Revenue operations charter
How it works
The rules the charter runs on.
Six mechanisms make the whole product: what the document is, how a signature binds, how a version is numbered, where a goal comes from, how a forecast is scored, and how the finished thing leaves the building. This page is generated from the same definitions the app enforces.
01 · The document
The charter is the document, not a dashboard.
Everything the org agrees on lives in one published, numbered document. Each section is edited in a focused panel — the Edit button beside its heading — and each can be included in or excluded from the published charter.
- Six parts, in the order the story is true. The company, then who it sells to, then how they move, then what is counted — and only then the numbers it has committed to, beside the chain that proves them.
- Numbered, and renumbered. Exclude the tech stack and the sections below it move up; the contents block and every cross-reference follow.
- Empty sections never print. A signed document does not carry instructions telling its readers to go and use an app.
- Your logo, your fiscal year. The cover carries the company that agreed to it, the version it is on, and the alignment score at that version.
- Grouped the way you named it. Stages appear under the deal and customer lifecycles you defined, as bands inside the funnel section — not as a flat list somebody has to re-sort in their head.
Every section above is generated from the same list the app publishes — exclude one and the numbering closes up behind it.
02 · Signatures
A signature is pinned to the wording it approved.
Approving a definition records an e-signature against a hash of the exact content approved. Edit that definition afterwards and the signature is marked stale automatically — there is no way to change what somebody agreed to without it showing in the document.
- Who must sign is derived, not assigned. A handoff needs both sides; a stage needs its gate owners; the goals section needs every department.
- Signing happens on the line. Either on the signature line in the document, or in the sign-offs & ratification matrix.
- Only the owner ratifies. Ratification is the one act that freezes a new agreed baseline — and the one that bumps the major version.
From the document · § 7 metrics, redlined against v2.0.0
| Qualified opportunity A deal with a named economic buyer, a written need and a date. | PNMT |
| Pipeline coverage Open pipeline in the quarter divided by the quarter’s | DRF |
| Activation, day 30 An account that has completed onboarding and used the product in week four. | CS |
One edit, two consequences: Dana’s signature reads stale until she looks again, and the charter’s alignment score drops to 86%.
03 · Versions
Version numbers are computed, never chosen.
Saving a version freezes an immutable snapshot of every definition. The next version’s number comes from diffing that snapshot against the live charter — nobody types it.
Pick any two versions and the document redlines itself between them: struck-out wording, underlined replacements, section by section, and grayscale-safe when you print it.
From the document · version history
- v2.1.0Added the Expansion stage and its three metricsminor · structure
- v2.0.0Ratified by the owner — the agreed baselinemajor · ratified
- v1.4.1Reworded the S2 → S3 gate criteriapatch · wording
- v1.4.0Customer Success added as a departmentminor · structure
Editing wording is 1.4.0 → 1.4.1. Adding or removing anything structural is 1.4.x → 1.5.0. Only ratification makes 2.0.0.
The three bumps, as the app computes them
Major — 2.0.0
The owner ratifies the charter. Ratification freezes the agreed baseline; nothing else bumps major.
Minor — 1.1.0
Structure changed: a stage, department, handoff, metric, offering, audience, stack tool, or dated goal was added or removed since the last version.
Patch — 1.1.1
Wording only: definitions, criteria, goals, or names edited with the structure intact.
04 · Goals
There is no field where you type a goal.
A dated goal states the money and the date. Everything under it is derived: campaigns commit targets against specific metrics, those targets ramp into the monthly goal lines on every chart, and the goal inventory is the automatic roll-up. If a number matters, a campaign has to own it.
From the document · § 10 goal inventory
| Metric | Owner department | Goal (start → end) | Via campaign | Window |
|---|---|---|---|---|
| Recognize $1,200,000 by Dec 2026 — New logo ARRtargets cover $1,240,000 of $1,200,000 | ||||
| How this is derived ratable over 12 months · full value requires booking by Jan 2026 · a contract booked this month recognizes 33% of its value by Dec 2026 · needs $3,610,000 booked ≈ 74 deals at $48,800 · at a 62-day sales cycle those deals must enter the pipeline by Oct 2026. | ||||
| Bookings · new business | Sales | $210,000 → $340,000 | Enterprise land, H2 | Jul – Dec 2026 |
| S2 → S3 conversion | Marketing | 40.4% → 47.0% | Evaluation content refresh | Sep – Dec 2026 |
Nobody edited a cell in that table. The goal row, its derivation prose and both target rows are computed from the dated goal and the campaigns pointed at it.
Attainment is pace-aware everywhere
On pace — 95% and up
Measured against the goal ramped to today, not the end-of-window number, so a back-loaded campaign is not scored as if it were linear.
Caution — 80% to 95%
Behind the ramp but inside recoverable distance. The same band drives the chip in the grid, the chart and the goal inventory’s coverage note.
Off pace — below 80%
Every band pairs its colour with a number and a word, so the verdict survives a grayscale printout and a colour-blind reader.
05 · Measurement
The forecast is frozen before the month it predicts.
Every metric carries its own basis — period-over-period or rolling, with a window — and its own forecast method, set in the metrics editor and signed by the function that owns it. The basis drives the computed baseline; the method drives the line.
From measurement · prediction scoreboard
| Month | Frozen forecast | Actual | Error | Goal attainment |
|---|---|---|---|---|
| Aug ’26 | $68,100 | $68,966 | +1.3% under | 103% of $67,000 |
| Jul ’26 | $52,400 | $47,903 | −9.4% over | 89% of $53,800 |
| Jun ’26 | $49,900 | $49,873 | on the nose | 99% of $50,400 |
| May ’26 | $51,200 | $53,400 | +4.1% under | 77% of $69,000 |
Model error carries no verdict colour — being wrong about a model is not a business outcome. Attainment does, because it is one.
Why that table cannot be edited after the fact
Frozen, then scored
On the first data save of each month, next month’s forecast and goal are written immutably. Later they are scored against what actually happened.
Self-explaining
Each forecast shows its arithmetic in your own numbers, so a number nobody in the room can explain never reaches a slide.
Backtested
Every method is walk-forward tested against your own history, per metric family, and the most accurate one is flagged.
06 · Sharing & exports
A charter that cannot leave the app is not an agreement.
The published charter goes out four ways, and every one of them honours the sections you chose to include and drops every editing and signing affordance on the way out.
A revocable link
One unguessable URL renders the charter read-only with no sign-in, and is never indexed. Rotate it or switch it off from the document’s Share menu.
Print and Word
The print stylesheet is a first-class surface — page breaks controlled, redlines grayscale-safe. Word export is a real .docx, not an HTML file with a new extension.
Structured, for machines
ProseMirror JSON for the document itself, a REST API for charters, metrics, time-series and campaigns, and an MCP server so a chat can drive all of it.