Public beta: the first 50 businesses get free forecasts
Forecast and plan anything from your spreadsheets
Tell us the decision you make every week, in plain words, and upload the spreadsheet you already have. 4castPlannr works out what to forecast, tests it on your own history, and sends back a plan: how much to order, make or staff.
You describe the decision. We do the math.
No forecasting vocabulary, no templates, no model settings. Write it the way you would explain it to a new hire.
"How much dough, cheese and sauce should we order for the next two weeks? Deliveries come Mondays and Thursdays."
-
We figure out what to forecast
We read your file, clean up messy dates, duplicates and refunds, and work out which numbers drive the decision.
-
We test it on your own history
Several methods, from simple rules to an AI forecasting model, are tested on your past data. The most accurate one wins.
-
We send the plan, and keep sending it
Order quantities, a production schedule or a shift plan, compared with your current rule. Updated whenever you send new data.
Built for the decisions you make every week
Each example below was run end to end on public data. The charts are the engine's real output.
Wholesale distributors
Inventory reorder
Know how much to reorder of every item, every week, from the sales export you already have.
Food and beverage producers
Production planning
Plan what to make each day on a shared line, with shelf life, changeovers and capacity built in.
Call centers and service businesses
Staffing plans
Forecast calls, visits or jobs per day and get a shift plan that covers them at the lowest cost.
Restaurants and food service
Restaurant demand
Forecast covers, orders or items per day and know how much to prep and order.
Sites with batteries or demand charges
Energy and batteries
Forecast load and decide when to charge and discharge a battery to cut peaks and cost.
Tested in public, including where we lost
We ran 13 case studies on public datasets and published every number. Where a simple rule or an existing forecast did better, we say so.
- 13 of 13 forecasts had less error than repeating last season, tested on past data
- 25% less error than repeating last season, the median across all case studies
- 1 of 4 plans cost more than a simple rule of thumb on data they never saw. We show that one too.
What you get back
- The answer first. One paragraph in plain English: what to do and why.
- A forecast with a range, so you know the likely high and low, not just one number.
- The accuracy we measured on your data, next to simple rules, so you know how much to trust it.
- A plan you can act on: order quantities in whole cases, batches per day or staff per shift.
- A side by side with your current rule, with the costs and assumptions spelled out.
- A spreadsheet with every number, line by line.
What happens next
Your file is used only for your forecast and never to train shared models. How we handle your data.
Common questions
What does 4castPlannr do?
You upload a spreadsheet and describe, in your own words, what you want to predict or decide. 4castPlannr works out what can be forecast, tests several forecasting methods on your history, and, when your question is a decision such as how much to order or how many people to schedule, builds a plan for it. You get a plain-English report with charts and a spreadsheet you can use.
Do I need to know anything about forecasting or data science?
No. That is the point. Describe the decision the way you would to a colleague, for example: how much flour should we order each week. We choose the method, set it up and explain the result in plain words.
Is it really free?
Yes. Your first forecast is free for the first 50 businesses. No credit card and no sales call. You get the full report and spreadsheet.
How accurate are the forecasts?
It depends on your data, so we measure it on your own history before you rely on it. We hide the last stretch of your data, forecast it, and compare with what really happened, several times over. Across our 13 public case studies, our forecasts had less error than simply repeating last season in all 13 tests. See the results page for every number, including where simpler methods won.
Is my data safe?
Your files travel over an encrypted connection, are stored separately per customer, and are used only to produce your forecasts. We do not sell your data or use it to train shared models. You can ask us to delete it at any time. The security page explains the details in plain words.