What does a board actually want from the forecast?
A board wants to know whether you understand your own business well enough to predict it, and whether you will tell them early when you do not. The number matters less than the reasoning attached to it.Directors have seen many companies miss a quarter. What separates a routine miss from a governance problem is whether management saw it coming and said so. A forecast presentation is a credibility exercise before it is a reporting exercise.
That shapes the content. Show the call, show how the call was constructed, show the variance history that says whether your calls have been reliable, and show the specific risks with names attached.
What slides do you need?
Three in the main deck, each answering a distinct question. Everything else belongs in the appendix.| Slide | Question it answers | Source |
|---|---|---|
| Number against plan | Are we going to hit it, and by how much | Forecast model plus rep-submitted roll-up |
| Composition of the quarter | Where does that revenue come from | Pipeline decomposition by source |
| Risks and corrections | What could break, and who owns the fix | Deal-level risk register |
Add one line to the first slide that most teams leave out: your called number against your closed number for the last four quarters. Volunteering your own accuracy record changes how the current call is received, because a board evaluating a forecast has no way to weight it without knowing whether your previous forecasts held. Directors who have to ask for that history read the omission as an answer.
How do you show the composition of the quarter?
Break the number into the three paths revenue can arrive by, and forecast each one separately.| Source | Definition | Typical failure |
|---|---|---|
| Carry-over | Deals already in pipeline on day one with close dates this quarter | Overtrusted, most of it does not close on time |
| In-quarter creation | Deals created, qualified, and closed inside the quarter | Under-modeled, often invisible on day one |
| Pull-forward | Deals brought in early from future periods | Discounted, and it borrows from next quarter |
Pull-forward deserves its own line because it has a cost. Rescuing a current quarter by pulling deals from the next one usually means discounting, and it removes revenue from a period you have already committed to. Boards should see that trade made explicitly rather than discover it when the following quarter comes in light.
Why should you avoid leading with pipeline coverage?
Coverage is an input, not a conclusion, and presenting it as the answer masks the risk it is supposed to reveal.The standard is 3x to 5x, and across ORM's customers coverage ranges from 1.4x to 5x with most sitting near 3.5x. Directionally it can be predictive in stable conditions. It is also dangerously incomplete, because it says nothing about the composition of what it counts.
A company can hold 4x coverage and miss badly when the pipeline is concentrated in early stages, dependent on a few large deals, inflated by stale opportunities, owned by the wrong reps, or built on close dates that keep moving. A company can start with thin coverage and outperform on a strong in-quarter motion.
Total pipeline coverage without context is the metric that creates the most noise in a board setting. It makes the room feel informed while hiding the actual risk. Present coverage segmented by stage, age, and owner, or explain in one line why the 3x coverage rule is not the basis of your call.
What do you say when you are going to miss?
Say it in the meeting where you first know, name the mechanism, and bring the correction with you.Forecasts miss because something in the business or the market changed and the plan was built on assumptions that no longer hold. Name which one applies:
- A competitor entered and created pricing pressure, so average deal size fell. - Rates rose, capital deployment slowed, buyers cut cost, and win rates declined with them. - Uncertainty extended decision cycles, so deals took longer from qualified to closed. - Territories changed and execution suffered while coverage still looked healthy.
Each explanation carries a different correction, which is why naming it matters. "We are behind on pipeline" is a symptom. "Our average closed deal size fell against a stable win rate after a competitor entered the mid-market" is a diagnosis, and it tells the board what you are going to do next.
Timing is the other half. Getting the forecast right in the final week of a quarter helps nobody, because the quarter has already happened. The value is in knowing the likely shape of the quarter on day one, early enough to act. Boards judge you on when you knew, not on when you reported.
How do you handle generated numbers in a board deck?
Require traceability on every figure, or leave it out.The gap in AI-assisted reporting is trust and traceability. When a model builds your slides, the validation work needed to confirm the numbers costs as much as building the deck manually. That trade only improves when the output can point back to the source of truth that produced each figure.
Set the standard before the deck exists. Every number in a board presentation links to a query, a definition, and a set of records that a director could ask to see. Numbers that cannot make that trip stay in the appendix or stay out entirely. That rule protects the presentation, and it protects the forecast accuracy record you are asking the board to believe.
Frequently Asked Questions
How many slides should the revenue forecast take in a board deck?
Three in the main deck and the rest in the appendix. One slide for the number against plan, one for the composition of the quarter, one for the risks with owners. Boards that want more detail will ask, and the appendix should be able to answer without a rebuild.
Should you present a single forecast number or a range?
Present a single number as the call and a range as the distribution around it. A range on its own reads as an unwillingness to commit. A number on its own hides the risk. State the call, then state what has to happen for the low end and the high end to occur.
What do you tell the board when you know you will miss?
Tell them early, name the mechanism rather than the symptom, and bring the correction plan to the same meeting. Boards react badly to surprise, not to bad numbers. A miss disclosed in week four with a plan is a management update, and the same miss disclosed at quarter end is a credibility problem.
How do you show forecast credibility to a board?
Publish your called number against your closed number for the last four quarters on the same slide as the current call. A track record of accuracy is the only thing that makes the current number believable, and volunteering it is more persuasive than being asked for it.
Can you use AI to build board reporting numbers?
Only when every number traces back to its source. The validation work on an unverified AI-generated figure costs as much as building the slide yourself. Require that any generated number links to the query and the underlying records before it enters a board deck.
See how ORM turns these insights into action
ORM builds custom revenue forecast models for B2B SaaS companies. Not dashboards. Prescriptive analytics that tell you what to do next.
Schedule a Demo