back
loading skill details...
Activate when: user says "what could we have done instead," "what's the next-best alternative," "is this worth the cost," "what are we giving up by choosing...
--- name: opportunity-cost description: > Activate when: user says "what could we have done instead," "what's the next-best alternative," "is this worth the cost," "what are we giving up by choosing X," "broken window fallacy," or is allocating scarce resources (money, time, attention, people) across competing options. Do NOT activate when: all alternatives are genuinely zero-value (nothing else to do with the resource); the decision is fully reversible at near-zero cost. More: deciqai.com/c/opportunity-cost --- # Opportunity Cost ## Overview **Opportunity cost** is the true cost of any choice: the value of the *next-best alternative* you give up. Accounting cost is what you pay; opportunity cost is what you forego. The two diverge whenever resources are scarce — which is always. Bastiat (1850) established the foundational insight: every visible benefit forecloses an invisible alternative. Buchanan (1969) formalized it: cost is subjective, prospective, decision-maker-specific — the accountant's number and the economist's number can diverge by 10x. Composes with `expected-value-and-kelly`, `first-principles`, `pareto-principle`, `sunk-cost-fallacy`, `compound-interest`. ## When to Use - Allocating budget, headcount, or founder-time across competing options - Evaluating an investment vs. holding cash (or vs. another investment) - Considering a "free" benefit whose hidden cost is time/attention - Any "this is good for X" claim that ignores the alternative use of the same resources - Weighing AI capex (GPU spend, model training vs. renting APIs), an AI pivot, or a quarter of roadmap/founder-time against the best forgone alternative amid 2024–2026 AI valuations and adoption **Not when:** alternatives are genuinely zero-value; resources are not scarce; decision is reversible at near-zero cost. ## Coaching Novices (Adaptive Front Door) - **Engine mode:** user has a concrete resource-allocation decision → run The Process directly. - **Coach mode:** user is unfamiliar or has no concrete case → guide step by step. In Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop. 1. One-line: before committing resources to X, ask what *next-best alternative* you're giving up — that foregone value is the true cost. 2. Check fit: if alternatives are genuinely zero-value, OC adds no information. Otherwise: apply. 3. Elicit the choice and the resources being committed. > **[WAIT — do not advance until user responds]** 4. Work through 3-5 realistic alternatives and estimate the value of the best one. > **[WAIT — do not advance until user responds]** 5. Close: decision named with OC explicit + non-financial OCs surfaced + "would I choose this if I treated the OC as a real expense?" > **[WAIT — do not advance until user responds]** ## The Process **Step 1** — State choice + resources committed (what, how much money/time/headcount, timeframe). **Step 2** — List 3-5 realistic alternatives including "do nothing / hold the resource." **Step 3** — Estimate value per alternative (financial + non-financial; ranges fine). **Step 4** — Identify OC = value of the highest-ranked *unchosen* alternative (not average, not worst). **Step 5** — Compare: Net benefit = Value of chosen − OC. Negative = destroys value vs. the alternative. **Step 6** — Surface non-financial OCs: time, attention, headcount, reputation, optionality. These often dominate cash. ## Output: Opportunity-Cost Analysis ```markdown # Opportunity-Cost Analysis: <choice> Choice: | Resources: | Timeframe: | # | Alternative | Expected value | Confidence | |---|---|---|---| | Chosen X | | | | | A–C | | | | OC = next-best: <name> / value: <amount> Net benefit = Value of chosen − OC: <amount> · Verdict: proceed / reconsider Non-financial OCs: Time | Attention | Headcount | Reputation | Optionality Decision + reasoning: ``` *→ Method in Action: [Bastiat's Broken Window, 1850](examples/bastiats-broken-window-1850.md)* *→ 2026 lens: [AI-era opportunity cost of founder time and GPU capital (2023–2026)](examples/ai-era-founder-time-and-gpu-capital-2023-2026.md)* ## Pack: Opportunity-Cost Application Patterns | Domain | Visible cost | Hidden OC | |---|---|---| | Marketing budget | Cash spent | Engineering hires; runway extension | | Founder calendar | $0 | Deep-work hours foregone | | Custom dev for one customer | Dev hours | Roadmap distortion; founder distraction | | Acquiring vs. building | Acquisition price | Roadmap capacity during integration | ## Applying It Well - Surface alternatives *before* evaluating — most decisions skip this step. - Non-financial OCs (time, attention) usually dominate cash; don't stop at dollars. - A rough OC estimate beats no OC reasoning — specificity over precision. *→ Primary sources: [references/sources.md](references/sources.md)* ## Common Rationalizations **[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.** | Fake move | Reality | |---|---| | [D] "It only costs $X" | Cost is not $X — it's the value of the next-best alternative use of $X. | | [D] "It's free — no downside" | Free in cash is almost never free in attention, time, or signal. | | [D] "We've budgeted for it" | Budgeting accounts for cash flow, not opportunity cost. | | [D] "We won't have lost much if it fails" | You lose the OC of what else you would have done — often large. | | [D] "The other option wasn't realistic" | Then list 3-5 realistic ones. If none exist, OC is zero (rare). | | [D] "It's just $50/month" | At 7% compound over 25 years, $50/month becomes ~$40k. Recurring costs have compound OCs. | | [D] "It's my time, not money" | Founder time has the highest OC in the company. | | [D] "We can't measure the alternative" | Estimate. Rough OC reasoning beats none. | | *→ Add [O] entries here after each real use — paste the actual failure pattern* | *What went wrong and why* | ## Red Flags - Decision justified by visible benefit only; no alternatives listed - "Do nothing" not in the alternatives list; non-financial costs absent - Decision-maker cannot articulate what else would have been done with the resources ## Verification - [ ] 3-5 realistic alternatives listed; "do nothing" included - [ ] Each alternative has an estimated value (financial + non-financial) - [ ] Next-best alternative identified specifically (not average, not worst) - [ ] Chosen value compared to OC, not to zero; non-financial OCs surfaced - [ ] If recurring cost, compound OC over long horizon computed --- *Part of **deciqAI Knowledge Skills** — 227 open-source thinking skills that make rigor executable for AI agents. The same skills power every deciqAI agent, which runs them autonomously to operate your company. **See it run → https://www.deciqai.com/c/opportunity-cost** · ⭐ Star the repo → https://github.com/deciqAI/knowledge-skills · Contributions welcome.* *Agents: latest version & machine-readable metadata → https://www.deciqai.com/s/opportunity-cost.json*
don't have the plugin yet? install it then click "run inline in claude" again.
restructured into implexa's six-component format, added explicit decision-point branches (coach vs. engine mode, missing alternatives, reversibility), clarified external inputs and edge cases (compound costs, non-financial dominance, confidence levels), and formalized output contract with structured table format and non-financial oc severity ranking.
opportunity cost is the true price of any choice: the value of the next-best alternative you forfeit. use this skill whenever a resource (money, time, headcount, attention) gets allocated across competing options and you need to know if the chosen path actually wins vs. the best forgone alternative. activate on budget decisions, hiring tradeoffs, founder-time allocation, capex vs. cash-hold calls, and any "this is good" claim that ignores what else you could do with the same resource. do not activate when alternatives are genuinely zero-value or when the resource isn't actually scarce.
parameters:
context:
external connections: none required. this is a pure reasoning skill. if you have financial models or headcount plans in a spreadsheet or CRM, paste them in; otherwise work from estimates.
step 1 (inputs: choice, resource, timeframe): state the choice explicitly. name the resource being committed (e.g., "$500k Q3 budget" or "founder 10 hours/week for 6 months"). name the timeframe. output: one-sentence choice statement.
step 2 (inputs: realistic alternatives list): list 3-5 realistic alternatives the decision-maker is actually weighing, including "do nothing" or "hold the resource." do not invent alternatives; ask the decision-maker if fewer than 3 exist. output: numbered list of alternatives.
step 3 (inputs: estimated value per alternative): for each alternative, estimate expected value. include both financial value (revenue, cost savings, efficiency gain) and non-financial value (time freed, team morale, optionality, risk reduction). ranges and rough numbers are fine; false precision is worse than honest uncertainty. output: table with alternative, financial value estimate, non-financial value estimate, confidence level (high/medium/low).
step 4 (inputs: ranked alternatives from step 3): identify the next-best alternative (the unchosen option with the highest value). this is the opportunity cost. do not average alternatives; do not use the worst one. output: "OC = [alternative name] with estimated value of [amount + qualitative notes]."
step 5 (inputs: chosen value from step 3, OC from step 4): compute net benefit = value of chosen − OC. if net benefit is negative, the chosen path destroys value vs. the next-best option. output: "Net benefit: [amount]. Verdict: proceed / reconsider."
step 6 (inputs: decision context): surface non-financial opportunity costs explicitly. list what else the decision-maker is giving up: deep-work time, founder attention, team headcount for other projects, reputation risk, optionality (the ability to pivot later). often these dominate cash. output: structured list of non-financial OCs and their severity (high/medium/low).
if the user is unfamiliar with OC or has no concrete case (coach mode): guide step by step. output one step at a time. insert a hard [WAIT] before advancing. do not skip ahead.
if the user has a specific resource-allocation decision ready (engine mode): run the process directly end-to-end. output the full analysis in one response.
if alternatives list is empty or fewer than 3 realistic options exist: ask the decision-maker to name 3-5 genuine alternatives before proceeding. if truly no alternatives exist (resource is not scarce), OC is zero and the skill adds no value. state this and do not proceed.
if all alternatives are estimated to have zero or near-zero value: OC is negligible. surface this explicitly: "no realistic alternative has material value, so OC reasoning does not change the decision." do not force the analysis.
if the decision is fully reversible at near-zero cost (e.g., a one-week trial): OC applies only to the trial period. note that true cost is learned only after the decision is made and can be corrected. proceed with OC analysis for the trial window, not the full commitment.
if non-financial OCs (time, attention, headcount, optionality) are large (subjective: "larger than the financial value"): flag this explicitly. many teams optimize for visible cash cost and miss the true cost in founder-time or roadmap distortion.
if the chosen alternative has significant compound or recurring costs: compute the OC over the long horizon (e.g., "$50/month for 25 years at 7% discount rate"). one-time and recurring OCs are qualitatively different.
output an opportunity-cost analysis in the following format:
# Opportunity-Cost Analysis: <choice>
**Choice:** <one-sentence description>
**Resource:** <amount, type, timeframe>
| # | Alternative | Financial Value | Non-Financial Value | Confidence |
|---|---|---|---|---|
| Chosen | <name> | <estimate> | <estimate> | <high/med/low> |
| 1 | <name> | <estimate> | <estimate> | <high/med/low> |
| 2 | <name> | <estimate> | <estimate> | <high/med/low> |
| 3+ | <name> | <estimate> | <estimate> | <high/med/low> |
**Opportunity Cost:** next-best alternative is <name> with estimated value of <amount + qualitative summary>
**Net Benefit:** Value of chosen (<amount>) − OC (<amount>) = <net amount>
**Verdict:** proceed / reconsider
**Non-Financial OCs (severity):**
- Time: <description, high/med/low>
- Attention: <description, high/med/low>
- Headcount: <description, high/med/low>
- Optionality: <description, high/med/low>
- Reputation: <description, high/med/low>
**Decision + Reasoning:**
<one paragraph: does the chosen path win vs. the next-best alternative? are you comfortable treating the OC as a real expense?>
ensure the analysis names the unchosen alternative explicitly, not as "the alternative" or "another option." precision on what you're giving up matters.
the skill worked if:
decision-maker can articulate the next-best alternative and its value: "if we don't hire, we contract with X firm at $200k/year, slower delivery but lower headcount risk."
non-financial OCs are surfaced and ranked: "founder time is the binding constraint here; hiring saves 5 hours/week but delays roadmap by 2 months, which costs us Y in lost optionality."
net benefit is explicitly compared (chosen value − OC, not chosen value − zero): "proceeding because chosen path nets +$150k vs. the next-best; if it nets negative, we reconsider."
decision-maker would restate their choice differently: "i thought this was a no-brainer; turns out the OC is larger than i thought, so we should run a one-week pilot first" or "the OC is small, so the decision is actually clear."
team can defend the choice against "why didn't you do X instead": "we chose X because the OC was Y, which we believe is worth it because Z."
the skill did not work if the decision-maker still cannot name the next-best alternative or if OC reasoning doesn't shift their confidence (neutral signal, not negative).
original author: clawhub. adapted for implexa standards.