Impact Over Noise: On the Name Drop Podcast with Molly Baker

If it doesn’t hurt, you’re not prioritizing

“If it doesn’t hurt, you’re not prioritizing. So I need to be discerning and say, all right, this is a cool idea, but it doesn’t necessarily fit us.”

The opening line of the episode, and the through line for the rest of it. Most marketing leaders are not bad at recognizing good ideas. They are bad at saying no to good ideas that do not fit. That is the job. There are infinite cool things to do on social, in AI, in content. The hard part is the discernment.

When I am thinking about my own ventures or working with a client, the question I keep asking is: is this just fun, or is it really going to make a business impact? Most of the time the answer is honest. Most of the time the answer is not what people want to hear. The discomfort of saying no is the signal that the prioritization is real.

The sales lesson that humbled me

“Who’s the key decision maker? What’s the timeline? What’s the budget? Who are the competitors?”

Years ago, I came back from a client meeting at a big media company feeling great about myself. I had controlled the room. The meeting ran an hour and a half over because they were enjoying the conversation. I sat down in Brian Wiener’s office to debrief and told him how well it went. He said, “No.” Then he asked me four questions. I did not know any of the answers.

That one moment was years of learning compressed into a single conversation. Sales is not performance. It is information gathering. Active listening. Knowing the room is yours even when someone tries to pull you down a side quest. Leaving 10 to 15 minutes at the end so questions become follow-ups and follow-ups become deals.

That lesson translates beyond sales. The same principles apply when you walk into a stakeholder meeting, a board update, or any conversation where the goal is to advance something. You can impress people all day. If you do not know the timeline, the budget, and the decision maker, you do not have a deal.

Don’t ever outsource your thinking

“Don’t ever outsource your thinking. Things that I’m passionate about, there are times when I’m working on something and I’m like, I probably could use AI for this. I’m like, don’t.”

I use AI all day, every day. Vibe coding, branding, image generation, copywriting, drafting. I have so many subscriptions I should probably cancel half. But I draw a line: the things I am genuinely passionate about, I do not outsource. I love what I do. If I let a model do all of it, I stop loving it. And at 41, I am very aware that if I outsource my thinking long enough, I will lose the ability to do the work itself.

The trick I have settled on is using AI to argue against me, not to praise me. Once I have an idea, I prompt the model to red-team it: tell me why this is wrong, where it falls apart, what I am missing. The default behavior of most assistants is to tell you you are brilliant. That is the worst thing they can do for your judgment.

If you want to see this play out in tools, I have shipped a handful of free ones built on the same philosophy: AI as a draft and ideation layer with the user still in the driver’s seat. The AI policy generator and the marketing text generator both work that way: AI fills the blank page, you do the editing and judgment. That is the right division of labor.

The fallacy of AI’s exponential growth

“I think there’s a fallacy that AI will continue to get better. The people telling you that are probably the ones selling you the AI.”

This is the contrarian piece, and the one I expect to get pushback on. AI has gotten remarkably better in the last year. That is true. The leap from where we were in 2023 to where we are in 2026 is real. What I am skeptical about is the assumption baked into most enterprise AI strategy decks: that this rate of improvement continues indefinitely.

The people most aggressively predicting that are usually the ones with the largest financial stake in the prediction being true. Foundation model companies, infrastructure providers, AI consultancies. They need it to be true. That does not make it false. But it does mean the prediction deserves more skepticism than it gets. Very few things in history have had genuine, sustained exponential growth, and most predictions of exponential growth have been wrong.

The practical consequence: I see companies wasting real money assuming AI will solve a problem in 18 months that they could solve today by hiring two more humans. The humans would deliver. The bet on AI may not. If I am wrong about this, I will come back on the podcast and say so. But I am not investing my own ventures or my clients’ roadmaps as if exponential growth is a given.

Study psychology, not business

“Tech changes; humans don’t. Same emotional drivers, same thinking processes. If you understand people, you’ll be good.”

When Molly asked what marketers and leaders should be thinking about right now, my honest answer was: people, philosophy, and the humanities. Technology has always changed and will keep changing. The platforms come and go. The tools cycle through. What people respond to has stayed almost identical for as long as we have records.

Especially in the AI era. If models really do replace categories of jobs, the people who remain will be the ones who understand what motivates and scares the humans on the other side. Change management. Buyer psychology. Why teams adopt or reject new tools. Sales itself comes down to understanding what someone is actually afraid of and what they are trying to win. Ad creative is the same. Brand strategy is the same.

So when young marketers ask me what to study, I keep saying study less business and more psychology. The marketing playbook can be learned in six months. The human stuff takes a career, and it is the part that compounds.