Upload a CSV with a date column and a metric. Get a Holt-Winters forecast with 95% prediction bands, a trend and seasonality breakdown, and a backtested accuracy score. Free.
or click to browse · max 3 MB
Fitting forecast model and backtesting...
Sent to — forecast chart with 95% prediction bands, trend and seasonality breakdown, seasonal pattern, backtested accuracy, R code, and AI insights.
Analyze another fileThe analysis parses your dates, aggregates duplicates, infers the series frequency (daily, weekly, monthly), and fills small gaps by interpolation. It then fits a Holt-Winters exponential smoothing model — with additive seasonality when the history supports it, falling back to a non-seasonal trend model or ARIMA(1,1,1) otherwise — and projects forward with 80% and 95% prediction intervals. A holdout backtest refits the model without the most recent points and scores it (MAPE/MAE) so the accuracy claim is earned, not assumed.
Use it when you have a dated history of one metric — sales, signups, traffic, demand — and want a defensible projection with uncertainty bands and a trend/seasonality readout.
Not for series driven mainly by known external events (promotions, launches), for multi-driver causal forecasting, or for histories shorter than ~15 points.
Built for: Operators, planners, and analysts who need a quick defensible projection of a business metric
Typical data source: Any CSV or spreadsheet with a date column and a numeric metric — sales by week, orders by day, revenue by month
A dated history of one metric. For example, weekly sales:
Minimum 15 rows · Best with 1-3 years of daily, weekly, or monthly history (30-1,000 points)
Standard-library analysis: forecast any metric from its own history. Maps a date column and a numeric column, infers the data's frequency (daily, weekly, monthly), fits a Holt-Winters exponential smoothing model with automatic seasonal and non-seasonal fallbacks, and projects forward with 95% prediction intervals. Includes a holdout backtest (MAPE/MAE) so you know how much to trust the forecast, plus a breakdown of trend and seasonal pattern.
Your history and the projection on one line — the shaded band is the 95% prediction interval, the honest range of where the metric could land.
The metric split into its moving parts: current level, trend per period, and the size of the seasonal swing.
The repeating within-cycle pattern — which days, weeks, or months run above or below trend, and by how much.
The model tested on data it never saw: holdout MAPE and MAE tell you how far off the forecast tends to be.
Plain-English interpretation — what the numbers mean, what's significant, and what to do next.
Where will this number be in a few months?
Map your date column and the metric. The model learns the trend and any repeating seasonal pattern from your own history, projects forward with a 95% prediction band, and backtests itself on held-out data so you know how much to trust it.
See our FAQ for details on pricing, data privacy, and how the analysis works. Every report includes a Methodology section showing the statistical test, assumptions checked, and diagnostics run.
Run any analysis on your own data — validated R analyses, interactive reports, AI insights, and PDF export.
Try Free — No Credit Card