Methods

Sixteen methods. Chosen for the decision, not the other way round.

Select any method to see a worked example from a real engagement type, with the chart and the result.

Structure

Dimensional modelling

Facts and dimensions structured so every metric has one definition and one source.

Structure

Metrics layer

KPIs defined once and reused by every report and model.

Structure

Data quality scoring

Completeness, consistency and timeliness measured before anything is modelled.

Describe

Cohort analysis

Behaviour tracked by the period a customer, job or campaign started.

Describe

Statistical process control

Control charts on cycle time, yield and throughput.

Describe

Anomaly detection

Control limits and model residuals flag what doesn't fit.

Describe

Attribution modelling

Marketing spend and channels weighed against pipeline and revenue.

Predict

Time-series forecasting

Trend, seasonality and cycle separated, then projected forward.

Predict

Regression

Models for drivers, elasticities and expected values.

Predict

Classification

Probability that a lead converts, a job runs late, a part fails inspection.

Predict

Probabilistic forecasting

Prediction intervals and quantiles, not single numbers.

Predict

Survival analysis

Time-to-event models for churn, equipment life and job duration.

Decide

Scenario analysis

Base, upside and downside cases with the assumptions that separate them.

Decide

Monte Carlo simulation

Thousands of runs across uncertain inputs give a distribution of outcomes.

Decide

Optimisation

Schedules, capacity and allocation solved within real constraints.

Decide

Model monitoring

Forecast accuracy tracked against actuals; models retrained when drift appears.

Examples are drawn from real engagement types. Clients are withheld and figures are illustrative.

Not sure which method fits?

That's what the business analysis decides. Start with a conversation.