Clarity Out of Ambiguity

In 2011, Larry Page became CEO of Google again and did something that sounds small and turned out not to be: he dissolved the company's functional organisation. Product managers, who used to be evaluated largely on the size of what they owned, suddenly needed a new yardstick — because "how big is your area" was already a flawed way to measure someone, and once the old structure was gone, it stopped working entirely. Shishir Mehrotra, who went on to lead product at YouTube and later co-found Coda, was one of the people who had to solve that problem. What he came up with is, I think, the clearest framework I've encountered for what a career actually is. I use it constantly, and I wanted to write down why.

Problem, Solution, How, Execution

The framework is usually just called PSHE, and it's built on a genuinely simple observation: any piece of work has four parts — someone has to name the Problem worth solving, propose a Solution, work out How to build it, and then actually Execute. Almost every job hands you some of those four and expects you to supply the rest. The size of your job isn't really about headcount or budget. It's about how many of the four letters land on your desk.

I like this because it explains something that used to bug me about the usual career-ladder language — "scope," "impact," "seniority" — which always felt like it was measuring the wrong axis. PSHE measures the right one: not how much you're doing, but how much of the thinking is yours. At the bottom of the curve, someone hands you a clear problem, a clear solution, and a clear plan, and the job — a real, honourable job — is making it work. At the top, nobody hands you anything except a space to look at, and the entire job is deciding what's worth solving in the first place. Mehrotra's own line for it has stuck with me since I first read it: "the best people create clarity out of ambiguity." Not the loudest people, not the ones with the biggest team. The ones who can walk into a fog and come out with a plan.

The billion-dollar version of this

The story that made the framework click for me isn't abstract. Lane Shackleton — who started in sales at YouTube and now runs product at Coda — was handed something close to a pure Problem-level assignment: figure out what to do about pre-roll ads that everyone, viewers included, quietly hated. There was no brief telling him to build a skippable ad format. He worked that out himself, then had to work out the how, then had to sell a plan that a fair amount of the internal sales organisation actively resisted, on the reasonable-sounding grounds that giving viewers a skip button seemed like giving away the product. He built it anyway. Skippable ads — TrueView — went from nothing to roughly a billion dollars in revenue within about eighteen months, and the format has quietly become the default shape of video advertising since. That's what the top of the curve looks like when it works: not a bigger title, but someone willing to own the whole fog, including the part where people tell you you're wrong.

Where I've actually operated on that curve

I want to be direct about my own version of this, because it's the part of the essay that's actually about me, and I'd rather show the trajectory plainly than describe it in adjectives.

Early on, the job was straightforwardly Execution — a clear brief, a clear architecture, build it well. I don't say that as a modest disclaimer; that stage matters, and doing it properly is its own skill most people undervalue once they've left it behind. My MSc research nudged me a rung up: nobody hands you the solution to "can a webcam read a student's attention in real time" — you're given the problem, and the solution and the how are entirely yours to find, including the several ways it doesn't work before the one that does.

The clearest example, though, is the KPN developer portal. Nobody handed me a finished product spec for that — it started as an internal incubator project, which is a polite way of saying: here's a space, something might be worth building in it, go find out. Working out what developers actually needed, what the portal should be, and building the plan that took it from that uncertain starting point to meaningful revenue was a Solution-and-How job wearing a Problem-shaped hat, and it's the piece of work I'd point to first if someone asked me where I actually operate on this curve. Co-founding Wynoot is the fuller version of the same thing — nobody is handing me the problem anymore. Finding the right one, and being willing to be wrong about it in public a few times before I'm right, is the entire job now.

What I think this actually says about leadership

I don't think leadership is a separate skill from everything above — I think it's what naturally shows up once you're operating high enough on this curve that other people's Execution now depends on the clarity of your How, or your How depends on the clarity of your Problem. You don't get handed authority for reaching the top of the curve. You get handed the fog, and your job is to be the person who comes back out of it with something the rest of the team can actually build against. That's a quieter definition of leadership than the usual one, and I think it's the truer one — closer to cartography than to command.

The honest, slightly funny thing about all this is that nobody hands you a certificate for moving up the curve. You mostly notice in retrospect, looking back at a stretch of work and realising the questions you were being asked had quietly changed from "can you build this" to "what should we build" — which is a much better problem to have, and a considerably less comfortable one.

Sources & further reading: Shishir Mehrotra, interviewed by Peter Yang, "The Best Framework for Career Growth," Creator Economy (creatoreconomy.so) — the primary source for the PSHE framework, its origin at Google in 2011 following Larry Page's dissolution of the functional organisation, and the Lane Shackleton / YouTube skippable-ads story. Figures on TrueView's revenue and timeline are as recounted in that interview.