How it works

PlanCore is a desktop tool that automates sales forecasting and inventory buying using your existing Excel spreadsheets.

01Data input

Connects straight to your existing Excel sales records. No complex integration needed.

02Data smoothing

Flags and fixes outliers like stockouts or entry errors, while preserving real promo spikes like Black Friday.

03Run forecast

Automatically tests, scores and blends multiple AI and statistical models to find the best fit.

04Forecast results

Displays editable, top-level forecasts. Manual changes are saved permanently without losing the original AI baseline.

05Disaggregate

Automatically breaks down high-level forecasts into channels, categories, styles, sizes and stores.

The shares editor, two panes with history on the left and the forecast on the right. The top pane holds the disaggregation node - every collection under its style, each with its own share of the channel total, and the blue cells are shares a planner typed over the average. The bottom pane drops to individual patterns, each with start and end dates and its own share of the collection above. Sale periods keep their own columns beside the months, and a legend under the grid names every fill.

06Purchase

Translates forecasts into specific weekly purchase orders based on lead times, order minimums and stock limits.

The purchasing inventory projection: a week-by-week walk-through of opening stock, inbound, buys, transfers and demand down to unmet demand, ending units and weeks of supply, shown for the total and for a single style beneath it. Buys land on their supplier's lead time rather than all at once, weeks below the cover target are called out, and any demand the buy could not cover in time gets its own row.

Amber where cover falls under target, red where stock runs out.

07Forecast engine

Combines 11 top-performing statistical models, machine learning and AI (including Chronos-2), running locally on your computer.

Three accuracy leaderboards for the Ecom channel - short, medium and long horizon - each ranking seven models on score, SMAPE, MAPE, WAPE, bias in percent and in units per month, and RMSE. Chronos2 leads the short horizon at 20.2% WAPE against a seasonal naive's 23.8%; over the long horizon Prophet leads at 27.8% and HW_add is last at 152.9%. The ensemble blend behind each horizon is named above its table.

Every model is scored on weeks it never saw, and a different model wins at a different horizon. Chronos-2 leads fev-bench across 100 real-world tasks at a 90.7% win rate, ahead of every statistical baseline and every other foundation model.

Backtest validation: actual demand as a dark slate reference line, with every model traced across the rolling origins from 2019 to 2026 and the ensemble among them. A legend names each series, and the horizon being validated is switchable between short, medium and long.

08Command-line workflow

Fully controllable from local terminal commands or AI agents like Claude Code and Codex, to automate repetitive workflows.

Agent transcript

> Test a scenario with $13M revenue next year and $8M during the
  Black Friday / Cyber Monday sale. Check the shares against
  history, then run it and compare against the base plan.

metzger status
metzger goals set --target 13000000 --year 2027
metzger goals set --target 8000000 --weeks 2026-11-27:2026-12-01
metzger api get "/api/shares?channel=DTC&level=L2&node=Total"
metzger forecast --run-name "FY27 Upside"
metzger allocate --run-name "FY27 Upside buy"
metzger insights run-comparison --run-a "FY27 Base" --run-b "FY27 Upside"

09Scenarios & BI

Compare multiple plans side by side in built-in pivot tables and charts, and export directly back to Excel.

The built-in BI dashboard: an Excel-style free-form pivot with a field list and drop zones for Filters, Rows, Columns and Values. Two saved plans, FY27 Base and FY27 Upside, sit one per row beneath each gender with their months side by side, and the chart above shows the pair diverging from early 2026 where the two plans separate.

Two backdated re-forecasts on the Home screen, each scored as a dense pivot: channel and metric down the left, periods across. Under Q4 Re-forecast, Amazon forecast 77,979 units for 2025 Q4 against 87,413 actual, a -10.8% bias at 38.4% SMAPE and +32% over a seasonal naive. A part-finished period splits in two: 2026 YTD scores the weeks that landed, and 2026 PLAN carries the 661,379-unit projection and scores nothing, so forecast and actual can always be read as a pair.

Every saved plan keeps its own accuracy score, measured on weeks it never saw.

10Goal seeking

Enter a dollar or revenue goal and the system calculates the required unit sales, margins and prices.

Key takeaways

No database required

Runs 100% locally on your machine and works directly with Excel files.

Smart automation

Lets AI build, test and run scenarios headlessly.

Audit-ready

Keeps historical data, manual adjustments and AI projections clearly separated.

Ready to plan with confidence?

Reduce waste. Catch the opportunities. Put your assumptions on screen so the room can agree on them.

See pricing