UX rarely happens in a straight line
The UX process makes sense on paper. In my experience, clients, contracts and the point at which you join a project rarely let it unfold in that order.
I can count on a handful of projects throughout my career where I’ve worked through the full UX process, from discovery to deployment, in the order we’re usually shown.
Discovery, wireframes, testing, high fidelity, then working with engineers through production and release. It’s a useful way to explain the work. But when you’re designing products, especially as a contractor, the project you walk into rarely follows that neat sequence.
A lot depends on the client, the organisation and the point at which you’re brought in.
When you work directly with clients, discovery can be a difficult sell. In my experience, many feel they already understand their business and their competitors. They don’t necessarily want to pay a designer to spend time investigating something they believe they know.
Sometimes that comes across as arrogance. Sometimes they simply don’t see what discovery will give them that they haven’t already got. Either way, you can find yourself moving towards design much sooner than the standard process suggests.
You still need to understand the business. I’ll often do some research myself so I know what I’m talking about, understand the competitors and can have an informed conversation. But that work may never become a formal discovery phase the client recognises or values.
In a large corporate organisation, the situation is different. You’re often brought in to work on one part of a product or a particular piece of a project. Discovery may already have happened. Other decisions may already have been made. You inherit that work and start from wherever the team needs you.
As a contractor, you don’t always get to choose the starting point, and you may not be there for the finish.
That’s why I find the expectation that every designer should follow the whole process from beginning to end difficult to reconcile with my experience. The sequence makes sense on paper. In practice, your role might cover only a small section of it.
When the client can finally see it
Even when I’m working directly with a client, there’s often a gap between the work I know is necessary and the work they see value in.
A grey wireframe is a good example. As a designer, you can see the thinking behind it: the structure, the choices, how someone will move through the experience. The client may struggle to see any of that. They also may not understand how the discovery work led you there.
Depending on who you’re working with, the project only starts to feel real when it reaches high fidelity.
That can be frustrating. You’ve already done a lot of the work that makes the design hold together, but the response changes once there’s a polished screen in front of them. It can almost feel as though everything that came before has been forgotten.
The earlier thinking should still be there, informing the design. You need to keep checking it and making sure the experience makes sense. But you can’t assume the client will value every stage equally, or that they’ll recognise the work behind what they’re looking at.
AI changes how quickly we can learn
AI has changed how I think about this process again.
As a designer, I now have more power to get something working, move it towards deployment and test it quickly. The distance between an idea and something people can use feels much shorter. That gives us more opportunities to test, learn and adjust.
I think this is particularly useful when working with startups and small or medium-sized businesses, where you’re building something from the ground up. There’s room to shape the product as you learn, rather than fitting your work into a company where much of it is already established.
But before using that speed, we still need to ask the questions that give the work a purpose.
What is the problem? What is the user trying to do? What are we trying to achieve with this design?
Those questions help us decide what to build and what we need to learn from testing it. Getting something into production quickly is useful when we understand what we’re trying to find out.
A process that fits the project
Do I believe in the UX process? Yes. Do I believe it happens in a straight line in the real world? Very rarely.
Most projects I’ve worked on have involved adapting to the circumstances: a client who wants to see finished designs, a corporate team that has already completed discovery, or a contract focused on one part of a larger product.
The responsibility to understand the problem stays with us, even when the project doesn’t give that work a formal stage. So does the responsibility to question our assumptions and check whether the design helps the user do what they came to do.
I still want that thinking to be solid, whether I’m showing a wireframe, a high-fidelity design or a working product. The client may only see its value once the screen looks finished. As the designer, I need to understand how we got there.