A human-centered approach to AI implementation

By Carlos Del Carmen · 6 min read

AI tools are everywhere, but adding them without a plan usually means wasted subscriptions and frustrated teams. The approach that works is the same one good designers use: start with people.

In my years as a UX designer and instructor, I followed a six-stage design process. When I started implementing AI and automating workflows in business, I found the same process works just as well. Here's how I apply it.

1. Understand

Start with people, not tools. What does the team actually do every day? Where do they lose time, repeat work, or wait on someone else? Map the real workflow before thinking about AI.

2. Research

Challenge your assumptions. Talk to the people doing the work, watch how tasks get done, and survey the team. The best automation ideas usually come from the people closest to the problem.

3. Analyze

Move from “what” to “why.” Look for tasks that are repetitive, rule-based and high-volume. Those are your best candidates for automation. Tasks that need judgment or a personal touch usually aren't.

4. Design

Combine the big idea with what you learned. Sketch the new workflow, choose the right tool for it (often the one your team already uses), and build a small prototype instead of a big system.

5. Launch

Test with a small group first, gather feedback, and refine. Train people with clear, short guides. A tool nobody understands won't get used, no matter how powerful it is.

6. Monitor

After launch, keep watching. Is it saving time? Are there errors or workarounds? Measure the results, listen to feedback, and keep improving.

The takeaway

AI implementation isn't about having the newest tool. It's about understanding how people work, solving real problems, and helping your team feel confident with the change.

Looking for tools to start with? Browse my AI tools directory, or get in touch if you'd like help with your workflows.