PROVIDER - Spreadsheet BI native to the cloud warehouse

Sigma Computing

Sigma Computing is warehouse native BI: live queries against Snowflake, Databricks and BigQuery, a spreadsheet surface for self-service analysis, and embeddable delivery. It removes extracts and copies, while query spend on the warehouse bill and upstream data hygiene become the governing costs.

Data Visualization & BI United States 2026.11.20
4.3pts

Detailed Notes

Pros 5 items

Queries run directly against the cloud warehouse with no extract, so reports stay consistent with 1 source of truth
A spreadsheet style surface lets business users build pivots and calculations without writing 1 line of SQL
Pushes queries down to 3 or more warehouse types including Snowflake, Databricks and BigQuery
Workbooks publish as embedded analytics pages under 1 permissions model
Compute lands in the warehouse, so the platform stores 0 extra copies of the data

Cons 4 items

Query cost shows up on the warehouse bill and can climb fast without governance
Depends on warehouse hygiene, so messy underlying tables degrade the experience
Deep modelling and complex preparation still need upstream tooling
Access from mainland China and local deployment support are limited and need a compliance review

Service Overview

Sigma Computing is a cloud warehouse native BI platform from the United States, representing the no-copy school of data visualisation and BI. It does not move data into its own store, instead pushing queries down to the warehouse that already holds it.

That architecture sets three traits: reports always match the source, the platform adds no storage cost, and performance plus spend follow the warehouse. For teams with disciplined warehouse modelling this is the shortest path to analysis; for teams without it, the underlying problems surface unchanged.

Key Strengths

  • Live queries: the warehouse executes, so reports share one source of truth
  • Spreadsheet experience: familiar cells for pivots and calculations
  • Multi warehouse reach: Snowflake, Databricks and BigQuery supported
  • Embeddable delivery: workbooks publish as external analytics pages
  • No copy overhead: no duplicate storage or sync pipeline to maintain

Product Ecosystem

Platform capabilities

Warehouse direct queries, spreadsheet modelling, charts and pivots, parameter inputs, role and row level permissions, embedded publishing, and usage monitoring.

Integrations

Connects to cloud warehouses, data catalogues and identity providers, working either as a standalone analytics entry point or embedded inside a product.

Typical scenarios

Business self-service data access, customer facing data products, weekly finance and operations reviews, and a shared reporting layer over the warehouse.

Limitations

  • Query cost: warehouse spend scales with usage without guardrails
  • Upstream dependency: poor table design shows up directly
  • Modelling depth: heavy ETL remains an upstream job
  • Regional support: mainland access and localised compliance need separate review

Use Cases

  • Teams with a governed warehouse (★★★★★): live queries give consistency
  • Business self-service (★★★★☆): spreadsheet surface lowers the bar
  • External data products (★★★★☆): embedded publishing suits client analytics
  • Small teams without a warehouse (★★☆☆☆): prerequisites are missing

Pricing

Item Billing model Notes
Platform subscription Per seat by edition Viewer and explorer roles
Warehouse queries Settled by the warehouse Appears on the cloud bill
Enterprise Contact sales Adds governance and embedding

FAQ

Must data be loaded into Sigma?

No, queries push down to the warehouse.

Do business users need SQL?

Not for most work, though SQL remains available for complex logic.

What if queries are slow?

Warehouse tuning and aggregate tables usually fix it; the platform does not cache extracts.

Can it embed in our own product?

Yes, embedded publishing respects the configured permission model.

Code Examples

-- Sigma pushes aggregates like this down to the warehouse; no data lands on the platform
SELECT date_trunc('week', paid_at) AS week,
       campaign,
       SUM(revenue) AS revenue,
       COUNT(DISTINCT order_id) AS orders
FROM analytics.orders
WHERE paid_at >= DATEADD('day', -30, CURRENT_DATE)
GROUP BY 1, 2
ORDER BY 1, 3 DESC;
# Compare warehouse-side runtime and cost for the same logic before publishing a workbook
# Example: review the last 7 days of query history in Snowflake

Community & Resources

Visit Sigma Computing Website → https://www.sigmacomputing.com/
Visit Website ↗