Kickstarting UX & Product with AI
AI is dramatically reshaping our world—not just through velocity-boosting tools, but by enabling more impactful, user-centered features and functions across digital products and services. Currently, with our consulting clients, I’m observing four distinct phases in how clients are integrating AI into UX and product development, ranging from initial alignment to design and delivery.
1. AI Workshops: Sparking Alignment and Opportunity
We begin by bringing teams together and asking simple yet powerful questions: Where is AI already helping? What challenges would we love AI to tackle? Through structured workshops—starting with discovery surveys and moving into collaborative prioritization—we help teams identify high-impact AI opportunities. Where are the processes that could best use AI/Agents to create a more seamless experience? Participants map out “lighthouse” projects: small, strategic pilots designed to prove value quickly and build momentum. By the end, organizations emerge aligned, energized, and ready with clear, actionable paths for AI-powered initiatives.
2. Directional AI‑Augmented Research: Speeding Up Ideation
AI excels at rapid idea generation. We’ve leveraged it early in the process to draft AI-Assumption‑Based personas, run JTBD frameworks, compare competitors, and even develop interview guides. This isn’t about relying on AI for final answers; it’s about harnessing it as a creative accelerator. Our teams treat AI-generated insights as drafts: helpful starting points to refine with human expertise. As long as it’s positioned as a starting point and not the final authority, this early input speeds things up, helping us explore angles quickly before diving into true qualitative/quantitative research.
The idea is to use AI to enhance your research, not replace it.
3. Data‑Driven Product Discovery: Validating Real Needs First
The next step is grounding AI in real user needs. Instead of relying on generic AI strategy decks, we utilize quantitative data analytics and usage statistics, as well as qualitative insights from various product discovery methods (journey maps, service blueprints, etc…). This discovery phase uncovers true customer or employee pain points. From there, we ask: Could AI help? (and the answer isn’t always YES!). The result is proposals like, “Our users struggle with X—and AI can deliver Y to solve it.” This direct linkage between user problem and AI solution builds credibility, focuses on ROI, and avoids vague “sprinkling AI” across product journeys.
Make sure you're linking AI to tangible, human-centered challenges.
4. Designing for AI: Building Trust Through Thoughtful UX
When the correct usage of AI is defined through research, the UX design phase begins. AI-driven experiences can require new patterns, such as explainability layers, refinement controls, fallback options, predictive suggestions, and transparent feedback loops. Interfaces might not change at all in cases where AI/agents work behind the scenes. However, even when interfaces remain familiar, they should communicate changes in AI activity. We design features that clarify what AI is doing, why it did it, and how users can guide or correct it. This fosters trust, maintains control, and supports user adoption.
A Four-Phase AI Journey That Scales
Successful AI integration follows a clear sequence:
Align through AI workshops to launch pilot projects.
Explore and accelerate with AI-aided research.
Validate using real user data and discovery.
Design thoughtful, transparent AI interactions.
This sequence accelerates design cycles and ensures AI enhances—not distracts from—the overall product experience.
Why This Approach Works
Human-Centered Focus: AI becomes a tool, not a directive.
Trust and Transparency: Users understand and shape how AI operates.
Clear ROI: Each AI effort is tied to a measurable benefit.
Scalable Framework: Teams can move confidently through each phase, adapting and learning as they go.
How About You?
Are AI workshops part of your practice? Have you used AI to spark research insights or to prototype solutions? What challenges have you faced designing AI-powered experiences?
Let’s compare notes—whether you’ve just begun experimenting or are scaling across products, I’d love to hear what’s working (and what’s not).