The situation
A manufacturer in the consumer goods market that needed to find the drivers behind its market strategy and production planning, for both product and region categories.
The challenge
- Little clarity over market shiftsInaccurate demand forecasts and no view of macroeconomic effects meant no full visibility over resource allocation, which led to unnecessary costs and lost revenue.
- No external driversA bottom-up approach limited the forecast to a simple univariate model, which cannot use leading indicators.
What Indicio did
- Identified predictive market drivers at product group and region level, such as consumer sentiment, crude oil prices, consumer prices and e-commerce retail sales.
- Built forecasts for each product group and region that need only a quick monthly update.
- Replaced a forecast without models with 50+ econometric models, weighted into one forecast.
The result
- 42%
- forecast accuracy improvement
- 50+
- econometric models weighted into one forecast
A double-digit MAPE improvement in forecast accuracy.
Sales performance optimised with forecasts validated by market and macroeconomic data.
Visibility of the significance of each driver, tracked as consumer and economic activity shifts.
Forecasts for every product group and region, updated in minutes each month.
How it was done
Seasonal patterns
The data was cleaned and the seasonal pattern identified and removed before modelling, then added back to the forecast.
Benchmark forecasts
Univariate models built on historical sales alone set the benchmark that the more advanced models had to beat.
Specific leading indicators
A VAR model with a Lasso penalty tested the possible combinations of indicators and kept the group that best predicts future sales.
Multivariate forecasts
A library of multivariate models forecast the main variable together with the indicators identified.
Weighted by accuracy
Some models are better at short, mid or long horizons. Indicio combines them into one forecast, giving more weight to the best model at each step.