The board report that eats a day every month

The assembly tax

The report itself is not hard. What takes the day is that the inputs live in four places and none of them agree on formatting.

Revenue comes out of the accounting system. Cash position comes from the bank, and probably from more than one account. Pipeline comes from the CRM, if the client keeps it current, which is its own conversation. Headcount and payroll come from somewhere else again. Then last quarter's deck has to be opened to remember which metrics this board actually asked for, because it is different from your other clients.

None of that is expertise. It is fetching, reconciling and reformatting, and it is the same fetching every month.

What is stable and what genuinely changes

This is the split that decides what can be automated. Get it wrong and you either automate something that needed judgment, or you keep hand-building something that never changes.

• Stable across months. The metric set, the section order, the chart types, the definitions behind each number, and this board's preferences about detail. These change once a year at most.
• Changes every month, mechanically. The values themselves and the period comparisons. Pure fetching.
• Changes every month, and needs you. The commentary. Why margin moved, whether the pipeline number is real, what the board should worry about. This is the part you are paid for.
• Changes per client, permanently. Which metrics matter, how formal the board is, and whether they want one page or fifteen. This is intake, not monthly work.

Cutting the assembly without touching the judgment

The aim is a draft on your desk with the numbers already in place and the obvious variances already flagged, so the day becomes an hour of thinking rather than six hours of gathering and one hour of thinking.

• 1. Write down the report definition once, per client. Metrics, definitions, section order, comparison periods, tone. This is the only genuinely new work, and you do it once.
• 2. Connect the sources rather than exporting from them. Accounting, bank, CRM, payroll. Exporting to spreadsheets is the step that quietly recreates the whole problem next month.
• 3. Generate the numbers and the variances together. Any figure that moved more than a threshold you set gets flagged for commentary, so nothing material reaches the board unexplained.
• 4. Draft commentary from the prior period's narrative. Not to reuse the words, but so the story is continuous. Boards notice when a concern raised last quarter silently disappears.
• 5. You write the judgment calls and the recommendation. Then the finished report goes back into the source set, and next month starts from a better baseline.

The commentary is the part people get wrong

It is tempting to have the whole report written for you, including the analysis. That produces confident prose about numbers nobody checked, which is the specific failure a board will catch and remember.

The useful division is narrow. Draft the descriptive sentences, the ones that restate what changed and by how much, because those are mechanical and boring to write. Leave every causal claim and every recommendation to you.

A good setup makes that division visible in the draft, so what needs your judgment is obvious rather than buried in fluent paragraphs.

Where the numbers come from

Most fractional finance practices run on the client's own stack rather than a dedicated FP&A platform. Accounting in QuickBooks or Xero, a bank feed, a CRM the client half-maintains, and a spreadsheet holding everything the other systems cannot.

That is fine. The workflow should read what is already there rather than requiring the client to adopt anything new. A reporting process that depends on your client changing their systems is a reporting process that will not survive contact with the client.

Client financial data stays on the client's accounts and yours. Nothing is uploaded to a vendor platform and nothing trains anyone's model. For finance work that constraint is not negotiable.

What this does not do

It does not close the books. It reads what the close produced. If month end is late, the report is late.

It does not replace your judgment about whether a number is trustworthy. It will happily report a pipeline figure from a CRM nobody has updated in six weeks, and it will not know that.

It does not pay off on the first client. The report definition takes real time to write down. The return shows up from the second month on the same client, and again on every client after that.

Questions