Data assessment
Two weeks mapping your systems, data quality and reporting needs into a costed roadmap.
We consolidate your systems into a warehouse, define the metrics once in version controlled models, and build the dashboards your leadership will actually open on Monday morning.
We build on open tools in your cloud account so nothing here depends on us continuing.
Two weeks mapping your systems, data quality and reporting needs into a costed roadmap.
Snowflake, BigQuery or Postgres with ingestion from your ERP, CRM and finance systems.
Version controlled dbt models with tested, documented metric definitions everyone shares.
A small number of dashboards designed with the people who will use them, not a 40 tab gallery.
Four weeks of pairing and training so your analysts can extend the models without calling us.
Pipeline monitoring, new source onboarding and a standing office hour for your team.
Meridian was founded in 2016 by analysts who had watched too many companies buy a dashboard tool and end up with 200 dashboards and no shared definition of revenue. We build on open, standard tools in your own cloud account, we document everything as we go, and enablement is part of every engagement rather than an upsell.
“We had four departments reporting four different revenue numbers to the board. Six weeks in there was one number and a documented definition behind it.”
Elaine V., CFO, healthcare services
“My two analysts were nervous about inheriting it. After the enablement month they were shipping new models on their own and I have not needed to call Meridian in five months.”
Trent B., Director of Analytics, distribution
“The assessment told us we did not need a warehouse yet and how to fix reporting for a tenth of the price. They came back a year later and built it properly.”
Nikki C., VP Operations, retail
Often not yet. If you are under a few million rows and reporting from two systems, cleaning up your existing reporting is usually the better spend. The assessment exists partly to answer this honestly, and we tell several clients a year to wait. We would rather build it when it will actually pay back.
Typically a warehouse such as Snowflake or BigQuery, an ingestion tool, dbt for modeling and a BI tool you likely already have. All of it is in your accounts and your name. For a mid market company the total tooling cost usually lands between 900 and 3,500 dollars a month, and we will size it with you before you commit.
It is, everyone's is, and the assessment is where we find out how much. Data quality work is scoped explicitly rather than hidden, because discovering a broken customer ID halfway through is what blows up these projects. We would rather show you an unpleasant number in week two than in month five.
That is specifically what we build against. Everything is in your cloud, in your repository, using standard open tools, with documentation written as we go. The four week enablement engagement exists to hand the keys over properly, and most clients drop to a light monitoring retainer or nothing at all inside a year.
Two weeks, a fixed fee, and a roadmap you keep. If the answer is that you should not build anything yet, we will tell you that.