€778K a year in our case
Your forecast already runs. We show how much better it can run, and what that is worth per year.
For supply chain and demand planning teams that already forecast. We benchmark against your current forecast on your own data, and you see the accuracy improvement before you see a quote.
No obligation, under NDA. No improvement, no cost.
71%
Accuracy uplift (our case)
€778K
Annual saving
€0
Upfront cost
4-6 wk
Backtest to readout
€0K
Saved per year
- GDPR & EU cloud
- NDA first
- Works with SAP IBP, o9 and Anaplan
The difference
Standard software, or a model on your data
Standard forecasting software optimizes one algorithm library for thousands of companies at once. We spend all of it on one dataset: yours. Same benchmark, on your own history.
Forecast accuracy in our reference case
40% to 68% accuracy over 1,247 products, worth €778K a year.
Curious what this looks like on your data?
Book a free 30-minute call with the solution architect who also builds the work. You walk through your processes together and get an honest answer on where the numbers are for you, and where they are not.
Free intro call
30 minutes, no obligation, directly with the builder.
Every pallet bought too early is working capital standing still.
Better forecasts mean less stock, fewer stock-outs and more margin.
01
Secure, and on your terms
NDA first
We sign an NDA before you share anything.
Your data stays yours
Processed in an EU cloud, GDPR-compliant. We never train on it.
No system access
Only your actuals and current forecast. No ERP connection, SKUs may be anonymised.
Only then comes the no-obligation analysis.
02
The analysis: proof before you pay
Most teams assume they're at the accuracy ceiling. Rarely true. Send us your actuals and current forecast; we backtest on a hold-out period and compare against your own forecast (Forecast Value Added). You see the accuracy improvement by segment first, and only a pre-agreed threshold leads to a quote. If not, no cost.
Want a rough number before you send anything? Our free calculator converts a percentage point of accuracy into freed working capital and recovered lost sales.
03
A standard model, or a model on your data
Standard forecasting software auto-selects the best-fitting model from a fixed library. Fine for the average SKU in an average company. We build a model that learns from your own sales data instead: your seasonality, your product mix, and the effect of your own campaigns and peaks like Black Friday.
We combine techniques for promotions, seasonality, new products and the volatile long tail where standard methods break down, all explainable to your team. No black box. We do not replace your software; we raise the accuracy of the forecast it produces, and prove it first on your own history.
04
Proof, and it fits your stack
A global leader in construction tools had no history for new product launches. Our models took forecast accuracy from 40% to 68% (a 71% relative gain) and saved €778K a year across 1,247 products. You keep your own systems and ERP; we connect to them, so our models run alongside with no integration risk. Read the case.
“Stratalytic helped us implement a data-driven approach that had a direct impact on our bottom line. They understood our challenges and delivered something that genuinely works.”
Process
How does it work?
- 1
Intake (30 min)
Together we set the segment, horizon and improvement threshold. No data shared yet.
- 2
Data under NDA
You send historical actuals plus your current forecast. No system access, SKUs may be anonymised.
- 3
Backtest & readout
We run our models on a hold-out period and show the accuracy improvement by segment. Numbers first.
- 4
Decision
Beat the agreed threshold and a quote follows. If not, no cost, no further ask.
FAQ
Frequently asked questions
What does this cost, and what if you don't beat my forecast?+
The analysis runs against an improvement threshold agreed upfront (for example a number of accuracy percentage points, or a minimum Forecast Value Added). If we don't clear it, there is no cost and no next step. Only once that threshold is cleared does a quote follow.
What data do you need, and is it secure?+
Your historical actuals plus your current forecast output over the same period. Nothing more: no system access, no integration. Transfer under NDA via secure transfer, SKUs may be anonymised. We process in an EU cloud, GDPR-compliant, and never train on your data.
How do you measure the improvement?+
We measure the accuracy improvement by segment, plus Forecast Value Added against your current forecast and a naive baseline. A single blended accuracy number says too little: the gain often sits in the long tail, intermittent demand and new products, so we break it out.
We already use forecasting or planning software. Are you replacing that?+
No. Standard forecasting software auto-selects a generic model from a fixed library, which is fine for broad planning. Our models sit alongside it, built on your data, and lift accuracy where a percentage point of error costs you the most. You migrate nothing and hand the forecast back through the interfaces you already have.
What makes your approach strong?+
Deep experience in machine learning, deep learning and statistics, and models tuned specifically to your data rather than one off-the-shelf approach. We combine techniques for promotions, seasonality, new products and intermittent demand, and keep everything explainable. No black box.
Show us your forecast. We'll show you the delta.
Book a 30-minute intake. NDA first, you share your data securely, and within weeks you see the delta against your current forecast, before any quote.
Measured at these companies
