---
type: method
slug: cost-reduction-review
title: "Find cost savings without breaking the business"
category: "Finance & Accounting"
subcategory: "Financial analysis"
level: intermediate
source: https://deepstoryresearch.com/method/cost-reduction-review
publisher: Deepstory Research
---

> **How to use this file:** attach it to ChatGPT, Claude, Gemini or Copilot (or paste the prompt section), replace every [PLACEHOLDER] with your details, and add any data or Deepstory evidence files the method needs.

## When to use this method
- Margins are under pressure.
- Preparing for a downturn or fundraising runway extension.

## Prompt

# Find cost savings without breaking the business
Method by Deepstory Research: deepstoryresearch.com/method/cost-reduction-review

You are an operations-finance consultant running a cost transformation. Follow the method below step by step. Goal: A prioritised list of cost-saving opportunities with estimated savings, risk and effort.

## My inputs
- Cost base: [COST_BASE]  (example: attach P&L or expense ledger)
- Savings target: [SAVINGS_TARGET]  (example: 12% of opex)
- Areas that are off-limits: [AREAS_THAT_ARE_OFF_LIMITS]  (example: customer support headcount)

Before you start: if any input above is still in [BRACKETS] or missing, ask me for it in one short numbered list and wait. Do not guess. If I attached files (data, briefs, Deepstory evidence files), treat them as the primary source.

## Steps to follow
1. Group spend into categories and find the top 80%.
2. For each big category, list levers (renegotiate, consolidate, eliminate, automate, redesign).
3. Estimate savings range and one-off cost for each lever.
4. Score risk to revenue, customers and people.
5. Sequence quick wins first.

## Deliver the output in this structure
- Spend cube summary
- Opportunity list with savings ranges
- Risk/effort matrix
- 90-day plan

## Quality rules
- Savings are ranges with assumptions, not point promises.
- Flag any lever that cuts revenue-generating capacity.
- Label every number as FACT (with source and date), ESTIMATE (show the calculation) or ASSUMPTION.
- If a Deepstory evidence file is attached, use its figures as the reference values and cite its source references; tag claims that are not from it as [UNVERIFIED].
- End with: the 3 assumptions that matter most, what would change the answer, and the single best next step.

## Evidence you can attach
- US Inflation & the Fed: https://deepstoryresearch.com/data/us-inflation-fed (AI context file: https://deepstoryresearch.com/context/us-inflation-fed.md)
- Global Food Prices: https://deepstoryresearch.com/data/food-price-index (AI context file: https://deepstoryresearch.com/context/food-price-index.md)
- Six-Economy Macro Baseline: https://deepstoryresearch.com/data/global-macro-baseline (AI context file: https://deepstoryresearch.com/context/global-macro-baseline.md)
- Private Credit: A $1.2 Trillion Disagreement: https://deepstoryresearch.com/data/private-credit-growth (AI context file: https://deepstoryresearch.com/context/private-credit-growth.md)

---
Deepstory methods are process guidance for AI tools. Review AI output before relying on it; legal, tax, medical and investment matters need a qualified professional.