Moving upstream: designers in product decisions
How AI frees designers to shape what gets built, not just how it looks.

Zander Whitehurst
8 min read
For most of my career, designers were handed problems that had already been decided. The roadmap was set, the feature was agreed and our job was to make it usable and make it look good. AI is quietly changing that. When the production side of design gets faster, the time we used to spend pushing pixels can go somewhere far more valuable: deciding what gets built in the first place.
That is what moving upstream means. It is not about abandoning craft. It is about bringing your craft to the earlier conversations, where the biggest decisions are made and where good design thinking has the most leverage.
Why AI makes this possible now
When a prototype takes a day instead of a fortnight, the cost of exploring an idea drops dramatically. That changes who can contribute to product decisions and when. A designer who can turn a rough idea into something people can click through by the afternoon becomes incredibly useful in a planning meeting.
It also changes what teams need from designers. If AI can produce a reasonable interface from a description, the scarce skill is knowing which interface should exist, for whom and why.
Start by framing problems well
The most valuable thing a designer can bring to a product conversation is a well framed problem. Teams often jump straight to solutions because solutions feel like progress. A clear problem statement slows that down in a useful way.
A good problem frame usually covers:
- Who is affected. A specific group of users, not everyone.
- What is going wrong. The behaviour or pain you have actually observed.
- Why it matters. The impact on users and on the business.
- What success looks like. How you would know the problem had been solved.
I remember sitting in a planning session a few years ago where the team was debating three competing features. I asked, rather nervously, which user problem each one solved. The room went quiet. Two of the three features were solving the same problem in different ways, and nobody had noticed. That one question saved us weeks of work, and it changed how the PM saw my role from that day on.
Join discovery and roadmap conversations
You cannot shape decisions you are not in the room for. If discovery and roadmap planning happen without you, ask to be included. Most PMs will say yes, especially if you offer something useful in return.
Come prepared. Bring a synthesis of recent research, a quick prototype that tests an assumption or a map of where users struggle today. AI is brilliant for this kind of preparation. You can synthesise interview notes in minutes, sketch out several directions in code and arrive with something concrete to react to.
Once you are in the room, listen more than you speak at first. Learn how decisions are actually made and what pressures the team is under.
Build a real partnership with your PM
Moving upstream does not mean taking over the PM's job. The best results come from a genuine partnership, where the PM owns the business case and priorities and the designer owns the understanding of users and the quality of the experience. There is plenty of overlap, and that overlap is where the good work happens.
A few habits help:
- Share early and often. Show half formed ideas rather than waiting for polish.
- Ask about their goals. Understand what they are being measured on and what keeps them up at night.
- Offer prototypes as decision tools. A working prototype can settle a debate faster than any slide deck.
- Disagree in private, align in public. Work through tension one to one, then present a united front.
Build business literacy
Designers sometimes lose influence because they talk only about users while everyone else talks about revenue, retention and cost. You do not need an MBA. You do need to understand how your product makes money and what the business is trying to achieve this year.
Read your company's strategy documents. Ask your PM which metrics matter most. Sit in on a sales call or a support shift. Learn the language of the people making decisions, then connect your design work to it. When you can explain how a better onboarding flow affects activation, people listen differently.
How to start this week
You do not need permission or a new job title to begin. Small, consistent moves build credibility quickly.
- Ask one framing question. In your next planning or kickoff meeting, ask which user problem the work solves and how you will know it worked.
- Prototype an assumption. Pick a risky idea on the roadmap and build a quick prototype to test it.
- Book time with your PM. Ask what they are worried about this quarter and how design could help.
- Learn one metric. Find out which number your team cares about most and how your current work affects it.
When I first started doing this, I felt like I was overstepping. I was not. Most teams are crying out for someone who can connect user needs to business goals and make ideas tangible quickly. Designers with AI in their toolkit are perfectly placed to be that person.
The shift is already happening
The designers I see thriving right now are not the ones producing the most screens. They are the ones helping their teams decide what is worth building. AI gives you the time and the tools to do that. The rest is having the confidence to step into the room and ask better questions. If you want to learn the product side properly, the AI Product Management Bootcamp covers discovery, PRDs and prioritisation.
Memorisely