For years I've argued that design gets a seat at the table by learning to speak in money and risk. I still believe that.
Not because money is the only thing that matters. Because it's the language the business already uses to make decisions, and design keeps asking to be judged in a different one.
The way I see it, a company is basically a machine for making bets. What to build, who it's for, how much to spend. Design's real job is to make those bets better before the money's committed. The screens and the prototypes are the vehicle. They were never the job.
We took the wrong lesson from Apple
Designers love to hold up Apple and Jony Ive as the patron saints of good design winning. But design didn't earn its seat at Apple because of the beveled corners.
It earned it because Apple has sold more than 3 billion iPhones and spent years as the most valuable company on earth, plus executive sponsorship from Steve. The craft was real. But the authority came from being downstream of staggering business results.
So the lesson most designers took from Apple is backwards. It's not make it beautiful and you'll get respect. It's get tied to business impact and you'll get a seat. Aesthetics were the vehicle. Money was the ticket.
The $3M bet
Most design teams aren't stuck because the work is bad. They're stuck because of where they sit. Design shows up after the decision's made, so it ends up decorating outputs, and you can't argue your way out of that by explaining how valuable design is. Everyone nods and nothing moves.
We're a little complicit, too. If you keep accepting tickets and specs, you train everyone to see you as a service desk. The way out is getting into discovery before the solution's locked, so you're shaping the bet instead of the pixels, and owning the problem and the risk instead of just the interface. Designers are hired to be the architects of the building, not the painter who shows up after it's built.
What actually changes the conversation is the language. A design director I know put it perfectly. She stopped saying "we need more research" and started saying "we're making a $3M bet on an assumption we haven't tested." Same request. But leadership funds the second one, because now she's not a designer asking for time, she's someone protecting the business from a bad bet.
The first version asks the org to value what design values. The second meets the org where it already makes decisions. Nobody has to be converted.
And it's not just a better line, there's real math under it. Douglas Hubbard's How to Measure Anything boils the value of information down to roughly the chance you're wrong times the cost of being wrong. "A $3M bet on an assumption we haven't tested" is that calculation said out loud in a meeting. Hubbard also points out that the uncertainties that matter most to a decision tend to be the ones nobody bothers to measure, and the untested assumption under a big bet is usually one of them.
One move, three directions
I've come to think it's one move, pointed in different directions.
Point it forward, at a bet you're about to place, and the question is what it costs if the assumption under it is wrong. That's the $3M framing. It buys you discovery time.
Point it at a problem nobody's solving and the question is what it's costing right now. Support load, rework, features nobody uses, building the wrong thing twice. Don Reinertsen has been saying for years that if you only quantify one thing, quantify the cost of delay, and in a 2011 interview he put the share of product developers who don't know it for their own projects at 85 percent. If design can put a number on the problem nobody's working on, that buys roadmap position.
Point it inward, at how the org works, and it gets a little uncomfortable. I once watched three senior people review a Jira ticket whose entire purpose was confirming that a date stamp updated. Don't make me get out a calculator and figure out what that cost the company, because I'll bet you it's approaching $1,000. For a date stamp. Shopify would probably back me up on that number. When it built a meeting-cost calculator into its calendars in 2023, the company put a typical half-hour meeting with three people somewhere between $700 and $1,600. Then there's the retro that cost $1,500 to conclude "we need to communicate better." That's not an answer. It's a question without a question mark.
Same tool every time. Take the design concern and put it in the unit the business already thinks in. The direction changes, the move doesn't.
How I actually measure it
This doesn't mean running design off a spreadsheet. I pitch in dollars because that's leadership's language. I work by laddering down to something a team can actually change.
In practice that means four kinds of measures, because they answer different questions for different people. The UX outcome is whose life gets better and how, in the user's own words. The success metric is the behavioral finish line, what done actually means. Progress metrics tell you how far along you are and give other teams something to line up behind. Problem value is what the thing costs while it stays broken, in dollars. That last one is what makes UX legible to people who think in money, because it lets leadership compare it against every other cost in the org.
The dollar figure tells you why it matters. What you actually steer by is the task-level stuff, time on task, completion rate, error rate, drop-off at a step, because those move the moment you change the design. The dollar figure catches up later.
Someone always says that's a business metric, not a UX metric. Yes. That's the point. A bad experience always shows up somewhere in the org's finances. Translating UX into dollars isn't leaving UX behind, it's proving UX was a business lever the whole time.
Where this goes wrong
The fastest way to lose all of this is to win it with bad numbers.
Our field has a favorite stat, that every dollar put into UX comes back a hundredfold. You've probably seen it on a slide. It gets pinned on Forrester constantly, and nobody seems able to trace it to an actual Forrester study. The one about bugs costing 100 times more to fix after release has the same problem: the IBM research it supposedly comes from has never turned up. Even the solid research, like McKinsey's design index, shows that companies good at design grow faster. That's a correlation, not proof that design did the growing.
Overclaiming is how design loses credibility with the people who can actually check, and somebody in finance always checks. Design rarely gets a clean causal number, so separate what it can claim directly from what it only contributed to. A boring local number you can defend, like what that date-stamp review cost, beats a famous one you can't.
There's a failure mode on the other side, and it has a name: the McNamara fallacy. You measure what's easy, give everything else an arbitrary number, decide the unmeasured stuff doesn't matter, and eventually decide it doesn't exist. A design team that puts a dollar sign on everything ends up there. Some work is just plumbing, and inventing a success metric for plumbing is worse than admitting it's plumbing. The math cuts both ways, too. If being wrong is cheap, the research isn't worth much either, and being willing to say so is part of what makes your $3M number believable.
You also can't research your way to a transformative bet. Data describes the world that already exists. At some point it's a conviction call, and you have to be willing to be wrong in a way incremental work never asks of you. Even Steve, everyone's favorite example of not asking customers, told BusinessWeek in 1998, "we have a lot of research into our installed base." His point was narrower than the legend. Focus groups can't design a complicated product for you, that's all.
So when you make a call the data doesn't support, say it out loud. "I'm making a bet the data doesn't support, and here's why" is defensible. Quietly ignoring inconvenient numbers while calling yourself data-driven isn't.
The craft still matters. It was always the vehicle. But design didn't get its seat at Apple by being beautiful in a language the business couldn't read, and it won't get one anywhere else that way either.
Money was the ticket. It still is.