// EXECUTION_PIPELINE
Implementation Pipeline
Our engine handles the complexity of data movement while you focus on high-level decision logic.
PHASE_1 // The VIP "Whales"
Identify the top 5% driving 40% of revenue. Exclude them from generic discount ladders to protect margins.
PHASE_2 // The "At-Risk" Churners
Spot users who bought monthly but haven't purchased in 60 days. Hit them with aggressive win-back flows.
PHASE_3 // The One-Hit Wonders
Isolate buyers who purchased heavily discounted items during Black Friday and never returned.
// STRATEGIC_SCENARIO
Deep Data Retrieval
How Arcli grounds AI in your exact schema to generate highly-optimized, dialect-specific execution logic.
The Engine Room: RFM Customer Scoring
How Arcli scores every customer in your database based on behavioral transaction data.
THE EXECUTIVE FILTER (ROI)
Eliminates wasted ad spend by allowing you to suppress recent buyers and aggressively bid on high-LTV lookalike seeds.
- Fully optimized for DuckDB SQL constraints.
- Bypasses semantic layer hallucinations via strict schema grounding.
DuckDB SQL_COMPILE
-- Generated by Arcli AI Semantic Router
WITH customer_stats AS (
SELECT
customer_id,
email,
MAX(created_at) AS last_order_date,
COUNT(id) AS frequency,
SUM(total_price) AS monetary_value
FROM tenant_workspace.shopify.orders
WHERE financial_status = 'paid'
GROUP BY 1, 2
)
SELECT
email,
frequency,
monetary_value,
DATE_DIFF('day', last_order_date, CURRENT_DATE) AS recency_days,
CASE
WHEN DATE_DIFF('day', last_order_date, CURRENT_DATE) > 90 AND frequency > 3 THEN 'At-Risk VIP'
WHEN monetary_value > 1000 THEN 'Whale'
WHEN frequency = 1 THEN 'One-Hit Wonder'
ELSE 'Standard'
END AS rfm_segment
FROM customer_stats
ORDER BY monetary_value DESC;