Product Engineers

Embedded Analytics Without the Warehouse Bill

Your product embeds SQL-powered dashboards. Every customer query shouldn't spin a $50K/mo warehouse. HatiData gives you Postgres-compatible SQL with per-second billing.

Run the Audit Script →View Sample Report →

THE PROBLEM

What You're Dealing With

60-Second Minimum Billing

Each dashboard tile fires a query. Legacy warehouses bill 60 seconds minimum per query — even for sub-second lookups. You're paying 20-60x the actual compute.

Cold Start Latency

Warehouse clusters take 5-30 seconds to resume from idle. Your customers see spinners while the meter is already running.

Multi-Tenant Cost Explosion

Each customer workspace needs isolated compute. Legacy warehouses charge per-warehouse, making multi-tenant analytics prohibitively expensive.

THE HATIDATA FIX

How HatiData Solves It

Per-Second Billing

A 200ms query costs 200ms of compute. No 60-second minimum. No rounding. You pay for exactly what your customers use.

Always-Warm, 5s Auto-Suspend

DuckDB's in-process engine starts instantly. No cold starts. Auto-suspend in 5 seconds, not 60. Your dashboards feel native.

Shared Compute, Isolated Data

One HatiData cluster serves all tenants with row-level security. No per-workspace warehouse sprawl. Costs scale linearly.

Side-by-Side Comparison

Feature	Legacy Warehouse	HatiData
Minimum Billing	60 seconds	Per-second (no minimum)
Cold Start	5-30 seconds	Instant (in-process)
Auto-Suspend	60 seconds	5 seconds
Multi-Tenant Model	1 warehouse per tenant	Shared compute + RLS
Typical Dashboard Cost	$2,000-5,000/mo per tenant	$200-500/mo per tenant
SQL Compatibility	Native	99%+ (AI transpiler)

CODE EXAMPLES

Drop-In Integration

import psycopg2

# Connect to HatiData — same as any Postgres database
conn = psycopg2.connect(
    host="hatidata.internal.yourcompany.com",
    port=5439,
    dbname="analytics",
    user="dashboard_svc",
    password="***"
)

cursor = conn.cursor()
cursor.execute("""
    SELECT customer_id, SUM(revenue) as total_revenue
    FROM orders
    WHERE created_at >= CURRENT_DATE - INTERVAL '30 days'
    GROUP BY customer_id
    ORDER BY total_revenue DESC
    LIMIT 10
""")

rows = cursor.fetchall()