What it delivers, measured on real data.

Every project starts with a business case and ends with a measurement. Here is what that delivered, starting with a manufacturer where the forecast went from 40% to 86% accuracy, worth CHF 778,000 a year. Clients are anonymous unless they gave permission.

2Orders and quotesFrom ongoing projects
Employee at an industrial sewing machine in a producer's workshopLive
WholesalerOrder intake

From inbox to ERP, without retyping.

Every order email used to be entered into the ERP by hand. Now the system reads the email, prepares a draft order and sets doubtful cases aside. The order desk approves instead of retyping.

67.7% of order emails become a correct draft order
5,017 order emails in the measurement
MaspartsQuote prediction

The sales team knows up front which quotes have a chance.

Quotes used to be judged on gut feel. Now a model gives every new quote a score, inside the system the team already works in.

84% Accurate quote prediction
9.445 Quotes scored automatically
Read the case
Sales figures on a tablet in a clothing storeLive
D2C webshop, NL and DEData foundation

Data from separate systems brought together, with a forecast on top.

Only once the numbers were in one place did it show how much discount leaked away each month. A forecast with purchase proposals and warnings for shortages now runs on it.

30 days of discounts in one view
Participants in a training session, one raising his handPilot
Training instituteLead scoring

Call first where the odds are best.

The model ranks open leads by closing probability. The outcome: half of the top 20% become customers.

6.1x better hit rate than without a model
632 closed leads in the test

Clients are anonymous unless they gave permission, like Masparts.Source: Stratalytic measurements and backtests, 2026. A backtest tests the forecast on a period from the past.

The same measurement on a company's own orders, stock or planning.

An intro call of 30 minutes, no obligation. If there is nothing to gain, there is no proposal.

Thirty minutes to see where the biggest impact is.

An honest estimate in numbers, even if the answer is that it does not pay off.