Six systems. One set of numbers to run the week on.

The business problem

Leadership had no single source of truth for clients, capacity, delivery, churn or margin. Metrics lived in department spreadsheets and differed depending on who prepared them. In one instance, inserting a column corrupted 352 cells without anyone noticing.

What changed

Yash built a shared Postgres warehouse, 16 dashboard screens and two API routes. Time tracking, tasks, the absence calendar, the HR roster, subscription invoices, Slack-sourced sentiment and finance sheets flow into the same model through scheduled sync jobs. Purpose-built aggregation queries define each metric. An AI analyst reads those metrics through tools, with read-only SQL enforced in the database. A sync-health view flags any feed that goes stale and posts an alert to Slack.

Before & after

The business value

Weekly leadership reviews run on the platform instead of the spreadsheet. After the time-entry clean-up, the table held 291,550 rows with zero duplicate IDs. A later audit found no ID collisions across 304,719 rows, and ingestion was checked against the source API before it was relied on.

Why the system is dependable

The useful part of a dashboard is the definition behind the number. Contract values are normalised by payment frequency, client names are matched through one shared key, and data-quality caveats are shown in the interface. The AI’s database access is enforced as read-only in Postgres, and 42 automated test files cover the metric, sync and AI logic. Part of the client-request reporting began life as a small standalone dashboard, and its logic was later ported into this platform.