/segment-spotlight

Revenue-opportunity-ranked segment analysis for any A/B test. Shows per-segment dollar values, verdicts, and rollout recommendations across device type, visitor type, and traffic source.

Works with both active and ended tests.

---

Step 0: Workspace Check

Check if the shared workspace exists and is ready:

test -d ~/intelligems-analytics/venv && test -f ~/intelligems-analytics/ig_client.py && echo "READY" || echo "NEEDS_SETUP"

If NEEDS_SETUP: Run the /intelligems-core skill first to set up the workspace.

---

Step 1: Get API Key

Check if an API key is already configured:

source ~/intelligems-analytics/venv/bin/activate && python3 -c "
from dotenv import load_dotenv; import os
load_dotenv(os.path.expanduser('~/intelligems-analytics/.env'))
key = os.getenv('INTELLIGEMS_API_KEY', '')
print('HAS_KEY' if key and key != 'your_api_key_here' else 'NO_KEY')
"

If NO_KEY: Ask the user for their API key and save it.

Never hardcode or assume an API key.

---

Step 2: Copy Spotlight Script

cp references/spotlight.py ~/intelligems-analytics/spotlight.py

---

Step 3: Select Test

The script accepts an optional test ID as an argument.

If the user already specified a test ID: Pass it directly. If no test was specified: The script lists active experiments for selection.

---

Step 4: Run Analysis

cd ~/intelligems-analytics && source venv/bin/activate && python3 spotlight.py [optional_test_id]

The script will:

1. Fetch overview analytics + all 3 segment types
2. Calculate revenue opportunity per segment
3. Rank segments by dollar value
4. Generate rollout recommendations

---

Step 5: Present Results

Read the script output and present conversationally:

1. Quick Summary

One-line takeaway on segment uniformity or divergence.

2. Revenue Opportunity Table

Ranked by annual dollar opportunity:

Segment	Type	Verdict	Lift	Confidence	Annual Value
Desktop	Device	WINNER	+15%	92%	$47,000
...	...	...	...	...	...

3. Rollout Recommendation

• Roll out everywhere — If all major segments win
• Segment-specific — If some win, some lose
• Hold — If mixed or insufficient data

4. Contradictions

Highlight any segments that contradict the overall result.

---

Step 6: Set Up Slack Automation (Optional)

Same pattern as other skills — use --slack <webhook_url> flag.

cd ~/intelligems-analytics && source venv/bin/activate && python3 spotlight.py <test_id> --slack "<webhook_url>"

---

Notes

• Revenue opportunity = segment daily visitors × RPV lift × 365. This is the annualized dollar value of the lift in that segment.
• All 3 segment types are analyzed: device type, visitor type, traffic source.
• Low-data segments show "Low data" for confidence instead of a percentage.
• COGS awareness: Uses Gross Profit per Visitor when COGS data exists.