analysis-next

/analysis next

Advance the current analysis to the next ALIVE stage.

Instructions

Step 1: Identify current analysis

Read .analysis/status.md to find active analyses.

• If only 1 active Full analysis → select it automatically
• If multiple active → ask user which analysis to advance (show ID + title)
• Quick analyses don't use /analysis next (all sections are in one file)
  • If user selects a Quick, remind them to fill sections in order within the file

Step 2: Determine current stage

Read the analysis folder to see which stage files exist:

• Only 01_ask.md → current stage is ASK
• Up to 02_look.md → current stage is LOOK
• Up to 03_investigate.md → current stage is INVESTIGATE
• Up to 04_voice.md → current stage is VOICE
• All 5 files exist → current stage is EVOLVE (analysis is complete)

Step 3: Review current stage checklist

Read the current stage's checklist (embedded at the bottom of the current stage file).

Check if there are any 🔴 (stop) items:

• If 🔴 items exist → warn the user: "There are stop signals in your {stage} checklist. Review before proceeding."
• If all items are unchecked → remind: "Don't forget to review the checklist before moving on."
• If items are checked → proceed

Step 4: Generate next stage file

Based on current stage, generate the next file.

ASK → LOOK (generate 02_look.md):

Note: If this is a 📈 Modeling analysis, use the Modeling-specific templates defined in analysis-new.md Section 4B instead of the Investigation templates below. This applies to LOOK, INVESTIGATE, VOICE, and EVOLVE stages. If this is a 🔮 Simulation analysis, use the Simulation-specific templates from analysis-new.md Section 4D for all stages.

# LOOK: {title}
> ID: {ID} | Type: 🔍 Investigation | Stage: 👀 LOOK | Updated: {YYYY-MM-DD}

## Data Sources
- Primary:
- Secondary:
- Access method: (MCP / exported file / manual query / BI dashboard)

## Data Quality Review
- Row count:
- Date range:
- Missing values:
- Known issues:

## Segmentation
Break down BEFORE drawing conclusions:
- By time:
- By cohort:
- By platform:
- By segment:

Key question: "Does the pattern hold across ALL segments, or is it driven by one?"

## Confounding Variables
- What else changed at the same time?
- Population comparison: Are we comparing the same groups?
- Survivorship bias risk:
- Simpson's paradox check:

## External Factors
- Calendar: holidays, seasonality, industry events
- Competitors: recent launches, pricing changes
- Platform: iOS/Android policy, algorithm updates
- Macro: economic indicators, regulation changes

## Cross-Service Check
- Related services that may be affected:
- Shared resources (auth, data pipeline, user base):
- Recent changes in adjacent services:

## Outliers & Anomalies
-

## Sampling
- Method:
- Size:
- Rationale:

## Initial Observations
-

---
{Insert LOOK checklist from .analysis/checklists/look.md}