Research built around the decision, not just the study.
This is the expanded index: four applied cases with a little more context than the homepage, followed by a research-practice lab and smaller signals of range.
When success could not prove itself.
A multi-month investigation traced the problem across customer practice, product capabilities, analytics, implementation, and organizational ownership.
How could teams make value easier to recognize, measure, and defend?
Customer evidence + internal landscape
Value-proof model + intervention map
Designing customer research at scale.
Interconnected activities captured measurement priorities, adoption barriers, AI readiness, capability use, and the organizational conditions behind each signal.
Which customer patterns needed product attention, readiness work, or a stronger measurement path?
Five linked participatory activities
Governed workbook + decision views
“I need help with a leave question.”
I can help clarify that. Before we continue…
Designing the handoff between AI and people.
The work moved beyond demo appeal to the operating conditions required for a voice assistant to resolve, transfer, recover, and remain accountable.
What would make a conversational assistant useful, recoverable, and accountable in real work?
Field reactions + workflow probes
Handoff packet + readiness model
Product response Short flow · visible status · safe exit
Designing beyond desk-based assumptions.
Foundational customer research developed into ecosystem mapping, kiosk concepts, authentication questions, guided help, and release-oriented validation.
What conditions make digital help genuinely reachable during the realities of frontline work?
Longitudinal discovery → validation
Access model + experience concepts
Research that remembers.
A living model of research memory—connecting source material, decisions, relationships, and bounded AI so teams can query evidence, preserve context, and carry trustworthy knowledge forward.
Research enters through many sources. The system keeps each one attached to its context, sensitivity, and reliability.
Smaller stories that add range.
Open a card for a concise description of the question, method, and capability it demonstrates.