There’s a moment in Rick Rubin’s 60 Minutes interview that I’ve thought about for years.
Anderson Cooper asks if he plays instruments. “Barely.” If he knows how to work a soundboard. “No. I have no technical ability. And I know nothing about music.”
Cooper laughs and says he must know something.
“Well, I know what I like and what I don’t like. And I’m decisive about what I like and what I don’t like.”
“So what are you being paid for?”
“The confidence that I have in my taste, and my ability to express what I feel has proven helpful for artists.”
This is a man who has produced the Beastie Boys, Johnny Cash, Metallica and Adele. He isn’t paid to play or to engineer. He’s paid to decide.
I relate to that more than to almost anything else I’ve heard about my own job. I don’t write the code and I don’t push the pixels. What I produce is user journeys: the path a guest takes from opening an app to a finished order. Brands hire me for the same thing Rubin is describing. In product, we call it product sense.
Product sense is taste with a job to do
Product sense is the judgment to look at everything a team could do and pick what it should. In practice it comes down to four calls:
- What to do. Of the twenty things on the list, which one actually moves the guest, and the business, forward.
- When to do it. The right fix at the wrong time is still the wrong call. Some things have to wait for the data, the platform or the season. Some can’t wait at all.
- Why it matters. If I can’t say in one sentence what a change does for the guest and for the business, it isn’t ready, no matter how good it looks.
- When to say no. This is the hardest one and the most valuable. Every no protects a yes. A roadmap that says yes to everything isn’t a strategy; it’s a queue.
None of that is technical. All of it is what I’m paid for.
Taste is mostly what you take out
Rubin has called himself a reducer instead of a producer. “I like the idea of getting the point across with the least amount of information possible.”
That is the best description of good product work I know.
Almost every bad user journey I’ve seen wasn’t missing something. It had too much. A sign-up form that asked for a birthday and a zip code before a guest had ordered anything. A password, for an app people use for 90 seconds at a time. An offer that had to be added in one place and its product found in another. Each of those made sense in a meeting. Each one cost guests.
At Jack in the Box, most of the work that cut ordering time in half was subtraction. Fields came out. Passwords became magic links. The offer and the product it applies to went into the order in one step. Nobody would call any of that a feature. Together, they’re why the app worked.
Adding is easy. Every stakeholder has something they’d like in the flow. Product sense is knowing which of those things the guest would never miss, and being willing to say no to the rest.
Taste is being decisive
The part of Rubin’s answer people skip is the second half: “I’m decisive about what I like and what I don’t like.”
Plenty of people have opinions. Fewer will commit to one when the room is split, the data is incomplete and someone senior wants the other option. That’s where most product decisions actually get made. If you wait for certainty, you’ve already decided: you’ve decided to ship whatever the loudest person in the room wanted.
Decisive doesn’t mean stubborn. It means making a clear call, explaining why, and changing it when you’re wrong. A team can work with a clear call. It can’t work with a shrug.
“The audience comes last” (and where I disagree)
Later in the interview, Rubin says something that sounds like the opposite of product management: “The audience comes last.” Cooper pushes back, and Rubin explains: “The audience doesn’t know what they want. The audience only knows what’s come before.”
I don’t put the guest last. In commerce, the guest is the whole point. But I agree with the second half more than you’d expect.
Guests are very good at telling you where something hurts. They’re not good at telling you what to build. Every one-star review I read at Jack in the Box was precise about the failure and useless as a design spec. “I couldn’t use my deal” is the most honest product requirement you’ll ever get (I wrote about that here). “Add a promo code field to the cart” is a guess, based on the last app they used.
So listen to guests for the problem, and use your product sense for the answer. That’s how the app went from 2.6 to 4.8 stars: not by building what reviews asked for, but by fixing what they described.
Product sense you can’t measure is just an opinion
Here’s where product work and record making split.
Rubin’s taste gets judged by whether the album moves people. My product sense gets judged by a dashboard. Conversion, time to order, rating, revenue. That’s a gift, not a constraint. It means product sense isn’t a mystery. It’s a track record.
When a monthly T-Mobile promotion at Popeyes was converting at 3.36%, someone proposed building a new promo code service. That was a when-to-say-no moment, and I said no. We built promo codes into the offer feature the app already had and cut the flow from eight steps to six. Conversion went to 9.82%. Across the brands I’ve worked with, those decisions add up to more than $1 billion in digital sales.
That’s the honest answer to “why should I trust your product sense?” Not because I said so. Because the calls kept working, and when one didn’t, I could see it in the numbers and change it.
Why product sense matters more now
AI has made it almost free to produce things. A prototype in an afternoon, a PRD in an hour, ten versions of a checkout flow before lunch. I use those tools every day.
But when making things gets cheap, choosing gets expensive. Someone still has to look at ten checkout flows and know which one a guest will finish. Someone has to decide what comes out. Volume is now free. Quality isn’t.
That’s Rubin’s point, applied to software. The tools will keep getting better at the technical part. What people will pay for is product sense: the confidence to say this one, not that one, not yet, and here’s why.