The problem
Keep conversations connected to what happens next.
I want to build a history of conversations with users that helps me revisit what was said, when it was said, and when it was addressed. This tool is part of improving research processes so they can create value for my colleagues as well as myself.
A concrete example of the difficulty in revisiting a conversation or tracking whether feedback was addressed.
Who it affects
The project is intended to improve research processes and create value for my colleagues as well as myself. It deals with conversations with users and the history of what was said and addressed.
Who would use the tool, how they currently retrieve or follow up on feedback, and what information they need to trust the history.
Decisions & tradeoffs
From a conversation to a useful history.
Transcription and redaction
Transcribe conversations with users and redact sensitive information as part of processing the material.
Takeaways from conversations
Process the conversation into takeaways I can return to as the work develops.
A timeline of feedback and follow-through
Track when something was said and when it was addressed in a timeline view.
Search across the research history
Find a specific study, quote, or person, and return to the relevant conversation history.
An important design decision about redaction, takeaways, the timeline, or search. Explain the options, the tradeoff, and why I chose one. The capabilities above describe the intended direction.
Progress & learning
A project I’m actively developing.
The project addresses an inefficiency I identified in the research workflow, with the aim of making user conversations easier to revisit over time and creating value beyond my own tasks.
Which capabilities work today, what remains in development, and an example of testing or feedback that changed the design.
Success criteria
The intended value is a research history that makes conversations easier to revisit and connects what was said with when it was addressed.
Define an evaluation task, such as finding a known quote or tracing a feedback item, and how I would assess accuracy, useful context, and effort. These are prompts for defining criteria, not reported results.