Book an intro callIntro call
Forecast orders, inventory and planning, without a data team of your own.
A planning that lives in one head, a spreadsheet next to the ERP. We build forecasts on the company's own orders and sales, compared with the method that runs today, and only continue when the improvement is in the figures. At a wholesaler, nobody retypes over 3,300 of 5,017 order emails anymore. Senior, hands-on, without agency overhead.
Rather talk right away? Book an intro call
- 2 out of 3Order emails at a wholesaler nobody retypes anymore
- 1.500+SKUs forecast weekly in production
- 2-4 wksFrom intake to live
- 0Systems to replace
A solution architect at the table.
At Stratalytic the solution architect at the table also builds the forecasts, no bench of consultants. In 30 minutes we find the prediction that improves the most expensive decision, with an honest verdict on whether the data can carry it.
Sound familiar?
The planning lives in one head
Purchasing, stock levels and lead times rest on what one experienced person expects. That works, until the person is away or the range grows.
There are dashboards, but no predictions
A dashboard shows what happened. The decisions that cost or make money (purchasing, pricing, keeping customers) need to know what happens next.
An earlier trial never made it into daily work
It scored well in a presentation and was never wired into the operation. A calculation that doesn't change a decision is a cost, not an asset.
The Data Impact Scan
Curious what this looks like on your data?
Book an intro 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.
30 minutes, no obligation, directly with the builder.
How it works, measured at every step
- 01
30 minutes with the solution architect
The call is with the solution architect who builds the forecasts, no layer in between. Together we pick the prediction that improves the most expensive decision, with an honest answer on whether the data can carry it.
- 02
Comparison on your own data, under NDA
We run the forecast against a simple rule of thumb and the current method on the real history. The result per product group is on paper before anything is committed.
- 03
Into daily use, documented, handed over
Forecast, connection and automatic checks alongside the existing systems, at a fixed price. Documentation and handover included; the team can run it without us.
Why not an agency or a hire?
| Agency | Own hire | Stratalytic | |
|---|---|---|---|
| Who does the work | Whoever is on the bench | One hire, if one can be found | The architect from the call |
| Proof upfront | Credentials and cases | CV and interview | Comparison on the company's own data |
| Cost | Partner rates, open-ended | Salary costs + recruiting | Fixed price per project |
| After delivery | Follow-up contract | Depends on retention | The team runs it without us |
Before the call
How is this different from a data science agency or staffing?
At an agency you pay partner rates and get whichever team happens to be on the bench; with staffing you rent hours without a results agreement. Here the seniors build it themselves, with a fixed scope and a measurable criterion: the forecast must demonstrably beat the current method.
We have no data team. Is that a problem?
No, that is the usual situation. We build and run the first version and write down how it works so the team can carry on without a specialist. There is no need to open a data scientist vacancy to get started.
How do you prove a forecast is actually better?
Every forecast is compared with a simple rule of thumb and with what runs today, on the company's own history, where the system does not know the outcomes of the test period in advance. The outcome per product group is on paper. If the forecast does not beat the current approach, that is said before any project starts.
What kind of problems fit?
Predicting and prioritising with the data a company already has: demand and inventory, customers who leave, the right price, quality control, which enquiries deserve follow-up first. No research-lab work; calculations that improve a concrete decision in the operation.
What does it cost?
A comparison on the company's own data or a first trial is fixed-scope; a full project to daily use has a fixed price after the Data Impact Scan. A proposal only comes when the saving in hours and errors clearly outweighs the price. You keep working, documented software.
What if it does not work?
Then we find out early and for free: the comparison runs on a share of the company's own data, before anything is invoiced. If the forecast does not beat the current method, there is no proposal.
How much of our time does this take, and is our data good enough?
The call takes 30 minutes, after that we ask for two years of history and handle the rest. Messy data is normal; the comparison shows what is usable and what needs cleaning first, and we say so honestly, even when the answer is 'not yet'. No new system is needed.
Won't this end up in a drawer like so many trials?
The project is only done when the forecast runs in the daily work: wired to the systems, with automatic checks, documentation and handover. The code goes to the company and the team can continue without us, ongoing support is available but never mandatory.