The portfolio talks about unifying the workflow automation system. This is the meta-story: what it took to ship anything inside a mature, high-craft design org without breaking it.
Companion teardown → the workflow builder, rebuilt end to end: four design moves and the one I'd changemonday.com had a sophisticated design org. Hundreds of designers, a mature design system, strong opinions about how design ships. Workflow automation was technically dense — automations and integrations had grown as separate products, and the team had been trying to unify them for years. The path was littered with failed attempts.
I came in as a Product Design Team Lead from a Director seat at a smaller, faster org. The skill gap I was paying attention to was not "can I design this." It was "can I earn the right to ship this here."
Team Lead for the Workflows area. We were a small team of designers, and I was one of them: leading and designing alongside the team, not managing it from a step removed. I owned the design direction for automations, integrations, and the new logic editor that was supposed to unify them. I didn't own the design system, but I had to design fluently inside it.
I also didn't own the politics. But the politics were real, and the people who'd tried this work before were still around.
Stop trying to rebuild from scratch. Find what's already working and earn the right to change it. In a mature org, the gravitational pull is to come in as the new senior person with a Big Reframe. That's the move that usually doesn't ship.
I bet that the unified mental model could ship if I let the existing patterns do most of the work and only changed what genuinely needed changing.
I read the design system end to end. Sat in three other teams' critiques. Shadowed an engineer for a day. The temptation was to start sketching workflow ideas. The right call was to internalize the conventions until I could break them on purpose, not by accident.
The product instinct was to ship the new model as the new default. I argued against it. Users with running automations would hit the new editor cold and resent it. We let users opt in, with clear migration paths. That single decision was the difference between adoption and churn. (It's also why the adoption number landed where it did.)
The engineering team had a list of "constraints" that turned out to be features. Trigger-based execution was easier than they made it sound, but only if the UI committed to a strict order of operations. We agreed on the order before any pixels. Half the design problem was already solved.
I pushed to add a "templates marketplace," the thing that would make the unified workflow builder usable from day one. A senior stakeholder wanted it cut entirely to de-risk the launch. I made the case for why it mattered, named the real tradeoff, and instead of letting it get dropped, won it as a committed fast follow: the core unification shipped on time, the marketplace a quarter later, cleanly. Fighting for the right scope rather than quietly accepting the cut is the part of the work I'm most proud of.
~15% increase in feature adoption in the most technically dense product area in monday.com. The unified logic editor became the most-used new capability in the area within two quarters. The portfolio has more on the design itself; what I'd add here is that the adoption came mostly from migration — people moving voluntarily, not new users defaulted in.
I onboarded too quietly. I was so focused on learning the conventions that I didn't ship a small visible thing in the first month. A short, opinionated piece of work early would've earned me trust faster, and made the harder conversations later (like the marketplace pushback) less expensive.
Joining a mature org as a senior leader is a craft of restraint. But restraint and invisibility aren't the same thing.