Virtual warehouses
Separately sized compute clusters that run your queries and suspend when idle.
Every account — and the single biggest factor in what you pay.
We build and run on thisEasy to start on. Easy to overspend on.
We design, build and operate Snowflake accounts: ingestion, transformations, roles, and the warehouse sizing that quietly decides your bill. Committed spend can be placed through our partnership.
Not a boxed licence — we design and operate workloads here, and can place your committed spend through us.
What we work with
The products, and who each one suits.
6 listed
Snowpark
Python, Java and Scala running inside Snowflake, next to the data.
Teams who would rather not export data to a separate compute cluster to transform it.
Snowflake Cortex
Managed model functions callable from SQL inside the account.
Analytics teams who want a model over governed data without standing up a second platform.
Dynamic Tables
Declarative transformations that refresh to a target lag instead of a schedule you maintain by hand.
Pipelines currently held together by cron jobs and manual reruns.
Secure Data Sharing
Live access to a table for another organisation, without sending anybody a file.
Anyone still exchanging data by SFTP, email attachment or shared drive.
Horizon governance
Masking, row access policies, tagging and lineage across the account.
Regulated organisations that have to show who can see what, and prove it.
What buying through us changes
A partnership is only worth what it changes.
Better than list on committed credits
Snowflake sells credits, and a committed contract prices below on-demand. We place the commitment through our partnership and size it against measured consumption, so you are not paying for a forecast that did not happen.
Named contacts for the vendor’s problems
A named route into Snowflake when the issue is Snowflake’s — a query planner behaving oddly, a region question, a contract change — instead of joining a general support queue.
Startup credits we sponsor
Snowflake runs a startup programme with credits attached. We prepare and sponsor the application. The award is Snowflake’s call.
How we help
The work around the product.
- Model the estate before any migration: what moves, what stays, and what should be deleted rather than carried across.
- Migrate off the existing warehouse — Redshift, Synapse, on-premises Oracle or SQL Server — with the old system live until the new one is proven.
- Set up roles, masking and row access policies so the access model is designed rather than inherited from whoever set up the trial.
- Right-size warehouses, set auto-suspend, and separate workloads that are quietly competing for the same cluster.
- Build ingestion and transformation with dbt or Snowpark, under version control, with tests that fail loudly.
- Review the bill every month and report what changed and why — in a page a CFO can read.
Where this is not the answer
Snowflake bills compute by the second, which makes it easy to spend a great deal without ever deciding to. If nobody owns warehouse sizing after we hand over, the saving does not survive the year — we would rather agree a small operations retainer than leave you with a bill that grows quietly. And if your workload is one dashboard over a few million rows, Snowflake is more platform than you need, and we will say so before the contract rather than after.
Next step
Buying Snowflake? Talk to us first.
Tell us what you are trying to do and we will come back with the sizing, the licensing and the price — from a partner, within two working days.