Automated data analysis
Connect your database or upload a file. You get a written report and an interactive dashboard — churn, LTV, CAC, forecasting and risk scoring, already interpreted.
What it replaces
Not a pile of charts to interpret yourself — a written report you can forward to the board, and a dashboard to dig into when you want the detail.
A document with the findings written out: what changed, why it likely changed, and what to do about it. Every figure is checked against the source data before it's published.
Section 1
Revenue is concentrated: the top six customers account for 44% of the period. Three high-severity anomalies were flagged, all on profit, all in the last week of April — pointing to volume rather than pricing.
Every metric the analysis produced, grouped by theme. Risk shows up in colour, the long tail is grouped so the chart stays readable, and the underlying tables come with a written summary.
Each run goes from basic aggregation through to causal inference. You pick the depth; the pipeline does the rest.
Multi-method forecasts with seasonality, confidence intervals and anomaly detection on every metric.
A weighted score across four dimensions, per category and per customer, so you know where to look first.
Retention by cohort, lifetime value by plan, acquisition cost by channel, and the LTV/CAC ratio for each one.
Groups built from behaviour rather than guesswork, each with its own value, risk and engagement profile.
Which metric actually moves which, with lag correlation and mutual information — not just what happened together.
Conservative, moderate and aggressive projections so you can see the range before committing to a decision.
Set it up once. Run it when you need it, or leave it on a weekly schedule.
Point it at a SQL or NoSQL database, or upload a CSV or Excel file.
Say what you want to find out. Relevant metrics are suggested from there.
The pipeline processes in the background and checks its own numbers.
Report and dashboard, ready. Schedule it to arrive every Monday if you want.
A 30-minute call. Bring a dataset and we'll analyse it live.