Skip to content
Data & Database

Granite Works cut data platform costs by 31% with warehouse scheduling and storage tiers

A 360-person analytics firm ran its cloud data warehouse around the clock and kept all historical data on hot storage. Scheduling, auto-suspend and tiering cut data platform spend from ₹18L to ₹12.4L a month.

  • Data engineering and analytics services
  • 360 people · Large
  • Pune
  • 6 weeks

Monthly spend

₹18L₹12.4L
Saved / month
₹5.6L
Reduction
31%
Payback
12 days

−31%

Data platform spend

₹67L

Annual saving

−82%

Warehouse idle time

−40%

Pipeline runtime

The challenge

Granite's warehouses ran 24 hours a day, though most jobs ran during business hours and in a nightly batch window.

Seven years of client data sat on premium storage even though queries rarely touched anything older than 18 months.

What we reviewed

  • Warehouse usage, queuing and idle time
  • ETL job schedules and runtimes
  • Storage by age and access frequency
  • BI seat usage
  • Data egress to client systems

The approach

  1. 1

    Schedule and auto-suspend

    Warehouses now scale to demand and suspend after five minutes idle, with separate sizing for batch and interactive use.

  2. 2

    Tier old data

    Data older than 18 months moved to an archive tier, still queryable when needed.

  3. 3

    Optimize the pipelines

    Incremental loads replaced full refreshes for the twenty largest tables.

  4. 4

    Right-size BI seats

    Viewer licences replaced creator licences for staff who only consume dashboards.

Before and after

Monthly cost by line item, in Indian rupees.

Granite Works — monthly data & database spend

Line itemBeforeAfterSaving
Warehouse compute (24/7 → scheduled)₹7,80,000₹4,90,000−₹2,90,000
Hot storage₹3,20,000₹1,62,000−₹1,58,000
Archive storage tier—₹38,000+₹38,000
ETL and pipeline compute₹3,10,000₹2,40,000−₹70,000
BI and dashboard seats₹1,90,000₹1,48,000−₹42,000
Data egress₹2,00,000₹1,64,000−₹36,000
Total₹18,00,000₹12,42,000₹5,58,000 (31%)

Monthly spend over time

₹67L saved per year

Two months before the engagement, then six months after the plan was delivered.

Timeline

  1. Weeks 1–2

    Usage analysis

    Warehouse, storage and pipeline telemetry reviewed.

  2. Weeks 3–4

    Design

    Scheduling, tiering and pipeline changes designed with data leads.

  3. Week 5

    Plan delivered

    Change plan with client-impact assessment.

  4. Week 6

    Support

    Scheduling and tiering live on the first warehouses.

“Our warehouse was awake all night with nothing to do. Aditya's team designed a schedule that cut a third of our data platform bill, and our pipelines got faster in the process.”
SKSneha KulkarniDirector, Data Platforms, Granite Works

Find out what your agency could save

Book a 20-minute call. We'll show you where the quickest wins usually are for a team like yours, with no data required and no obligation.

  • Free 20-minute assessment
  • NDA before any data is shared
  • Fixed-scope quote, 30/70 payment terms
  • 7 days of post-delivery support
Chat with us