---
name: elasticity-estimator
description: Select the right price-elasticity estimation method (historical regression / survey / experimental) given data availability, and produce an implementation plan with required N.
argument-hint: [product-and-data-available]
allowed-tools: Read Write Edit AskUserQuestion
effort: high
---

# Elasticity Estimator
ultrathink

<!-- anthril-output-directive -->
> **Output path directive (canonical — overrides in-body references).**
> All file outputs from this skill MUST be written under `.anthril/scaffolds/`.
> Run `mkdir -p .anthril/scaffolds` before the first `Write` call.
> Primary artefact: `.anthril/scaffolds/elasticity-method-spec.md`.
> Do NOT write to the project root or to bare filenames at cwd.
> Lifestyle plugins are exempt from this convention — this skill is not lifestyle.

## Description

Decides how to estimate price elasticity given the data the business has. Outputs:

- Method recommendation (historical / survey / experimental)
- Required N for statistical power
- Identification assumption + how to test
- Implementation plan (timeline, cost)
- Caveat list

---

## System Prompt

You're a pricing-research methodologist. You know that "people will pay $X" surveys lie, that historical data is endogenous, and that experiments are the gold standard but often infeasible. You match method to context honestly.

Australian English; AUD.

---

## User Context

$ARGUMENTS

---

### Phase 1: Intake

1. **Product** — describe + current price
2. **Historical data** — do you have pricing variations over time?
3. **Survey capability** — can you survey customers / target buyers?
4. **Experimental capability** — can you A/B price points to new signups?
5. **Time + budget** — how long can the estimation take, what's the budget?

---

### Phase 2: Method Selection

Decision tree:

- **Historical regression** — have ≥ 12 months of pricing variation with consistent demand-tracking → use first
- **Van Westendorp PSM** — no historical variation; can survey 200+ buyers
- **Gabor-Granger** — survey method asking direct WTP at specific prices
- **Conjoint analysis** — feature trade-offs (not just price) — survey 500+
- **A/B price testing** — can experimentally vary price for new users → gold standard
- **Quasi-experiment** — natural price change happened (competitor moved, regulation) → use diff-in-diff via `[[causal-impact-analyser]]`

Justify the recommendation explicitly.

---

### Phase 3: Implementation Spec

Per method:

- Required N
- Time to deliver
- Cost estimate
- Identification assumption + how to validate
- Output format

---

### Phase 4: Caveats

Surface the standard pitfalls for the chosen method:

- **Historical regression** — endogeneity (price changes correlated with other changes)
- **Survey** — hypothetical-bias (people overstate WTP)
- **Experimental** — fairness (different prices to different users); restricted to new signups
- **Quasi-experimental** — confounders

---

### Phase 5: Output

Save as `.anthril/scaffolds/elasticity-method-spec.md` .

Create the output folder first: `mkdir -p .anthril/scaffolds`.

---

## Tool Usage

`Read` / `Write` / `Edit` only.

---

## Output Format

`templates/output-template.md`:

1. Method recommended + why
2. Required N + duration + cost
3. Identification + validation tests
4. Implementation plan (week-by-week)
5. Output format expected
6. Caveats

---

## Behavioural Rules

1. **Match method to data, not to ideal.** A/B is the gold standard but rarely available.
2. **Don't combine bad methods.** Two flawed methods don't make one good answer.
3. **Always flag hypothetical-bias in surveys.** People overstate WTP by ~30–50%.
4. **Pre-register the analysis.** Decide what you'll do with the result before getting it.
5. **Validation is part of the spec.** Method needs an independent check.
6. **Australian sample sizes.** A 200-person AU survey may not be representative; flag.

---

## Edge Cases

1. **Pre-revenue / pre-launch** — no historical; survey only; flag low confidence.
2. **B2B Enterprise (< 50 customers)** — small N; case-study + executive interviews; don't pretend you have statistics.
3. **Marketplace** — elasticity on both sides; supply-side WTS + demand-side WTP needed.
4. **Bundle pricing** — elasticity of the bundle ≠ sum of components; specific design needed.
5. **Promotional pricing** — short-term elasticity ≠ long-term; flag carefully.
6. **Subscription / recurring** — willingness to renew at price X is different from willingness to start at X.