Grafana Playbook

Business finance,
visualized properly.

A runnable playbook for putting revenue, cash flow, expenses, and the executive KPIs that sit on top of them in front of the people who make decisions. One command brings up Grafana, a seeded Postgres, and four provisioned dashboards.

$ git clone …/grafana-playbook-finance && docker compose up -d

The dashboards

Four views, each answering one question a finance team actually asks.

Executive Overview dashboard
Executive Overview

How is the business doing this month?

The headline read: revenue, net profit, cash balance, and runway up top, then revenue versus expenses, monthly profit, the cash trend, and revenue by segment below. The one screen you'd send the board.

Revenue & Sales dashboard
Revenue & Sales

Where is revenue coming from?

Filterable by region and segment. Monthly revenue trend, revenue by product and region, and a ranked table of the top customers with color-graded values.

Cash Flow dashboard
Cash Flow

Are we cash-positive, and what are we owed?

Cash in versus cash out, net cash flow, the running balance, and an accounts-receivable aging table that flags the 90+ day bucket in red.

Expenses & Budget dashboard
Expenses & Budget

Are we on budget, and where is money going?

OpEx versus budget with variance, expenses stacked by category over time, spend by department, and a budget-vs-actual breakdown per category for the month.

What's in the box

Not just dashboards — the reasoning and the data to back them.

One-command stack

Docker Compose brings up Grafana, PostgreSQL, and Prometheus, fully provisioned. No manual clicking to see it work.

Seeded finance data

~18 months of realistic invoices, expenses, and budgets generated on first boot, so every panel renders immediately.

Readable SQL

Every metric maps to SQL you can read. The queries double as the definition of how each number is calculated.

The written playbook

Six chapters on data sources, metric definitions, dashboard design, alerting, and running it in production.

Source-agnostic

SQL by default, with patterns that transfer to MySQL, BigQuery, Snowflake, and Prometheus.

Versioned as code

Dashboards live as JSON in the repo. Changes are reviewable in pull requests, not buried in a UI.