Questions we get asked.
The ones that come up on almost every first call, answered properly rather than deflected. If yours is not here, ask it directly and you will get a straight answer.
The engagement
Is the price really fixed?
Yes, once the assessment is done. The free assessment is what makes a fixed price possible: it inventories the estate and classifies every dataflow by complexity, so the scope is known before anyone commits. You get a fixed price and a fixed timeline, not an open-ended time-and-materials bill.
What does the free assessment actually cost us?
Read access and about two weeks of light involvement from someone who knows your estate. There is no fee and no obligation to proceed. If you decide not to migrate, you keep the report.
What if we decide not to migrate?
Then you keep the assessment report, and it is still worth having. Knowing exactly what your estate contains and what leaving would cost is real leverage at your next renewal, whether or not you ever use it to walk.
How long does a migration take?
It depends entirely on what the assessment finds, which is the point of running it first. The assessment returns a fixed timeline alongside the fixed price, so you know before committing rather than discovering halfway through.
Can we migrate part of the estate rather than all of it?
Yes, and it is often the right call. The assessment classifies pipelines by complexity and flags dead work, which makes it straightforward to carve out a phase. Many teams start with the highest-cost or highest-risk slice and decide about the rest afterwards.
Technical
Which warehouses and tools do you target?
SQL, dbt, and Airflow on the warehouse you already run: Snowflake, Redshift, BigQuery, Databricks, or Postgres. The output is open, version-controlled, and standard, so any engineer can read and change it without a proprietary builder.
What happens to our dashboards?
Dashboards are the easy part. The hard, valuable part is the transformation logic underneath them, which is what we migrate. Your BI layer keeps reading from tables, except now those tables live in a warehouse you own rather than behind a vendor's consumption meter.
How do you prove nothing changed?
Every migrated pipeline runs side by side with its original through our parity harness, which checks row-level and aggregate equivalence and produces a diff report. Nothing is cut over until that report comes back clean. You sign off using your own numbers, not our assurance.
Do you use AI to convert our logic?
Partly, and only where it earns its place. The structural majority of an estate is converted by deterministic transpiling, because a parser is more reliable than a model for joins, filters, and aggregations. AI-assisted conversion handles the dense bespoke formula logic, with an engineer reviewing every output and the parity harness checking it against the original. That review is a sign-off, not a formality: real accountability for correctness has to sit with a person, not a model's confidence score.
What about custom macros and bespoke logic?
They get unpacked and converted with review rather than treated as a black box. Macros are exactly where estates hide their real complexity, which is why the assessment walks them rather than counting only top-level workflows.
Will our team be able to maintain the result?
That is the objective. The output is ordinary SQL, dbt models with tests and documentation, and Airflow DAGs. Any competent data engineer can read it. On assisted migrations your engineers review the work as it lands, so the knowledge stays in your building rather than leaving with us.
Risk, access, and data
Do you need access to our production data?
The assessment needs read access to inventory the estate. The migration runs against your environment with your team in the loop. Nothing is deleted or cut over without your explicit sign-off, and your data stays in your infrastructure.
What happens if something breaks after cutover?
Cutover happens on a schedule you set, with the original pipelines available as a fallback during the stabilisation window. There is no big-bang weekend where the old system is switched off before the new one has proven itself in production.
Our workflows are business critical. Is this too risky?
That case is exactly what the parity harness exists for. Business-critical pipelines are the ones where evidence matters most, and where a spot-check is not good enough. Every pipeline gets a diff report, and your finance or audit team signs off on their own numbers before anything moves.
Do we have to tell our current vendor?
That is your call and your commercial relationship. The assessment is a read-only inventory of your own estate, and plenty of teams run one purely to understand their position before a renewal conversation.
Still unanswered?
The detail on process lives in how it works, the detail on scope and pricing lives under services, and the platform-specific answers are on the Domo and Alteryx pages. Anything else, ask us.
Find out what leaving would actually cost.
The lock-in assessment is free, takes two weeks, and needs read access. You keep the report whether or not you migrate.