A board packet is written by management and read by the people responsible for overseeing management. That structural fact governs everything about how it should be built.
The temptation runs one direction only. A packet that shows steady progress produces a comfortable meeting. Nobody instructs anybody to write it that way, and it gets written that way anyway.
Volume as concealment
The most common defect is not a false statement. It is a hundred and forty pages.
A board receiving that much material four days before a meeting will read the summary and skim the rest. Anything difficult, if it appears at all, appears on page ninety-one in a subordinate clause. Technically it was disclosed. Functionally it was buried, and everyone involved understands the difference.
Directors should treat sudden growth in packet length as a signal in itself. Management that has begun sending substantially more material is frequently management with more to explain, and the volume is doing work that a shorter document would not.
What a board actually needs
A board has three jobs: hire and evaluate the executive, ensure the organization remains solvent and lawful, and approve matters of consequence. A packet should serve those and little else.
What has changed since last time. Not a recitation of ongoing activity. The deltas.
What is not going well. Explicitly, in its own section, at the front rather than distributed through the narrative where it can be absorbed.
What requires a decision, and what happens either way. Each with a recommendation, because a board asked to decide without one has been handed management’s job.
The four or five numbers that indicate health. The same numbers every time, so that a director can see a trend rather than a snapshot. Changing which metrics appear is itself a way of managing a board, and directors should notice when it happens.
What is coming that they should know about now. The risk that has not yet materialised is the item boards most often say they wish they had heard about earlier.
The bad news section
Every packet should contain a standing section for things that are not going well, and it should never be empty.
Making it structural removes the judgment call about whether a particular item is serious enough to raise, which is the judgment executives get wrong under pressure. It also normalises disclosure, so that reporting a problem is routine rather than an event that implies management has lost control.
An empty section in a real organization is not a good month. It is evidence that the filter is operating, and a board that receives one should ask what was considered for inclusion and left out.
Why concealment fails the executive too
The argument against managing a board is usually put in terms of duty. The self-interested version is stronger and less often made.
A board that learns of a problem from you, early, while it is still small, is a board that can support you through it. They have time, options, and the sense that they are being dealt with straight.
A board that learns of the same problem from a funder, an auditor, or a newspaper has learned two things: that there is a problem, and that they were not told. The second is the one that ends executives, and it ends them in circumstances where directors feel personally exposed and are therefore least inclined to be generous.
Nearly every executive removed over a concealed difficulty would have survived the difficulty disclosed.
For directors
Reading a packet well is a skill, and a small number of habits do most of the work.
Read the financials before the narrative, so the narrative is being checked against something rather than framing it. Notice what has disappeared since the last packet, since an item quietly dropped is more informative than any item added. Ask what is not in here that you considered including. And treat consistently positive reporting as a question rather than as reassurance.
A director who cannot say what would have to appear in a packet for them to become concerned has not established a standard, and will not recognise the moment when it arrives.
The document is not the relationship
A well-built packet cannot fix a board that is not told things, and a poorly built one is usually a symptom rather than the disease.
The condition that produces good packets is an executive who believes disclosure will be met with support rather than alarm, and a board that has behaved accordingly on the previous three occasions. Where that has been established, the format matters much less than anything written here suggests. Where it has not, no template will substitute for it.
Edited by Patrick J. Wolf, PhD