
Stop chasing deliverables. Start answering real questions.
Designers love to talk about process.
The frameworks, the stages, the artefacts - double diamond, discovery, definition, delivery - all get referenced. Ask any designer in an interview what their process is, and you’ll usually get a well-structured answer including one of these. I’ve given that answer plenty of times, but in reality, the honest answer has always been: it depends.
The problem with “perfect” process. I’ve spent years working with design teams of different sizes - embedding designers into delivery teams, trying to balance good design practice with the reality of how businesses actually work.
At one point, I was leading a large design team - over a hundred people. We had structure, specialisms, and a well-defined process. The double diamond held up well, with clear stages, gates, and deliverables. And it worked. We had:
High-quality outputs
Clear expectations
Confidence in what we were producing
But it also relied on something important. Time, people & buy-in from the business.
Then things changed. In a smaller, faster-moving SaaS environment, that same process started to feel heavy. Too many steps. Too many dependencies. Too many artefacts. Naturally, the question came up: “Do we really need all of this?”
Some even suggested skipping discovery altogether and jumping straight to design. That’s always a red flag. But it did force a more useful question: What is the process actually for?
The turning point. I went back through it all. Every stage. Every deliverable. Every activity. And they all had value. Of course they did - they’d been built over years to protect good design practice. But the realisation came in a conversation with one of the more sceptical designers in the team. The process wasn’t valuable because of the deliverables. It was valuable because of what those deliverables gave us.
Confidence.
Each stage existed to answer a set of questions- so we knew it was safe to move forward, or when we needed to stop and rethink.
From deliverables to questions. That shift changed everything.
Instead of asking: “Have we produced the artefact?”
We started asking: “Have we answered the question?”
Questions like:
Do we actually understand the problem?
Do we know who we’re designing for?
What assumptions are we making?
What risks are we comfortable taking?
Once those were answered, the format didn’t matter as much.
What changed in practice. A few things happened quite quickly.
First, teams started talking more. Designers, engineers, and product managers had to align on answers - not just review outputs. It created better conversations and a clearer shared understanding.
Second, deliverables became more flexible. Different projects needed different levels of depth. Some required detailed research. Others didn’t. The team could adapt without feeling like they were “breaking the process”.
Third, it created momentum. Once the key questions were answered, we could move forward with confidence. It felt lighter, but not less rigorous.
The role of the designer shifts (again). This also connects back to a broader shift I’ve been thinking about.
If AI is helping us produce artefacts faster - and in many cases, to a solid standard - then the value of design moves elsewhere. Not in the deliverables. But in:
Asking the right questions
Framing the problem properly
Knowing when we have enough information
Guiding the team forward
In other words, the designer becomes less of a producer and more of an architect. Not just shaping the solution - but shaping the thinking that leads to it.
A simple framework that stuck. We now use a set of questions across the four stages of the process. Not as a checklist - but as prompts for discussion.
You know it’s working when:
A product manager brings them into a session
An engineer challenges an assumption
The team debates whether a question has really been answered
That’s when the process is doing its job.
Final thought. A well-defined design process can give you confidence. But it can also become heavy, rigid, and disconnected from how teams actually work.
Shifting from deliverables to questions keeps the intent of the process - but makes it adaptable. Because ultimately, good design isn’t about producing the right artefacts.
It’s about making the right decisions.
And you only get there by asking the right questions.
