Every organization I’ve worked for has been in the business of asking people to believe something.
A utility asking a million customers to believe the power will come back on. A hospice asking a family to believe they’re making the right decision at the most vulnerable moment of their lives.
Different stakes … same work.
Nearly 30 years in, I’ve stopped believing you can message your way to that kind of belief. That belief is built in small, unglamorous places:
- The outage notification that arrives before people start guessing.
- The intake conversation that does not rush.
- The storm update at 2 a.m. that says what is actually known.
Communications is the discipline of being worth believing before you need to be believed.
Here is how I run that discipline.
I start with the worry, not the message
Most communications briefs open with what the organization wants to say. That brief is already behind. The useful version opens with what the audience is quietly afraid of, and the only reliable way to find that out is to ask them.
So I lead with voice-of-the-customer research, stakeholder interviews and whatever listening infrastructure the organization will let me build. Then I write toward the worry. Campaigns that address a real fear outperform campaigns that address a business objective, every time, and they tend to be shorter.
I build teams that survive bad weeks
At JEA I grew a team from two professionals to 11. The growth was not the hard part. The hard part was making sure the team produced the same quality on the week when two people were out and a hurricane was crossing the Gulf (true story times … three?).
Resilience is a design choice, not a personality trait. I cross-train so no single channel has one point of failure. I document decisions, not just deliverables, so anyone picking up the work understands why the last version looked the way it did. I set escalation paths in advance, in daylight, so nobody has to invent one at midnight. When the pressure comes, and in utilities and healthcare it always comes, the team leans on the system or the playbook instead of on adrenaline.
I hire for curiosity over credentials. Curiosity is the trait I cannot teach. Nearly everything else I can.
I treat trust as an operating decision
Trust is not a value I put on a slide. It’s a choice about how many approval layers sit between a smart person and her finished work, and the honest answer at most organizations is too many.
My approach: give people the full context, the real constraint and the actual deadline, then step back. I would rather review a finished draft I disagree with than approve an outline that produces something safe. The first conversation is more useful, and the person learns more from it.
That means coaching rather than correcting, in the moment, with the reasoning attached. It means the person who did the work presents the work. It means credit is specific and public, and criticism is specific and private. And it means when something goes wrong, and it will, the first question is what the process missed, not who missed it.
Teams that expect to be trusted take calculated risks. Teams that expect to be second-guessed produce work engineered to survive review, which is a different and much worse goal than producing work that moves someone.
I use AI where it provides real value
I use AI tools daily, and I keep an honest ledger of where they help.
They are genuinely good at research synthesis, stakeholder and segment analysis, structuring a channel plan, generating variants to test, and putting structure on a blank page. They give a small team the analytical reach of a much larger one, which is the real argument for them.
They are not good at judgment when the stakes are high and the tone is delicate. I do not hand a crisis statement to a model. I do not let AI be the last set of eyes on anything that goes to a customer, a regulator or a grieving family. And I do not use it to replace the relationship work, because the value in this job has always been in the rooms, not the drafts.
With teams, I set three norms: disclose where you used it, verify every fact it hands you, and never let it write in a voice you would not defend in a meeting. I also insist that the time AI saves goes back to the people who saved it, in the form of work only they can do. Used as a headcount argument, these tools make communications worse. Used as a leverage argument, they make it considerably better.
I measure what actually matters
Impressions are a proxy for a proxy. I would rather report on whether people trusted us more.
At JEA that meant a top national J.D. Power ranking for digital customer experience and a mobile app that nearly 8,000 customers have rated 4.9 stars*. Those numbers are hard to move and hard to fake, which is exactly why they’re worth reporting.
What I ask of an organization
I work best where I can hear the decision being made rather than get briefed on it afterward. Communications leaders who arrive at the end of the process can only decorate it or worse, second-guess it.
I also ask for permission to say the uncomfortable thing once, in the room, on the record. After that I commit to the decision and execute it fully, whatever it was. Every organization I have helped build trust with the outside world was one that could tolerate a small amount of friction on the inside.
Want to talk about whether that fits what you’re building? Schedule a conversation with me.
* Figures as of August 2026