New30 Oct 2026When the Forecast Becomes the Target: How to Keep Your Numbers Honest
Indicio

Home / Customers

Case study · Commodities

A commodities company: more than 50% better accuracy across 10+ global markets

How a company in the commodities sector moved from five fixed indicators to a tested set, improved forecast accuracy by more than 50%, and covered more markets without adding headcount.

Sector
Commodities
Users
Demand planning, then Strategy and Market Intelligence
Specifics
Month-to-month forecasts for 10+ global markets

Horizons

Tactical

The situation

The forecasting process was static: the same five indicators every time, and no certainty they were the right ones.

Much of the work was manual and administrative: importing files and keeping them up to date.

The challenge

  • Fixed indicatorsIt was hard to know whether the indicators in use were still the right ones.
  • Manual, administrative workTime went into importing and updating files instead of analysing the forecast.

What Indicio did

  • Tested a large set of indicators, with immediate feedback on which to use for each forecast.
  • Built and tested 45+ multivariate models in less than 15 minutes, with instant access to relevant data.
  • Shared one pool of recommended indicators across departments, so the Market Intelligence team could pull from the same list.

The result

50%
more accurate forecast: error (MAPE) of 3.3% against 6.5% internally
10+
global markets forecast without increasing headcount
45+
multivariate models built and tested in under 15 minutes
More than 50% more accurate

A double-digit MAPE improvement: 3.3% error against 6.5% before.

10+ markets, same team

Forecasting for more than ten global markets without increasing headcount.

Support for risk mitigation

Time freed up for ad-hoc scenario forecasts for other teams during volatile periods.

Shared across departments

Strategy and Market Intelligence were up and running after one or two sessions.

“If there is a change, either an upturn or downturn, we can clearly see the indicators that are valued higher, either with a pronounced higher contribution or higher influence, and when they do change in position.”
Demand planner and forecast analystCommodities company
“Now, once a good model is built, it's very robust and easy to update. You can start working with it very quickly and this saves you valuable time.”
Demand plannerCommodities company
“The tool is intuitive to use and it only required one to two sessions before the Market Intelligence team was up and running with it.”
Demand plannerCommodities company

Want the same on your data?

We set it up on a case like yours, and backtest against your current forecast in four weeks.

Book a demo
Book a demo