Turn a Spreadsheet of Figures Into a Plain-English Findings Report

Finance & Accounting Claude advanced

Paste raw financial figures and get a written analysis: what moved, why it matters, and what to look into next.

When to use it: When you have monthly sales, costs or other financial data in a spreadsheet and need the story it tells — for yourself, a business partner or the bank.
You are a financial analyst writing for an Australian small-business owner who is time-poor and not an accountant. You analyse only the data provided.

<context>
Business: [ONE LINE, e.g. "cafe, single site, 8 staff"]
What this data is: [e.g. "monthly P&L summary for the last 18 months" or "weekly sales by product line"]
Why I'm looking now: [e.g. "profit feels thinner despite busier trade" or "preparing for a bank meeting"]
What counts as significant to me: [e.g. "any move over 10% or $2,000"]
</context>

<data>
[PASTE THE SPREADSHEET DATA — copied cells, CSV, or typed rows. Include column headers.]
</data>

<task>
Before writing, do the analysis silently: check the data for gaps, one-off spikes and inconsistent columns; compute period-on-period changes; separate trend from noise using the significance threshold I gave.

Then write the report:
1. HEADLINE — the single most important finding, one sentence, numbers included.
2. WHAT CHANGED — the 3-6 significant movements, each with: the numbers, the direction, and why it matters in dollar terms to this business. Distinguish "trend" (3+ periods) from "blip" (one period).
3. WHAT'S PROBABLY DRIVING IT — hypotheses clearly labelled as hypotheses, grounded only in patterns visible in the data (e.g. "cost rise coincides with the March supplier change you'd need to confirm").
4. WHAT I'D CHECK NEXT — 3-5 specific look-ups, each tied to a finding (e.g. "pull the supplier invoices for March-May").
5. DATA QUALITY NOTES — anything missing, inconsistent or suspicious in the data itself, listed as [NEEDED: …] or [CHECK: …].
</task>

<output_format>
Under 600 words. Headings as above. Every claim carries its number. No tables unless the data demands one comparison table.
</output_format>

Rules: never invent figures, benchmarks or industry averages. If the data can't support a conclusion, say so plainly. Do not give tax, valuation or investment conclusions — flag those as questions for my accountant. Australian spelling.

Copy the block above straight into Claude — anything in [BRACKETS] is yours to fill in.

Want it tuned to your business? Bring it to the free weekly call and we'll adapt it live.

Join the free call

More finance & accounting prompts

Overdue Invoice Chase Sequence

A 3-touch payment chase that keeps the relationship and gets you paid

Cashflow Forecast Questions Pack

Prepare properly for a cashflow conversation with your accountant or bookkeeper

Supplier Price Negotiation Prep

Walk into a supplier renewal with leverage, numbers and a walk-away plan