AI Won't Fix Your UX Bottleneck (But These 3 Things Might)

During an interview last week, the discussion turned to tracking UX department efficiency. The company was very concerned that the design department needed to move faster and was looking for specific metrics to track UX, outside of engineering and product metrics (they were already "doing fine"), to highlight where the issues were. I pushed a little and correctly guessed that what they were really after was an easy way to see where they could plug AI into the design flow to make things move faster. I think there are some leaders who think we can just speak a design into being and BOOM site done!

Sigh.

Throwing "AI" at problems isn't going to solve them. There are some very useful newish tools available right now, but there are no magic wands. If there's a pinch in the productivity cycle, here are some things that might help:

Listening

There's how things are supposed to work, and how they're actually working. Maybe there's a team doing the "smile and nod" during refinement instead of speaking up with concerns. Maybe there's a designer who's obsessed with pixel-perfect, but that's not what the dev team needs to start. Maybe there are unclear expectations all around. Ask questions, get curious, and listen to understand before making changes.

Data

Only tracking the timing of design work doesn't give you all of the context you need, especially if you don't have a solid integration of UX into your entire department's success metric framework. Are you giving ample time to include research? Does design make estimates of their work against which efficiency success rates can be compared? Are your developers empowered to provide feedback to design early, or are they set up for a "handoff," which can create time-consuming back and forth? Metrics in a vacuum are just numbers, not information.

Upskilling

It's not magic, but there are a lot of ways that the tools we have available now can take time off of our hands and help us all move faster. But like anything new, we have to learn how to use it in general, then understand how it can solve our particular problems, and then make sure that our leadership understands how and why we're using it in this way. That last piece is crucial - your leaders need to know that you are stepping up and you need them to let you do your thing.

You're hearing a lot about AI right now, and that's probably not going away anytime soon. But you can be part of shifting the conversation to be more productive by asking better questions. Next time someone suggests AI as the solution to a workflow problem, try asking: "What have we learned from the people doing the work about where the actual bottleneck is?" Because here's the thing - people are still at the heart of the best work. AI is just another tool, and tools are only as effective as the strategy behind using them.