From ai-adoption-playbook
Calculates board-ready ROI for AI tool adoption from founder-provided data across cost efficiency, revenue optimization, and new revenue buckets.
How this skill is triggered — by the user, by Claude, or both
Slash command
/ai-adoption-playbook:roi-calculatorThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Produces a board-ready ROI calculation from the founder's actual data — not industry benchmarks. Separates ROI into three buckets so the founder can tell a complete story. This is a calculation tool, not a strategy session.
Produces a board-ready ROI calculation from the founder's actual data — not industry benchmarks. Separates ROI into three buckets so the founder can tell a complete story. This is a calculation tool, not a strategy session.
Core principle: Use the founder's real numbers. If a number is estimated, label it as estimated. Never substitute industry averages for missing data — flag the gap instead.
Every AI investment produces value in one or more of these buckets:
| Bucket | What it measures | Examples |
|---|---|---|
| Cost efficiency | Time saved, spend reduced | Hours saved per engineer per week, reduced contractor spend, fewer tools needed |
| Revenue optimization | Existing revenue protected or grown | Faster feature shipping, reduced churn from faster bug fixes, shorter sales cycles |
| New revenue | Revenue that wouldn't exist without AI | AI-powered product features, new service offerings, markets entered faster |
Most early-stage AI adoption lives in bucket 1. That's fine — but naming all three buckets shows the board you're thinking beyond cost cutting.
Ask for these if not already available from prior skills:
Cost efficiency ROI:
Annual tool cost = (seats × price/seat × 12)
Annual time value saved = (active users × hours saved/week × 50 weeks × hourly rate)
Cost efficiency ROI = (time value saved - tool cost) / tool cost × 100
Important notes for the founder:
Revenue optimization and new revenue: These require founder-specific data. If the founder has revenue impact data, include it. If not, note the bucket as "not yet measured" — don't estimate.
Symptom: "GitHub research shows 25-30% productivity improvement" used as the basis for ROI. Consequence: Board asks one follow-up question and the number falls apart. Fix: Use only this team's data. If they don't have productivity data, say "we measured time savings of X hours/week (self-reported)" — not "industry benchmarks suggest."
Symptom: ROI denominator includes all seats, but only half are used. Consequence: ROI looks lower than it is, and the real problem (unused seats) gets hidden. Fix: Calculate ROI on active users separately. Show unused seat cost as waste to address.
Symptom: ROI shows AI tools save money vs. doing nothing. Consequence: Board asks "what else could we spend $10K/year on?" and founder has no answer. Fix: Include the cost-per-engineer-per-month and let the board compare to other investments.
Symptom: A metric looks green — "95% of engineers have Copilot" — but the underlying reality hasn't changed. Access ≠ usage ≠ adoption. Consequence: Board thinks adoption is working. Next quarter the numbers don't move and credibility takes a hit. Fix: If the founder's data shows high access but low usage, flag it: "This number measures access, not behavior. The board will ask what people actually do with it."
Symptom: A metric becomes a target and people optimize the number instead of the outcome — e.g., team accepts every AI suggestion to hit "AI-assisted commits" while code quality drops. Consequence: The metric improves, the goal doesn't. Board eventually notices. Fix: Pair every activity metric with an outcome metric. If you report "AI-assisted commits," pair it with "defect rate" or "review cycle time." If they move in opposite directions, flag the tension.
Symptom: Usage spikes during the pilot month when everyone's watching, then drops when attention moves elsewhere. Consequence: ROI calculated from the pilot month overstates the real impact. Fix: If usage data covers only a monitored period, label the number in the output: "(pilot period — not yet confirmed as sustained)." The board sees it's provisional. Don't present pilot-month data as a trend.
Produce the ROI calculation in this exact format:
## AI Investment ROI
**Company:** [name] | **Period:** [timeframe] | **Date:** [date]
### Investment Summary
| Item | Monthly | Annual |
|------|---------|--------|
| [Tool 1] — [X] seats | $[X] | $[X] |
| [Tool 2] — [X] seats | $[X] | $[X] |
| Unused seats ([X] of [Y]) | $[X] wasted | $[X] wasted |
| **Total investment** | **$[X]** | **$[X]** |
| **Effective investment** (active seats only) | **$[X]** | **$[X]** |
### ROI by Bucket
#### 1. Cost Efficiency
- Active users: [X] of [Y] engineers
- Time saved: [X] hours/engineer/week [estimated/measured]
- Total weekly capacity recovered: [X] hours
- Value of recovered capacity: $[X]/year (at $[X]/hr fully loaded)
- **Cost efficiency ROI: [X]%**
- **Breakeven:** Tool needs to save each active user [X] minutes/day to pay for itself
#### 2. Revenue Optimization
[Specific data if available, or: "Not yet measured. To calculate: track feature shipping velocity, bug resolution time, or customer-facing cycle times before and after AI adoption."]
#### 3. New Revenue
[Specific data if available, or: "Not applicable yet. This bucket activates when AI enables product features or services that generate revenue directly."]
### The Math
[Show each calculation step so the board can verify]
### What's Missing
[List any data gaps that would strengthen the case — e.g., "Objective time measurement (current numbers are self-reported)", "Quality impact data (bug rates before/after)"]
[If team stability is "Restructuring in progress": add this line:]
- **Team restructuring in progress.** Headcount changes may inflate or obscure AI-driven savings. If roles were eliminated alongside AI adoption, separate "capacity recovered through AI" from "capacity removed through restructuring" before presenting to the board. Mixed numbers invite hard follow-up questions.
[If team stability is "Restructuring completed": add this line instead:]
- **Team restructured during measurement period.** Baseline headcount changed. ROI calculations use the current team as the denominator, but prior-period comparisons should note the team was larger/smaller then.
### Board-Ready Summary
[2-3 sentences: investment amount, return, and what it means. Use only defended numbers.]
board-narrative-coach — uses this calculation in the board update narrative90-day-plan-builder — Phase 2 begins measuring ROI, Phase 3 consolidates itadoption-scorecard — provides the usage data that feeds the active user countnpx claudepluginhub adimango/ai-adoption-playbookBuild ROI analyses, TCO comparisons, and value calculators for prospects. Use this skill when a rep needs to justify the investment to a prospect, build a business case with numbers, create an ROI model, compare total cost of ownership, says "build me an ROI calculator", "what's the payback period", "help me show the value", "TCO comparison", or when a prospect asks "what's the return on this?". Also trigger when building value engineering tools, cost-benefit analyses, or investment justification documents.
Calculates monthly token savings and ROI for Context Optimizer by AI model using sessions-per-day input, real data or 35% industry waste average. Reports efficiency multiplier, dollar savings, and team scaling.
Guides product managers through evaluating feature investments using revenue impact, cost structure, ROI, and strategic value. Use for data-driven prioritization decisions.