When I wrote about the forward-deployed designer squad model earlier this year, I was working from the outside in: what the model should look like, who it serves, why it matters. Ron Bronson ran it for four years as director of a 40-person design division at 18F, the now-defunct US government’s in-house digital services agency. His post is the inside view and he diagnoses why most orgs never get there:
The real reasons that design roles aren’t being considered for this is the ways orgs constrain how designers show up on cross-functional teams. If your designers are only good for handoffs, you’re not going to invest in the headcount.
The people are the key, but you have to be opinionated about what you’re looking for your designers to do. If you’re looking for pixel-perfect, portfolio polish then you’re doing it wrong. Due to the quirks of federal hiring rules, we weren’t allowed to consider portfolios. It didn’t mean we couldn’t look at them, they just couldn’t be part of the criteria someone got an offer or not.
Take the portfolio rule: federal hiring restrictions sound like the kind of constraint that makes a practice worse, and instead they forced 18F to evaluate designers on the things that actually predict forward-deployed performance—ambiguity tolerance, collaboration, low ego, willingness to work in the open. The portfolio gauntlet that dominates tech-industry design hiring optimizes for the opposite skill: producing pixel-perfect artifacts in isolation. Bronson’s team got better signal because they were prevented from looking at the worse one.
Bronson on the multidisciplinary bar:
hired designers who can do more than one thing. Some impressive UX researchers would show up on our doorstep often, and if they talked to me, I’d be very direct with them about how we worked and that our designers often had to wear more than one hat out of necessity. The other constraint? Headcount. Design often has to justify itself more than other practices, so we couldn’t afford people who were too “special” to be staffed to a broad array of partner engagements. What this meant in practice? Designers who could code, researchers with content strategy & information architecture chops, service designers who could lead and/or PM projects, and every designer being a strategist on some level.
Generalist breadth in this context is a structural requirement of the engagement. That’s what Bronson means by “wear more than one hat out of necessity.” You can’t deploy a specialist into a six-week problem-scoping sprint and expect them to be useful for more than one week of it.
Bronson on where designers should sit:
As I explained in Design as Repair at IxDA Oslo last September: we need designers embedded where problems happened, not downstream after it’s been scoped, broken and all the framing has been done and asked to execute.
Most design orgs are structurally downstream: invited in after PM and engineering have already decided what’s being built, given a brief that pre-resolves the questions design should be asking. Bronson’s 18F was built to refuse that posture by default, which is why the model worked there before it had a name.
