Before a board meeting, give an AI reviewer the whole board pack along with the decision you are asking for, and ask it to quote every claim your own appendix doesn't support. The useful window is the few days before the pack goes out, while a slide can still change and no director has formed a view.
What directors tend to ask
Directors usually read the pack a few days ahead, often late at night or on a flight, and arrive with notes in the margins. Those notes tend to catch a figure on a slide that doesn't match the appendix, a forecast built on a hiring plan the last meeting cut, or a risk section that calls something manageable after the audit committee has already seen the underlying letter. Many packs make the ask plain and leave out what it costs the company if the board says no.
Directors who back the plan still ask how you will know it has failed and who owns the first month after the vote. Directors who doubt it start with the one number everything else rests on.
The team that built the plan often leaves these answers out of the pack, because everyone on that team already knows them and stopped writing them down.
A worked example
This example is illustrative and doesn't describe a customer. The chief executive of a 60-person software company plans to ask the board to approve a second sales team for mid-market accounts, at about $1.8 million over the next year. The pack is a 14-slide deck and a spreadsheet appendix. Before sending it, the chief executive puts four things into one request:
- The deck and the appendix, attached or exported as text.
- The minutes from the last meeting, where the board asked management to keep at least 18 months of runway.
- The ask in one sentence: approve the hiring plan and budget, with a go or stop review after two quarters.
- One instruction: read as the director most likely to vote no, quote each claim the appendix doesn't support, and work out the runway if mid-market deals close at half the forecast rate.
A review that helps points at a specific gap. It might find that slide 6 shows a 45-day sales cycle while the appendix uses 90 days for every deal over $50,000, which covers most of the mid-market pipeline. It might calculate that the slow case leaves 14 months of runway, below the floor the board set last time, and point out that no slide mentions it. The chief executive can check both against the spreadsheet in a few minutes, then correct slide 6 and add the slow case with a stop condition. A review that says the deck reads well gives the chief executive nothing to check.
A request you can copy
Below are a board pack, the decision I am asking the board to make, and the minutes from the last board meeting. Read the pack as the director most likely to vote against the ask. List each claim the appendix or the minutes don't support, quoting the slide or line. Then list the first five questions that director would ask, and for each one say whether the pack already answers it and where. Rerun the main numbers assuming [the assumption that matters most, for example that deals close at half the forecast rate] and tell me what changes. Don't rewrite the slides.
Leave the appendix and the earlier minutes in. Board objections often come from a gap between the slides and the backup, and a reviewer that only sees the slides can't find it.
How the Pingpong web review app runs it
The web review app sends your request through several models in order. Each later model receives the original request and every earlier answer, with instructions to assess the work so far. With a board pack, an early pass might list unsupported claims and a later pass might check whether the arithmetic behind those objections holds up. You get a final answer and can open each earlier pass.
Each reviewer reads the earlier answers, so a confident mistake in the first pass can carry forward. A model can also misread a spreadsheet that was exported as text. Check every flagged number against the source file before you change a slide.
What a model review can't check
A model doesn't know what was said after the last meeting ended or which director has history with the market you are entering. Disclosure questions and anything touching a pending legal matter belong with your counsel before the pack goes out. The review can give you a list of questions to take into those conversations.
If the meeting is over and you are answering a director's follow-up question, see reviewing your reply to a board question. For a general check on any draft, see getting a second opinion on a draft. For a choice between options that hasn't reached the board yet, see reviewing a decision between options. For the steps in the app, see running your first review.