Northgrain Data

dbt consulting services for safer analytics releases.

Audit, implement or refactor dbt Core and dbt Cloud projects across Snowflake and BigQuery. Improve model structure, testing, contracts, CI, documentation and release confidence without rebuilding everything from scratch.

Talk to us about your dbt projectBack to all services
Engagement
Focused specialist consulting
Starting point
Your existing platform and team

Is this the right fit?

A strong fit if

This consulting work is a strong fit when…

✓

No tests that mean anything

✓

No layering, just one file

✓

Brittle runs

✓

Lineage nobody trusts

Probably not a fit if

It is probably not the right fit when…

×

The platform and delivery constraint are still completely unknown.

×

The work needs permanent on-call coverage or an unrestricted backlog.

What changes

dbt consulting services

Choose the dbt problem that needs to change. Each service connects implementation to a reviewable output.

dbt project audits and health checks

Review project structure, model grain, tests, CI, documentation, performance and ownership, then prioritize the fixes.

dbt Core and dbt Cloud implementation

Set up or refactor projects, environments, jobs and pull request checks for the way your team ships changes.

Unit tests, contracts and data quality

Turn business assumptions into tests and explicit model interfaces that fail before a dashboard does.

Legacy SQL migration

Move stored procedures, views and repeated dashboard SQL into layered, reviewable dbt models.

Incremental model performance

Tune materializations, keys and warehouse usage around the actual data shape and update pattern.

Orchestration and project standards

Connect dbt to CI, Airflow or Dagster and document conventions the team can continue using.

Technical example

A reviewable fix, not a template.

A common failure mode is a refund join that silently changes revenue. The revised model uses references, an explicit key, an enforced contract and a singular test that returns any broken rows.

models/marts/fct_revenue.sql (before)

-- models/marts/fct_revenue.sql (before)
-- Grain: invoice_id. Source: raw.invoices joined to raw.refunds.
-- Tests: not_null(invoice_id), unique(invoice_id).
-- The reconciliation between fct_revenue and the GL runs in a
-- spreadsheet. The gap is "explained" by a footnote the day before
-- close. The dbt test suite does not catch the double-count.
select
  i.invoice_id,
  i.amount,
  r.refund_amount
from raw.invoices i
left join raw.refunds r using (invoice_id)

models/marts/fct_revenue.sql (after)

-- models/marts/fct_revenue.sql (after)
{{ config(
  materialized='incremental',
  incremental_strategy='merge',
  unique_key='invoice_id',
  contract={'enforced': True},
  on_schema_change='append_new_columns',
  snowflake_warehouse='transforming_xs',
  tags=['finance', 'marts']
) }}

with invoices as (select * from {{ ref('stg_invoices') }}),
     refunds as (select * from {{ ref('stg_refunds') }})
select
  i.invoice_id,
  i.amount                                            as gross_revenue,
  coalesce(r.refund_amount, 0)                        as refund_amount,
  i.amount - coalesce(r.refund_amount, 0)             as net_revenue,
  i.updated_at
from invoices i
left join refunds r using (invoice_id)
{% if is_incremental() %}
  where i.updated_at > (select max(updated_at) from {{ this }})
{% endif %}
  • Sources through refs and staging models, not raw tables
  • Incremental merge, not a full daily refresh
  • An enforced contract makes the model interface explicit
  • A singular test checks the business invariant, not only not_null

Reviewable delivery

A dbt change is more than a merged model.

Before sign-off, the project should show the model grain, dependencies and acceptance criteria alongside the pull request. Relevant tests should run successfully, documentation should describe the interface, and any remaining migration or performance risk should be explicit.

How we work

How the work enters your repository.

Use a defined project for one dbt outcome or recurring support for a backlog that keeps growing.

01

Embedded delivery

Northgrain works inside your team: models, contracts and pull requests in your repository, with a clear communication rhythm and handoff.

02

Project delivery

A scoped dbt build with a clear start, end, and handoff: a migration, a cleanup, or an audit with a prioritized roadmap your team can run.

Questions

Answered before the first call.

Discuss your dbt project.

Share the repository, the failure mode and what needs to change. We will reply with the most useful next step.

Get in touch