Skip to main content

build-a-rolling-forecast

Build and maintain a rolling financial forecast from reconciled actuals, explicit operating drivers, scenarios, and measured forecast error. Use when leaders need a regularly refreshed view of profit, cash, liquidity, or capacity.

ソース情報

リポジトリ
protosphinx/sphinxstack
ソースの最終更新活動
2026年7月29日 13:55
検出された SKILL.md の言語
英語
スター
2
フォーク
0

インストール方法

デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。

ソースファイルを確認

インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。

ファイルエクスプローラー
2 ファイル

SKILL.md を表示中

SKILL.md
ソースの指示 · 読み取り専用プレビュー
name
build-a-rolling-forecast
category
money
description
Build and maintain a rolling financial forecast from reconciled actuals, explicit operating drivers, scenarios, and measured forecast error. Use when leaders need a regularly refreshed view of profit, cash, liquidity, or capacity.
# build-a-rolling-forecast Keep assumptions visible and replace forecast periods with actuals on a fixed cadence. ## When to use - Use for monthly or weekly planning when a static annual budget is no longer enough. - Keep accounting results, cash timing, and management adjustments distinguishable. ## Procedure 1. Define entities, currencies, accounting basis, forecast horizon, time buckets, update cadence, owners, and decision thresholds. 2. Load closed actuals and reconcile revenue, costs, working capital, debt, tax, capital spending, and cash to the controlled source. 3. Model revenue through observable price, volume, customer, contract, pipeline, or usage drivers. 4. Model headcount, compensation, vendors, infrastructure, facilities, tax, capital spending, financing, and working capital with timing. 5. Separate controlled inputs, formulas, actuals, overrides, and outputs. Record source, owner, date, confidence, and rationale for each material assumption. 6. Build base, downside, upside, and action cases. Apply coherent scenario changes instead of isolated percentage shocks. 7. Connect profit, balance-sheet movements, and cash. Expose minimum liquidity, covenant, capacity, and funding thresholds. 8. Review formula integrity, signs, currency, period alignment, circularity, double counting, and reconciliation to actuals. 9. Compare each prior forecast with realized results. Attribute error to driver, timing, data, model, or decision changes. 10. Roll the horizon forward, lock the prior version, obtain approvals, and publish decisions plus unresolved uncertainty. ## Done - A versioned forecast workbook or model records reconciled actuals, drivers, scenarios, approvals, and decision thresholds - Formula, cross-statement, cash, scenario, and backtest checks verify the output and quantify forecast error
GitHubで見る