variance-bridge-analyzer
Skill15 starsChanged 3 months agoMITView on GitHub
Decomposes a budget-to-actual revenue variance into price and volume effects per product via a bundled calculator that reconciles exactly to the total variance, so "we missed by $400K" becomes a specific, actionable story instead of one unexplained number. Use whenever the user needs to explain why actuals differ from plan, is preparing a variance explanation for leadership, or has a variance reported as a single dollar figure with no breakdown of what actually drove it.
In this repository23 other files it gives its agents.
- annual-operating-planSkill
- ar-ap-aging-analyzerSkill
- balance-sheet-health-checkSkill
- board-reporting-pack-builderSkill
- capex-roi-evaluatorSkill
- cash-runway-burn-analyzerSkill
- covenant-headroom-monitorSkill
- credit-risk-scoringSkill
- debt-schedule-modelSkill
- driver-based-budget-builderSkill
- fx-exposure-advisorSkill
- kpi-tree-dashboardSkill
- liquidity-stress-testSkill
- margin-decompositionSkill
- month-end-close-acceleratorSkill
- peer-benchmarkingSkill
- recurring-report-builderSkill
- rolling-forecast-builderSkill
- scenario-planning-engineSkill
- thirteen-week-cash-forecastSkill
- three-statement-modelSkill
- working-capital-optimizerSkill
- zero-based-budget-plannerSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 9
Use whenever actuals diverge from plan and someone needs to know why, not just by how much — replacing a variance report that states …
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Variance Bridge Analyzer
When to use
Use whenever actuals diverge from plan and someone needs to know why, not just by how much — replacing a variance report that states "revenue missed by $400K" with no explanation of whether that came from selling less, selling for less, or a mix of both.
What it does
Decomposes revenue variance into volume effect (the impact of selling more or fewer units than planned, at planned price) and price effect (the impact of realizing a different price than planned, at actual volume) per product via a bundled calculator, with the two effects reconciling exactly to the total variance — no unexplained residual.
Method
- Gather plan and actual volume and price by product — the bridge only works at this level of granularity; a single blended company-wide number can't be decomposed into volume and price effects meaningfully.
- Run the bundled calculator (
scripts/variance_bridge.py), which computes volume effect as the change in units times planned price, and price effect as the change in price times actual units — a standard, consistent sequencing convention that ensures the two effects sum exactly to the total variance. - Check that the bridge reconciles. Volume effect plus price effect should equal the total variance exactly, for every product and in total — if it doesn't, an input is wrong; the whole value of a bridge is that it ties out.
- Read the ratio between volume and price effect to characterize the story: if one effect is more than 1.5x the other, it's primarily a volume story or primarily a price story; if they're closer in size, both levers need investigating.
- Investigate the actual driver behind whichever effect dominates — a volume story might trace to a specific lost account, a channel underperforming, or a competitive loss; a price story might trace to discounting, a mix shift toward lower-priced products, or list price execution failing.
- Report the bridge, not just the conclusion. Showing the product-level table lets a skeptical reader verify the story themselves rather than trusting a one-line summary.
- Carry the finding into the next forecast cycle — if this month's miss was a volume problem in one specific product, next month's forecast should reflect that reality, not just extrapolate the prior plan forward.
Inputs
- Plan and actual volume by product
- Plan and actual price by product
- Config saved as JSON matching the format documented at the top of
scripts/variance_bridge.py
Output format
Product-level table of plan revenue, actual revenue, volume effect, price effect, and total variance; company-level totals; a plain-language read on whether the variance is primarily a volume story, a price story, or both.
Example
A $9,400 total revenue miss looks like a small, uniform shortfall until the bridge shows it's actually two very different stories: Product A grew volume but gave back price (net positive $4,500), while Product B lost significant volume with price roughly on plan (net negative $13,900). The real finding is Product B's volume loss, not a company-wide pricing problem the blended number might have suggested.
Common pitfalls
- Reporting variance as a single dollar figure with no breakdown, leaving leadership unable to act on the finding.
- Computing volume and price effects with an inconsistent sequencing convention, so they don't reconcile to the actual total variance.
- Stopping at "volume was the driver" without tracing to the specific account, channel, or product behind the volume miss.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

