’t tell someone’s story from a distance.” What follows is the working method behind it — a before-and-after framework, a way to advocate for an idea without sounding self-serving, and two prompting habits that stop AI flattening your voice.
Peter came to this the long way round: a politics PhD in South America, a professorship, a travel company started at 27 because nobody would hire him, a CNN en Español analyst slot, and then a Spanish-language TV show in the mould of Dirty Jobs that ran to more than 120 jobs on camera. He now runs fractional CMO engagements and produces documentary storytelling projects. There is a written companion to this episode in our article on school marketing that retains families.
Key Takeaways
1. Data does not change minds — stories do, and Ken Burns said it first
The hinge of the episode. “Ken Burns says you’re never going to change anybody’s opinion with more data,” Peter says, then makes it personal: his own views on HIV and AIDS changed because he worked as an HIV/AIDS caseworker, not because anyone handed him a report. He applies this straight to higher ed. A sector arguing its value with enrolment figures and funding gaps sounds like an executive asking for money. A sector that puts a student or a professor on camera and lets them describe what changed sounds like something worth defending.
2. The before-and-after framework, and why it works at any depth
Peter’s whole storytelling method reduces to three questions: who was I before, what was the thing that changed, and who am I after. He demonstrates it twice in one breath — once trivially (he ate mochi for the first time at 44 and now his house has nothing else) and once seriously (he became a certified nursing assistant at 40, which changed how he thinks about long-term care and how much time he wants to spend with America’s seniors). The superficial version is the point: the structure is easy enough that anyone can run it on themselves.
3. Tell the frontline story, not the executive one
Peter describes pitching a state healthcare executive on a documentary told through the eyes of caregivers rather than residents or administrators. What he learned afterwards is the transferable part: people who have given decades to an industry are intensely passionate and have simply never been handed a microphone. And an executive telling the same story invites the opposite reaction — you only do this for profit, you don’t work Christmas. The credibility sits with the person doing the work.
4. Advocate for someone else’s idea first
His answer to the person who has a good idea and cannot bring themselves to pitch it: don’t start with your own. Champion a colleague’s. “It’s pretty easy for us to sell somebody else’s, because it doesn’t come off as arrogant, self-centred, what am I getting?” You get the reps in without the exposure, and it turns out to be contagious — people start advocating for you when you are not in the room.
5. Imposter syndrome does not go away, so aim to be the most useful person in the room
Peter, at 45, with a team behind him and a speaking career, says he still gets nervous and still sometimes thinks he shouldn’t be there. His counter is not confidence but usefulness: carry the chairs, help someone else feel less exposed, take one person from zero to one. His concrete version for anyone building a public voice — think of five people at this event you could each help with one question, and start by documenting that.
6. AI is not the threat to authenticity; ungoverned AI is
Asked directly whether leaning on AI costs you your own voice, Peter says the output quality is downstream of what you put in — your context, your framing, the way you ask, all of which came from your own before-and-afters. He offers two habits that keep it honest: make the model ask you questions before it answers, and make it rank its claims from most to least certain and argue why each might be wrong. “Treat ChatGPT as if you’re that fifteen-year-old getting ready to break up with it.”