AI

When to use AI for design (and when to bring in a human)

Tools like Claude Design are impressive. They generate polished interfaces quickly, but polished output is not the same as clear thinking.

Where AI design tools help, where they fall short, and why UX strategy still saves time, money, and rework.

21 April 20268 min read

In short

Where AI design tools help, where they fall short, and why UX strategy still saves time, money, and rework.

The illusion of momentum

When screens start materialising quickly, it feels like progress.

In reality, you have accelerated the easy part. The hard part is understanding what those screens actually need to do, and that part remains untouched.

What decisions are you asking the user to make?

Where do journeys break down?

What shaped this experience before anyone opened a design tool?

AI does not answer those questions. It works with what it is given. Give it shallow thinking, and you get a polished of shallow thinking back.

AI can accelerate output, but it does not replace the work of understanding what the experience actually needs to do.

What these tools genuinely cannot do

Claude Design will not challenge the brief. It will not question whether a journey should exist at all. It has no visibility into , organisational , or the competing priorities that quietly shape every experience.

This plays out in predictable ways. At Co-op Bank, the problem was not visual. It was how journeys had been distorted over time by internal and policy.

At the NHS, no amount of better-looking pages would fix hundreds of structurally inconsistent .

In both cases, the was a symptom. The structure was the problem.

Generating new screens would have made things look better. It would not have made them work better.

Key takeaway

If the underlying structure is weak, AI will usually make the problem look better before it makes it work better.

Where AI fits, and where it does not

These tools earn their place when the thinking is already done.

Visualising a defined idea, exploring directions quickly, and executing low-risk work are all genuine contributions.

They fall short when the problem is still fuzzy, journeys are complex, are misaligned, or the work spans and teams.

At that point, you do not need faster output. You need clearer thinking.

Why hiring a UX consultant saves time and money in the long run

Most design cost does not come from building the wrong thing once. It comes from discovering it was wrong after launch and then rebuilding it.

A UX consultant's job is to make sure that does not happen.

The real cost of skipping UX rarely shows up neatly on a design budget. It shows up in higher development costs, lower , more support demand, slower , and launches that never quite solve the right problem.

built on unclear requirements get rebuilt. A confusing journey does not just frustrate users, it loses them. Ambiguous UX creates ambiguous tickets, which creates delay, uncertainty, and rework.

The most expensive outcome is shipping something that does not solve the right problem and having to start again. It happens more often than most organisations admit.

The cost of good UX usually looks small compared with the cost of fixing the wrong decisions later.

What a UX consultant actually does

A UX consultant does not just design screens. The value is in what happens before that.

They identify where and why users are dropping off or struggling. They map existing journeys to surface where the real sits. They align around a shared understanding of the problem before solutions are explored.

They translate and business into a clear design direction. And they make sure what gets built is solving the right problem, not just the most visible one.

That kind of structured thinking at the start of a project consistently reduces costly changes later. It also tends to surface quick wins: low-effort, high-impact improvements that can be made before any large-scale redesign is needed.

When to bring in a UX consultant

The earlier, the better. But there are moments where the value becomes especially obvious.

When you are planning a new product or and want to validate the approach before committing to development.

When you have launched something and it is not performing the way you expected.

When your product has grown organically and the experience has become inconsistent or hard to navigate.

When you are about to invest in a significant rebuild and want to make sure you are rebuilding the right thing.

When internal teams are misaligned on what users actually need.

In each of these situations, the cost of a consultant is usually a fraction of what gets spent fixing decisions that were made without one.

UX consultant vs in-house designer

In-house designers are valuable. They bring continuity, , and a deep understanding of the product over time.

But they are also close to it, which can make it genuinely difficult to question foundational decisions or challenge how things have always been done.

A UX consultant brings external perspective. They have seen how similar problems have been solved across different industries and organisations. They are not invested in preserving the existing approach, which makes it easier to identify what is actually causing the problem rather than working around it.

For complex challenges, audits, or moments of significant change, an external consultant will often move faster and more objectively than an internal team working on familiar ground.

The principle holds, regardless of how good the tools get

AI will keep improving. Design tools will become more capable, more integrated, and more seamless.

None of that changes the sequence that matters: understanding first, output second.

Start with output and you risk producing a more polished of the original problem. Start with understanding and everything that follows, including the AI-assisted parts, becomes more useful.

Good design has always been about how something works. Most of that work happens before any tool gets involved.

LET'S WORK TOGETHER

Ready to improve your product?

UX, research and product leadership for teams tackling complex digital services. The work usually starts where things have become harder than they need to be: unclear journeys, inconsistent products, competing priorities, or teams trying to move forward without a clear direction. I help simplify the problem, shape the right next step, and turn complexity into something people can actually use.

Previous feedback

Will Parkhouse

Senior Content Designer

01/20