The question behind this project

Why did this month’s revenue change?

Some customers join, others leave, and some change their subscription. In a fictional software business, we follow those movements to explain the monthly result and spot changes worth investigating.

Read the analysis

Painting of an analyst preparing a monthly report with subscription additions, changes, cancellations, and revenue charts.

Business reporting An analytical case study

Monthly revenue, customer losses, and business performance

What changed in recurring revenue, and which changes deserve investigation?

Simulated data from a fictional business

The finding

Revenue growth and customer health need separate explanations. A revenue bridge shows the contributors; anomaly baselines identify unusual observations.

Analytical work

Build a reproducible reporting workflow from generated monthly metrics to explanatory charts and bilingual reports.

FIG. 01Retained evidence

What changed monthly recurring revenue?

952.2Opening MRR

+8.2New

+34.2Expansion

−3.4Contraction

−21.5Churn

969.7Closing MRR

USD thousands · December 2025 · net movement: +17.5

NovaCRM synthetic scenario, seed 42. This bridge uses the same figures as the December report; displayed components are rounded. Growth is a net movement: expansion can conceal contraction and churn. It does not establish a cause. Source bridge · 2026-09-05.

Reconcile the revenue movement

New business and expansion add recurring revenue; contraction and churn remove it. Read these flows together with the reported balance, and check any reconciliation difference before attributing growth to a single driver.

Ask what “unusual” is measured against

An anomaly is a comparison with a statistical baseline, not a diagnosis. Inspect the metric, reference level, detection method, and period together. A pricing event or product launch can explain an unusual value without making the measurement wrong.

Keep projection separate from observation

The forecast extends a short monthly history under a smoothing model. Its band expresses model-based uncertainty, not a commitment. The executive report should carry the same period, segment, and assumptions as the charts used to explain it.

Decision supported

What to do with this evidence

Use the report to assign an investigation: identify the revenue component, affected segment, and event to verify. Validate alert behavior against the simulator before applying it to operational data.

What this analysis cannot establish

The company, customer segments, metrics, and narrative events are generated. They demonstrate reporting behavior, not measured business impact.
A short series limits forecast validation. Health-score weights and alert thresholds are configuration choices, not universal business standards.

Continue the investigation