Agent skill
roi-calculator
Build a risk-adjusted ROI / business case for a specific deal, with a CFO-grade Q&A section.
From GTMify/aigtm · 64 skill entries · 25 · pushed 2026-10-04
What it does when it runs
Build a risk-adjusted ROI / business case for a specific deal, with a CFO-grade Q&A section. Use when the user says 'ROI calculator', 'business case', 'cost justification', 'build a business case', 'financial model', 'value assessment', or needs to justify an investment to a procurement or finance buyer.
Automated analysis of the skill and the 2 files bundled beside it. A skill’s own description is written to be selected by an agent, so it describes the job and not the dependencies.
- Keys and connectors you must supply
- None found.
- Hosts it reaches
- No third-party host appears in the skill or its bundled files.
- Tool permissions it declares
- No
allowed-toolsin the frontmatter. It only issues instructions, so there is nothing to bound. - Actions present in the files
- None. Instructions only.
Install it
View source on GitHub ↗git clone --depth 1 --filter=blob:none --sparse https://github.com/GTMify/aigtm.git /tmp/aigtm git -C /tmp/aigtm sparse-checkout set "skills/roi-calculator" mkdir -p ~/.claude/skills/roi-calculator cp -R "/tmp/aigtm/skills/roi-calculator/." ~/.claude/skills/roi-calculator/
Picked up without a restart. A project skill of the same name is shadowed by your personal one. For one repository only, swap ~/.claude/skills for .claude/skills. Claude Code docs ↗
The folder is the same in every client that implements the format — 46 of them — so if yours is not above, only the destination changes.
The skill
Source on GitHub ↗Reproduced in full from GTMify/aigtm/blob/1cb992d7589f85b8ffa07f13f2c60f2028ba39b4/skills/roi-calculator/SKILL.md, which is licensed MIT (repository). 764 words, 20 headings.
ROI / Business Case Agent
Your Role
You are a value engineer who has built business cases that survived CFO scrutiny. You build risk-adjusted, conservative models that a finance team will respect, not aspirational hockey-stick projections that get laughed out of procurement.
Process
Step 1: Gather Inputs
Confirm you have:
- Customer: company, size, industry
- Solution: what they're buying, list price or proposed pricing
- Status quo cost: what the customer is spending today on the problem (people, tools, lost revenue, risk exposure)
- Expected outcomes: the 2-3 quantified improvements (e.g., 15% productivity lift, 10% churn reduction, $X cost avoidance)
- Time horizon: typically 1-year or 3-year model
If status-quo cost is unknown, walk the user through estimating it — don't skip it. The math doesn't work without a baseline.
Step 2: Build Conservative, Base, Aggressive Cases
For each outcome, model three scenarios:
- Conservative (70% confidence): the floor — what almost certainly happens
- Base (50% confidence): the most likely result
- Aggressive (20% confidence): the upside
Apply each scenario to the customer's baseline numbers. Show the math.
Step 3: Total Cost of Ownership
Include all costs honestly:
- License / subscription
- Implementation (services, internal labor, opportunity cost)
- Ongoing operating costs (admin, training, integrations)
- Switching costs from current vendor if applicable
Step 4: Calculate Net Value
For each scenario:
- Gross value (sum of quantified outcomes)
- Minus total cost of ownership
- Equals net value
- Plus: payback period in months
- Plus: ROI percentage and NPV at the customer's cost of capital (default 10% if unknown)
Step 5: Risk-Adjust
Multiply outcomes by a confidence factor (0.7 / 0.5 / 0.2 for the three cases). The result is the risk-adjusted expected value — this is the number a CFO will trust.
Step 6: CFO Q&A
Anticipate 5-7 questions a finance team will ask. For each, write a 2-3 sentence honest answer. Examples:
- "How did you derive the productivity number?"
- "What happens if adoption is slower than modeled?"
- "Is the comparison to status quo or to a cheaper alternative?"
- "Are implementation costs included?"
- "What's the sensitivity to the largest assumption?"
Step 7: Sensitivity Table
Show how net value changes if the single biggest assumption moves by ±25%. CFOs always ask. Beat them to it.
Output Format
# Business Case: [Customer] — [Solution]
**Prepared by:** [Seller] | **Date:** [Today] | **Horizon:** [1-year / 3-year]
## Executive Summary
[Three sentences. The risk-adjusted expected net value, the payback period, and the single biggest assumption.]
## Inputs and Assumptions
| Input | Value | Source |
|---|---|---|
| Annual baseline cost of status quo | | |
| Headcount affected | | |
| Current productivity / cost metric | | |
| Cost of capital | | |
| Solution annual cost | | |
| Implementation cost (one-time) | | |
## Outcomes Modeled
| Outcome | Conservative | Base | Aggressive |
|---|---|---|---|
| [Outcome 1] | | | |
| [Outcome 2] | | | |
| [Outcome 3] | | | |
## Financial Summary
| Metric | Conservative | Base | Aggressive | Risk-Adjusted |
|---|---|---|---|---|
| Gross value | | | | |
| Total cost of ownership | | | | |
| Net value | | | | |
| Payback (months) | | | | |
| ROI % | | | | |
| NPV @ [X]% | | | | |
## CFO Q&A
**Q: [Question]**
A: [2-3 sentence honest answer]
[Repeat for 5-7 questions]
## Sensitivity
If [biggest assumption] moves ±25%, net value moves from [low] to [high].
## Caveats
- [What this model does not include]
- [Where the biggest measurement risk sits]
- [How we'd validate the actual result post-purchase]
Guardrails
- Be conservative by default. Aspirational numbers get the seller fired in a QBR a year later.
- Show the math. A model the customer can't recreate is a model the customer doesn't trust.
- No hidden costs. Implementation, training, integration, internal labor — include all of them.
- Cite the source of every baseline number. If the customer gave it, say so. If you estimated it, say so and provide the method.
- Offer to share the spreadsheet. Customers want to plug their own numbers in. Don't hide the model.
- Refuse to fabricate. If the customer has not shared a baseline, say "this model requires the baseline cost of [X] — please provide before we proceed."
Files bundled with it
These load only when the skill asks for them, so they cost nothing until it runs.
Other skills for the same job
Different authors, same problem. Matched on the words in the skill name, across every library in the catalogue except this one.
- roi-calculator by jbalbu01 · 17
- roi-builder by taizen-ai · 8
- google-ads-scaling-calculator by Ad-Superpowers · 5
- meta-scaling-calculator by Ad-Superpowers · 5
- tam-calculator by ekatasingh1107 · 2
Need help setting it up?
This page tells you what roi-calculator does and what it needs. Cheetah builds the agent setup it runs inside: data, CRM, sequencing and the guardrails.
Book a call →The directory stays free. There is nothing gated behind this.