Analyze multilingual feedback without flattening language or segment differences by preserving each original response, producing traceable translations and classifications, and turning the reviewed findings into a cited product brief.
Source episode · 04:00
How I AI: Gamma's 3-Step AI Workflow for Global Feedback, Art Direction, and Hiringwith Claire Vo
Zach uploads roughly 550 pieces of image editing feedback in many languages to ChatGPT Deep Research, answers clarifying questions, receives translations, classifications, themes, quotes, and citations, then turns the structured results into a Gamma presentation for product and design partners.
A row level analysis with original text, translation, category, sentiment or rating, evidence, confidence, and citations, plus a presentation that explains coverage, segment differences, limitations, and product implications.
OpenAI conversational AI
Step by step
Follow the sequence once, then adapt the prompts, checks, and handoffs to your own setup.
3 steps
Export the permitted fields with a stable row ID, original response, language or locale, date, segment, feature, model, and rating. Remove unnecessary personal data and document the sampling frame before upload.
Ask Deep Research to translate and classify each row under a fixed rubric, then summarize themes and segment differences with citations and uncertainty.
Analyze this multilingual feedback for [research question]. First report rows, languages, dates, segments, missing fields, and exclusions. For each row preserve ID and original text, provide translation, category from [rubric], sentiment or rating, evidence, confidence, and NEEDS_REVIEW when ambiguous. Then summarize themes with counts, rates, exact citations, counterexamples, and segment differences. Do not invent quotations.
Review the row level output and key translations, then build a Gamma deck from a locked source outline. Include coverage, methods, findings, representative citations, differences by language or plan, limitations, decisions, and research questions without adding unsupported claims.
Create a product research presentation from this reviewed outline. Preserve every source row citation, qualifier, and unknown. Include coverage and method before findings, distinguish count from rate, show counterexamples, and do not add customer quotes or regional conclusions not in the outline.
Turn an idea into a PRD, user stories, and a plan.
After the steps
How to recover when the loop fails and where human judgment helps.
Recover
Keep building
Create context-aware guardrails for your enterprise AI agent using a 'Project' governance layer. As a project owner, define rules, scope tools, and require human-in-the-loop approval for sensitive actions, ensuring safe and effective AI use.
Transform a successful AI chat session into a reusable skill that anyone in your company can use. Package prompts, tools, and code into a standardized workflow, making it easy to share and scale automation across your organization.
Use a natural language AI agent to instantly generate and iterate on custom data dashboards. This workflow queries your company's data securely, allowing non-technical users to perform their own 'last mile' data analysis without filing tickets.
Join 100,000+ product managers who use ChatPRD to write better docs, align teams faster, and build products users love.