Revenue governance: the cadence that makes a plan real
Every company has a revenue meeting. Almost none of them have revenue governance. The difference is whether anything changes as a result: what an operating review asks, what a charter has to name, and which questions belong weekly, monthly and quarterly.
Summary
- There is one test for whether you have governance or a status meeting: at the end of it, has a decision been made that changes what somebody does next week?
- Governance needs an artifact, or it dies when the person who cared about it goes on leave. The charter names every number, its owner (a person, not a function), the threshold that triggers a conversation, and the forum that reviews it.
- Different questions belong at different frequencies: weekly for deals, monthly for decomposition against plan, quarterly for the plan reset. Collapsing them into one meeting is why that meeting decides nothing.
Every company we work with has a revenue meeting. Almost none of them have revenue governance.
The meeting happens. People attend. Numbers get read out. Somebody says the quarter is going to be tight and everybody agrees it will be tight. Then the meeting ends, nothing has been decided, and the same conversation runs again next week with a different set of numbers and the same absence of a decision.
Governance is the fourth pillar of a revenue operating system, and it is the one that makes the other three real. Planning without governance is a document. Demand generation without governance drifts. Execution without governance decays back to whatever each person finds comfortable. The cadence is what holds them in place.
The difference between a readout and a review
There is a simple test for which one you are running.
At the end of the meeting, has a decision been made that changes what somebody does next week?
If the answer is no, if the output was shared understanding rather than changed action, it was a readout. Readouts are not useless. They are just far more expensive than they look, because they consume the most senior hour in the company and produce nothing an email could not have delivered.
An operating review answers three questions in order, and it answers all three.
What happened? Against plan, not against last month. The comparison that matters is the commitment, not the trend.
Why? Decomposed into the drivers that produce the number: coverage, conversion, cycle time, capacity, realized price. "Sales was soft" is not a diagnosis. "Stage-two-to-three conversion fell from 34% to 21% in the mid-market segment" is.
What changes? A named action, with a named owner, and a date. If nothing changes, say so explicitly: deciding to hold is a legitimate decision. Deciding nothing is not.
Most reviews do the first well, the second badly, and the third not at all.
The FP&A charter
Governance needs an artifact, or it becomes personality-dependent and dies when the person who cared about it goes on leave.
That artifact is the charter: a single document naming what gets reviewed, at what cadence, by whom, and who owns each number. It is unglamorous and it is the difference between a system and a habit.
A workable charter answers four questions for every metric on it. What is the number. Who owns it, an individual, not a function. What is the threshold at which it triggers a conversation. And which forum reviews it, at what frequency.
"Sales owns pipeline coverage" means nobody owns pipeline coverage.
The ownership line is where most charters quietly fail. A name owns it. When a number has no name against it, the review has no one to ask, and the question dissolves into general concern.
The cadence stack
Different questions belong at different frequencies, and collapsing them into one weekly meeting is why that meeting runs for two hours and decides nothing.
- Weekly, the deal and pipeline review. Short. Thirty to forty-five minutes. What moved, what slipped, what is at risk, what needs help. This is a working session, not a reporting session. The output is unblocking, not understanding.
- Monthly, performance against plan. This is where the decomposition happens. Coverage, conversion by stage, cycle time, average deal size, realized price against list, win rate by segment. The point of the monthly is diagnosis: not that the number missed, but which driver moved and what that implies for the next two months.
- Quarterly, the plan reset. Capacity, coverage requirements, segment mix, pricing, territory, headcount. This is where the plan is re-based on what the last quarter actually taught you, rather than defended out of pride.
The most common structural error is running everything monthly. Deals move weekly, so weekly problems get discovered three weeks late. The second most common is running everything weekly, which turns the quarterly strategic question into a topic nobody has time for.
Forecast discipline
A forecast is a claim. Governance is what makes people accountable for the claims they make.
Three practices do most of the work.
- Categorise honestly. Commit, best case, pipeline. Commit means you would stake your credibility on it. If commit routinely comes in below what was committed, the category has stopped meaning anything and the forecast has become a negotiation.
- Tie categories to [exit criteria](/resources/articles/revenue-execution-interested-to-closed), not to feeling. A deal is in commit because specific, verifiable things have happened: the economic buyer has engaged, procurement has been briefed, a start date is agreed. Not because the rep is optimistic.
- Track forecast accuracy as a metric in its own right. Not just whether you hit the number, but how far each forecast was from actual, by person, over time. A rep whose forecast is consistently 30% high is not a bad salesperson; they are a miscalibrated instrument, and calibration is coachable. You cannot coach it if nobody is measuring it.
The four failure modes
In our experience these account for most of what goes wrong.
- The deal that has slipped four times and nobody says it. Everyone in the room knows. The rep reforecasts it into next month again. The absence of a governance norm that permits naming it is itself the failure.
- The review with no data integrity underneath it. If the CRM is filled in the night before, you are governing fiction with great discipline. Governance sits on top of clean data, and where the data layer is broken, the review can only ever be theatre.
- The concession nobody approved. Discounting, extended payment terms, waived implementation, scope expanded after signature. These rarely appear in a revenue review at all, because they are not revenue events in the reporting, and so the margin quality of what is being closed never gets governed.
- The review that only looks backwards. If every minute is spent on what happened, the leading indicators (coverage for two quarters out, capacity ramp, segment mix drift) never get discussed until they are no longer leading.
What it looks like by archetype
The pillars do not change across business models; the instrumentation does.
A sales-led B2B company governs pipeline coverage, stage conversion and forecast accuracy. A D2C business governs by channel, return rates and blended acquisition cost, at a far higher frequency, because the data arrives daily. A hospital governs occupancy, ARPOB, denial rates and payor mix. A manufacturer governs order book, capacity and realization against price list.
Same cadence architecture, different dials. What is constant is the charter: a named number, a named owner, a threshold, and a forum.
Where to start
Most companies we meet do not need a new meeting. They need the meeting they already have to produce decisions.
Start by writing down what is currently reviewed, how often, and who owns each number. The gaps are usually obvious within an hour: numbers with no owner, owners with no threshold, thresholds with no forum.
That is the honest read governance requires, and it is what ROSA is built to produce: a short diagnostic across the five pillars of a revenue operating system, so you can see where your cadence is real and where it is aspirational before you decide what to fix.
How useful was this article?
One tap. It tells us what to write more of.
The next one
Get what we publish next, by email.
Working notes on raising, borrowing, protecting, growing and structuring capital in India. One email a week at most, and you can leave any time.
We use your address only to send this. See our privacy policy.
We store your address to send you these emails and nothing else. See our privacy policy.
Related reading
