Numbers your board can act on
Instrumentation, pipelines and executive dashboards. Built so that two people asking the same question get the same answer.
The problem
Finance, product and sales each have a number for the same metric, and all three are defensible. The monthly pack takes nine working days to assemble, and by the time it lands the decision has already been made without it.
You are probably here if
If three or more of these are true, this pillar is where your constraint lives.
- 01Two dashboards disagree and nobody knows which is right
- 02The monthly report is assembled in a spreadsheet
- 03Product analytics were installed once and never maintained
- 04Regulatory reporting is a manual process with a deadline
- 05Nobody can answer a new question without an engineer
What we actually do
Analytics setup
A tracking plan, then instrumentation that matches it, then a check that catches drift.
Data platform
Ingestion, modelling and a warehouse your team can query without asking us.
Executive dashboards
The eight numbers that drive the business, defined once, refreshed automatically.
Regulatory reporting
Scheduled, reproducible, and auditable back to the source row.
How it runs, and where you can stop
Each phase ends with something you own. Stopping after any of them leaves you better off than before it.
- 011-2w
Define
Metric definitions agreed in writing by the people who will argue about them later.
- 022-4w
Instrument
Events and pipelines built against the definitions, with tests.
- 033-6w
Model
Marts that answer the actual questions, not a copy of production tables.
- 042w
Publish
Dashboards, scheduled reports, and training for the people who read them.
What you receive
Artifacts, not a slide deck. Each one is usable by your team without us in the room.
- Signed metric dictionary with owners
- Tracking plan and instrumented events
- Pipelines with data quality tests
- Warehouse models and documentation
- Executive dashboard set
- Runbook for adding a metric without us
What this is usually built on
One engagement from this pillar
Regulatory reporting that stopped eating nine days
Monthly regulator submissions assembled by hand across four systems, with no audit trail back to source.
Read the full case- Cycle time
- 8.5 days6 hours
- Restatements
- 90
- Traceable rows
- 12%100%
Compared against the 12 monthly cycles preceding the engagement. Source: internal close log, June 2025
Not a fit if
We would rather lose the engagement on this page than at week six.
- The organisation is not ready to agree on a single definition per metric
- You want a dashboard but not the instrumentation underneath it
- The data lives in a system nobody will grant read access to
Asked on almost every first call
Do we need a warehouse?
Not always. Under a few million rows a well indexed replica is cheaper and faster. We size it to the data you actually have.
Why start with definitions?
Because every disagreement about a dashboard is really a disagreement about a definition. Fixing the tool does not fix that.
Can you use our existing analytics tool?
Yes. Most engagements keep the tool and fix what is feeding it.
How do you prevent drift?
Tests on the pipeline and a weekly check that compares event volume against expectations. Drift becomes an alert instead of a surprise.
Who maintains this afterwards?
Your team, using the runbook. We offer a retainer but the design assumes you will not need one.
Bring the constraint, not the brief
The first call is 45 minutes and free. If data and intelligence is the wrong pillar for your problem we will say so and point you at the right one.