Files
2025-11-29 18:30:14 +08:00

51 lines
3.3 KiB
Markdown
Raw Permalink Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
---
name: run-survey
description: Launches customer surveys with sampling plans, question sets, and automation hooks.
usage: /customer-feedback-orchestration:run-survey --audience "admins" --goal adoption --channels "email,in-app" --deadline 2025-12-15
---
# Command: run-survey
## Inputs
- **audience** persona or segment target (admins, champions, exec sponsors).
- **goal** objective (adoption, roadmap validation, satisfaction, messaging test).
- **channels** comma-separated distribution methods (email, in-app, community, CS-led).
- **deadline** fielding end date or fixed duration.
- **incentive** optional reward description or budget.
- **sample_size** desired number of responses (numeric).
### GTM Agents Pattern & Plan Checklist
> Mirrors GTM Agents orchestrator blueprint @puerto/plugins/orchestrator/README.md#112-325.
- **Pattern selection**: Survey orchestration typically runs **pipeline** (brief → question design → sampling → monitoring → handoff). If question design + sampling prep run in parallel, note a **diamond** block with merge gate in the plan header.
- **Plan schema**: Save `.claude/plans/plan-<timestamp>.json` capturing audience, goals, sampling logic, dependency graph (legal, CS, research ops), error handling, and success metrics (response rate, completion %, demographic balance).
- **Tool hooks**: Reference `docs/gtm-essentials.md` stack—Serena for MAP automation diffs, Context7 for research SOPs + historic surveys, Sequential Thinking for pre/post-readouts, Playwright for survey QA.
- **Guardrails**: Default retry limit = 2 for distribution/QA failures; escalation ladder = Research Lead → Customer Marketing → Exec sponsor.
- **Review**: Run `docs/usage-guide.md#orchestration-best-practices-puerto-parity` before launch to confirm dependencies, approvals, deliverables.
## Workflow
1. **Brief Assembly** capture goals, personas, hypotheses, and stakeholder approvals.
2. **Question Design** generate question bank with logic, scales, and branching recommendations.
3. **Sampling & Distribution** build target lists, dedupe suppression lists, schedule sends.
4. **Monitoring** track response rates, data quality, and audience balance; trigger boosts if needed.
5. **Handoff** package clean dataset + metadata for `synthesize-feedback` command.
## Outputs
- Survey plan (audience, sample, channels, incentive, timeline).
- Question set with logic map and translation/localization notes.
- Distribution checklist + automation steps for MAP/CS tooling.
- Plan JSON entry stored/updated in `.claude/plans` for audit trail.
## Agent/Skill Invocations
- `research-lead` validates methodology and question quality.
- `cs-analyst` ensures sampling + data integrity.
- `survey-design` skill provides templates and guardrails.
- `stakeholder-ops` skill outlines approvals + comms.
## GTM Agents Safeguards
- **Fallback agents**: document substitutes (e.g., CS Analyst covering Research Lead) when leads unavailable.
- **Escalation triggers**: if response rate misses guardrail twice or sampling bias detected, escalate via GTM Agents rip-cord and log corrective actions in plan JSON.
- **Plan maintenance**: update plan JSON/change log when audiences, question sets, or tooling stacks change to preserve audit parity.
---