Forecast demo
A baseline you can see the reasons for
Pick a dataset. Five methods are tested on its own history, the most accurate is pre-selected, and you can overrule it. Then work on the chart: click a forecast month to adjust it, with a reason, and open Dayta's notes on the line. Everything runs in your browser.
Total
Click a forecast month to adjust it.
ActualForecast (sum of each SKU's method) Explained
Cases per month. Year labels mark January. Synthetic data: every number is invented. Forecasts never go below zero: if a method's line would, Dayta shows 0.
The numbers
| Row | Oct2026 | Nov2026 | Dec2026 | Jan2027 | Feb2027 | Mar2027 | Apr2027 | May2027 | Jun2027 | Jul2027 | Aug2027 | Sep2027 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Baseline | 67,976 | 72,422 | 72,376 | 77,638 | 80,201 | 74,946 | 67,780 | 68,078 | 66,450 | 67,246 | 69,469 | 71,054 |
| Adjusted | 67,976 | 72,422 | 72,376 | 77,638 | 80,201 | 74,946 | 67,780 | 68,078 | 66,450 | 67,246 | 69,469 | 71,054 |
| Difference | – | – | – | – | – | – | – | – | – | – | – | – |
| Same month last year | 61,489 | 70,070 | 70,698 | 79,139 | 81,893 | 73,104 | 63,146 | 64,499 | 64,202 | 64,636 | 68,073 | 72,571 |
Lock and export
Locking freezes the 0 adjustments and stamps the forecast with a fingerprint, so anyone can check later that the numbers weren't changed.
Everything here lives in this browser tab only. Nothing is saved or sent, and refreshing the page starts over.