A board reporting package is the document you send your directors before a meeting so the meeting itself can be about decisions instead of discovery. That is the entire design brief. Every choice about what goes in it and when it lands traces back to one question: does this help the board direct the company, or just prove it has been busy?
Under Delaware law, the business and affairs of a corporation are managed by or under the direction of a board of directors, except as otherwise provided in the statute or the certificate of incorporation. That is the job description. A board being told what happened is not directing anything, it is attending a presentation. The pack gets the telling out of the way in advance, so the room can do the part that needs a room. This reaches past venture-backed companies: any fast-growing company with outside directors or a PE sponsor has the same problem.
The send-ahead norm, and why it is the whole game
The pack goes out with enough time for a director to actually read it. Not the night before, not at the meeting. Common practice is several days ahead, and the reason is mechanical: if it arrives late, directors read it in the room, and your scarce hour becomes a reading session with commentary. The same logic kills the walkthrough. If you sent it ahead, they read it, so open with the asks and go straight to discussion.
Sending early costs something. Your numbers must be final earlier, which means your monthly close has to be fast enough to support the deadline. That is the unglamorous constraint behind every good board pack, and it is why companies that close their books in a week tend to have better board meetings. The close calendar is the board calendar. The other reason to send early is pre-alignment: a director who reads on Tuesday and emails on Wednesday has told you what Thursday's hard part will be.
An annotated table of contents you can copy
Here is a working structure. The order front-loads what directors read first and pushes detail to the back. Adapt the metrics to your model, keep the shape.
| Section | What goes in it | The question it answers |
|---|---|---|
| 1. The asks | What you need from the board today. Decisions, approvals, introductions, judgment calls. First, not last. | Why are we in this room? |
| 2. Headline | Three to five plain sentences: the state of the business, the one thing that changed, the one thing worrying you. | If a director reads nothing else, what must they know? |
| 3. Cash and runway | Cash on hand, net [burn rate](/glossary/burn-rate), [runway](/glossary/runway) in months, assumptions stated. Include the date the money runs out on the current plan. | How long do we have, and has that changed? |
| 4. Scorecard | The same handful of metrics every period, with plan, actual, and prior period side by side. Same definitions, same order. | Are we on the plan we agreed to? |
| 5. Financials vs plan | P&L, balance sheet, cash flow, against the board-approved budget, with written explanations of material variances. The variance note matters more than the statements. | Where did we deviate, and do we know why? |
| 6. The hard thing | One page on the biggest current risk, written honestly, with what you are doing and what you do not yet know. | What should we be worried about? |
| 7. Functional updates | One page each from product, sales, marketing, people. Progress against goals set last period, not activity logs. | Are the functions doing what they said? |
| 8. Appendix | Cohorts, pipeline detail, cap table, metric definitions. Reference, not narrative. | Where do I dig if I want to? |
If you are venture-backed, some of this is contractual. The NVCA model Investors' Rights Agreement (October 2025 version) sets a delivery covenant for major investors: annual financials within 180 days of fiscal year end, quarterly within 45 days, and, as a bracketed option, monthly within 30 days. It also has management submit an annual budget to the board, and the version the board approves becomes the Approved Annual Budget, which the annual comparison clause asks you to report actuals against, with material differences explained. These are investor information rights rather than board pack rules, and optional model terms rather than law.
The no-surprises rule
A board member should never learn something materially bad for the first time in a board meeting. If a key customer churned, a hire fell through, or the quarter is going to miss, the board hears it when you know it, by email or a call, not in a slide three weeks later.
The reason is not politeness. A surprise forces a director to process news and respond in the same breath, in front of their peers, which produces worse judgment than one who has had two days to think. Saving bad news for the meeting protects nobody: it guarantees the least useful reaction and spends credibility you will want later. If the pack contains a real surprise, the pack was late.
Reporting is not narrating
Two failure modes, mirror images. A pack that only reports hands the board a wall of numbers and makes them interpret it live, badly and out loud. A pack that only narrates tells a confident story with no evidence underneath, which good directors discount immediately. Do both, in that order: state the number, then write the sentence that explains it. Not a bare jump in customer acquisition cost, and not "we are seeing headwinds in acquisition efficiency" floating free of any figure, but the number, the cause, and what you are doing. The interpretation is management's job, and skipping it is how packs get long and useless at once.
Common failure modes
The pack as performance
The tell is polish. Custom illustration, animated builds, a designer's hand on every page. Effort spent on the pack looking good is effort not spent on it being true, and boards learn to read production value as a proxy for something being managed. A useful test: read any page and ask what a director is supposed to do with it. If the honest answer is "be impressed," cut it.
Burying bad news
Bad news migrates. It starts in the headline, gets softened in review, moves to a functional update, then the appendix, then a footnote under a chart. Nobody decides to hide it, it drifts. The fix is structural rather than moral: give the hard thing its own numbered section at a fixed position, every period. A slot that must be filled is harder to quietly empty.
Metrics that change definition every quarter
This destroys trust fastest. If ARR includes services revenue this quarter but not last, or net revenue retention quietly starts excluding a churned cohort, the board cannot compare anything to anything, and cannot tell whether the change was housekeeping or an attempt to flatter a bad number. Write the definitions down once and keep them. If one must change, say so and restate prior periods on the new basis. A redefined metric without restated history is a broken time series, worse than no metric.
Cadence and ownership
Cadence follows your meeting schedule, and many fast-growing companies settle on quarterly meetings with a written monthly update between. The monthly update is not a mini pack. It is a short email: scorecard, cash and runway, what changed, what you need. Its job is to make sure the quarterly pack contains nothing new.
Ownership is what people get wrong. The pack has one owner, the CEO or the senior finance lead, not a rotating committee and not whoever has time. Functional leaders supply pages against an internal deadline well before the send date, and the numbers come from closed books rather than parallel spreadsheets, which is where drift begins. Someone must own the close, the definitions, and the calendar, every period. That standing workload is much of why companies bring in fractional CFO support once board reporting starts to matter, and it is a clear signal the role has become necessary.
Build the template once, then change it as rarely as you can stand. A board pack compounds through repetition: same sections, same metrics, same definitions, same order. A board that knows where to look can spot a change in your business at a glance. A board re-learning your format every quarter never gets past orientation, and you will mistake their silence for confidence.