Home / Tools / Demand forecaster

Demand forecaster

Forecasts the next 30 to 90 days of demand from your history, with a range and a back-test so you can see how far to trust it.

Hospitality · Machine learningLive

Open the tool

What it does

Most staffing and stock decisions start with one question: how busy will we be? The forecaster learns the weekly and yearly pattern in your own history and projects it forward, showing a range instead of a single number. The back-test replays past weeks so you can see how wrong it would have been.

Who can use it

  • Operations and revenue managers
  • Store and restaurant managers planning stock and staff
  • Analysts who need a baseline forecast

Where it fits

  • Hotel occupancy
  • Restaurant covers
  • Store footfall or units sold
  • Any daily count with a weekly pattern

How to use it

  1. Load a daily history as CSV with a date column and a count column, or use the sample data
  2. Choose how far ahead to forecast
  3. Check the back-test error before using the numbers
  4. Export the forecast to feed a schedule or a purchase order

Limitations

  • It needs at least a year of daily history to learn the yearly pattern
  • It doesn't know about one-off events unless you mark them
  • Large changes like a renovation or a new competitor break the history

What's next

  • Event and holiday markers you can edit
  • Restaurant and retail editions with sector-specific inputs
  • Forecasts refreshed daily from a warehouse table

Run it on your own data

  1. In the browser (today)Free, no sign-up. What you enter stays on your device. Good for trying it and for small teams.
  2. On your files or databasePoint the same logic at your exports or a database view (PostgreSQL, SQL Server, Azure SQL or DuckDB), on your own machine or network.
  3. As a pipeline and warehouseA scheduled ETL job pulls from your systems (reservations, point of sale, inventory, payroll) into a star-schema warehouse, runs data-quality checks, and the tool or a Power BI report reads the result every morning.

What that adds here

  • A scheduled job reads yesterday's actuals and updates the forecast every morning
  • Forecasts are stored next to actuals so accuracy can be tracked over months

How I'd set it up for you

I'd map your exports or database to the tool's model, package the job in a container that runs on your server or cloud account, add the checks your data needs, and hand over the code and a runbook. Your data stays on your systems. Get in touch if you'd like it fitted to your business.

Updates

No logged updates yet. Changes to this tool will appear here and in the activity log.