Skip to content

Summarize a six-month P&L with trend commentary

Lays out a six-month P&L as a table and adds a three-sentence read on what is growing and what is compressing.

Workflow · Close & Reporting | Role · FP&A Analyst | Intermediate | 6 min | Updated Jun 2, 2026
The prompt

Copy and customize

prompt.txt
You are a senior FP&A analyst preparing a P&L overview for management.

Context: amounts are in the Amount field, P&L lines are in Account Group L2, and
the reporting date is Reporting Month. Use the last six months.

{pl_data}

Task: present the P&L as a table, then read the trend in exactly three sentences.

Output format:
- A table: rows are Account Group L2 lines, columns are the six months plus a
  six-month total. Include Revenue, Gross Profit, and Operating Income subtotals
  where the groupings allow.
- Directly below the table, a three-sentence trend read covering what is growing,
  what is compressing, and the single line that explains most of the movement.
Open in
We’ll copy the prompt and open the chat.
How to use

Run it in four steps

  1. Export six consecutive months of P&L from the Amount field, grouped by Account Group L2 on a consistent Reporting Month.
  2. Paste it into {pl_data}.
  3. Run it for the table plus the three-sentence trend read.
  4. Tie the subtotals back to your source total before circulating; rows with a blank L2 drop out silently.
When to use

When to reach for this prompt

Use this when someone wants the P&L and the takeaway in one artifact, where a bare table is not enough but a full commentary deck is too much. The three-sentence cap is the point: it forces a position rather than a recap. It depends on settled account groupings, so it is better suited to a business whose L2 mapping is stable than one mid-way through a chart-of-accounts cleanup.

Example output

What you can expect back

P&L overview โ€” last 6 months (Amount, by Account Group L2)

LineJanFebMarAprMayJun6-mo
Revenue1,840K1,910K2,050K1,880K1,920K1,820K11,420K
COGS(552K)(573K)(615K)(564K)(576K)(546K)(3,426K)
Gross Profit1,288K1,337K1,435K1,316K1,344K1,274K7,994K
OpEx(1,010K)(1,040K)(1,065K)(1,090K)(1,120K)(1,150K)(6,475K)
Operating Income278K297K370K226K224K124K1,519K

Revenue was broadly flat to slightly down over the half while gross margin held near 70%. Operating income fell about 55% from March to June, driven almost entirely by OpEx, which rose each month. Overall the half points to cost growth outpacing a steady top line, though further analysis would help confirm the drivers.

Limitations · Worth knowing

This prompt has real limitations you should understand.

A table plus a tight read looks authoritative, but both halves rest on the L2 mapping being clean and stable. If a cost moved from one Account Group L2 bucket to another in the fourth month, the table shows one line shrinking and another growing, and the trend sentence narrates a reclassification as a business shift. The model cannot distinguish the two.

There is also a silent-loss problem. Rows with a blank or unrecognized L2 do not raise an error; they simply fail to appear, so a subtotal can fall short of the ledger with nothing on the page to say so. The three-sentence format compounds this by stripping the hedges: "margin held near 70%" reads identically whether the margin was genuinely stable or whether there were too few clean months to say.

For a P&L summary to be safe to forward, every line needs to map to one stable statement category and tie back to the ledger total before the prompt sees it. Consistent account grouping across the full window is the precondition, and it is a data-layer property rather than something a better-worded prompt can supply.

Prerequisites

What your data needs to look like

  • An Amount field at monthly grain across six consecutive months
  • A populated Account Group L2 that does not change category assignments mid-window
  • A Reporting Month field aligned to the same period definition throughout
  • A consistent sign convention for costs, with subtotals that tie to the source total
See it run on real data

See how FinanceOS handles this prompt on real financial data.

Book a 20-minute walkthrough. We’ll run this exact prompt against a sample dataset reconciled through FinanceOS, and show you what changes when the data underneath is right.

Book a walkthrough
The Hub

Join the FinanceOS Hub.

Subscribe to get new sessions, use cases, and AI workflows for finance teams โ€” straight to your inbox, every Monday.

  • Weekly ยท ~4 min read
  • No spam, unsubscribe anytime
  • 2,400+ finance leaders