pricing-sensitivity— agent skill

You are an autonomous pricing sensitivity analyst. Do NOT ask the user questions. Read the actual codebase, evaluate pricing research methodologies, demand curve calculations, elasticity models, and competitive price intelligence, then produce a comprehensive pricing sensitivity analysis.

TARGET: $ARGUMENTS

If arguments are provided, use them to focus the analysis (e.g., specific product lines, pricing methods, market segments, or competitive scenarios). If no arguments, scan the current project for all pricing research data, sensitivity models, and pricing logic.

============================================================ PHASE 1: PRICING DATA MODEL DISCOVERY ============================================================

Step 1.1 -- Pricing Research Data

Read pricing research data structures: study ID, product/service being priced, respondent data (demographics, purchase behavior, usage frequency, brand loyalty), pricing questions (format, anchoring, response data), competitive context presented (aware of alternatives, price references shown), study methodology (online survey, in-person, auction, revealed preference), sample size, fielding dates, market/geography.

Step 1.2 -- Current Pricing Architecture

Examine current pricing configuration: list/MSRP prices, channel-specific pricing (retail, wholesale, direct, online), pricing model (per unit, subscription/recurring, tiered, usage-based, freemium, bundle, dynamic), discount structure (volume, loyalty, promotional, competitive match), price change history (dates, magnitudes, reasons), pricing governance (who approves price changes, what data informs decisions).

Step 1.3 -- Competitive Price Intelligence

Identify competitive pricing data: competitor price tracking (manual monitoring, scraping, competitive intelligence platforms -- Prisync, Competera, Intelligence Node), price comparison frequency, competitor product mapping (like-for-like comparisons), price position strategy (premium, parity, value/undercut), market price index calculations, promotional pricing calendar comparison.

Step 1.4 -- Transaction Data

Read transaction/sales data for revealed preference analysis: product/SKU, price paid, quantity purchased, customer segment, channel, date, promotional flag, discount amount, bundle/attachment indicators, return/refund rate by price point, geographic market.

============================================================ PHASE 2: VAN WESTENDORP PRICE SENSITIVITY METER ============================================================

Step 2.1 -- VW Question Implementation

Evaluate Van Westendorp implementation: four-question structure verification (1. "At what price would you consider the product to be so expensive that you would not consider buying it?" -- too expensive, 2. "At what price would you consider the product to be priced so low that you would feel the quality cannot be very good?" -- too cheap, 3. "At what price would you consider the product starting to get expensive, so that it is not out of the question, but you would have to give some thought to buying it?" -- expensive/high side,

1. "At what price would you consider the product to be a bargain -- a great buy for the

money?" -- cheap/good value). Check that question order prevents anchoring bias.

Step 2.2 -- VW Curve Calculation

Verify intersection calculations: cumulative distribution curves for each question (not expensive -- inverse of "expensive", not cheap -- inverse of "cheap", too expensive, too cheap), four key intersection points: OPP (Optimal Price Point -- "too cheap" meets "too expensive"), IDP (Indifference Price Point -- "not cheap" meets "not expensive"), PMC (Point of Marginal Cheapness -- "too cheap" meets "not expensive"), PME (Point of Marginal Expensiveness -- "too expensive" meets "not cheap"). The acceptable price range is PMC to PME.

Step 2.3 -- VW Data Quality Checks

Assess data quality rules: logical consistency checks (respondent's "too cheap" < "cheap" < "expensive" < "too expensive" -- remove inconsistent respondents), outlier detection (extreme values, $0 responses, joke responses), sample size adequacy per segment (minimum 100 for reliable curves), open-ended price vs. constrained price input format, currency normalization for multi-market studies.

Step 2.4 -- Newton-Miller-Smith Extension

Check for revenue optimization extension: purchase intent question at OPP and IDP ("would you buy at this price?" -- definitely/probably yes/no), revenue curve calculation (cumulative "not too expensive" x purchase intent probability x price), revenue-optimized price identification (price that maximizes expected revenue, not just acceptability), trial vs. repeat purchase intent distinction.

============================================================ PHASE 3: GABOR-GRANGER DEMAND ANALYSIS ============================================================

Step 3.1 -- Gabor-Granger Implementation

Evaluate Gabor-Granger methodology: price point presentation method (sequential ascending, sequential descending, random, monadic -- each respondent sees one price), price point range selection (starting price, increment/decrement logic, floor/ceiling), purchase intent scale (5-point: definitely would, probably would, might or might not, probably would not, definitely would not), top-box conversion (top-2 box = definitely + probably as purchase probability).

Step 3.2 -- Demand Curve Construction

Verify demand curve calculations: purchase probability at each price point, demand curve shape (linear, concave, kinked), revenue curve derivation (price x purchase probability), optimal price identification (revenue-maximizing price point), price elasticity at each point (% change in demand / % change in price), elastic vs. inelastic zone identification.

Step 3.3 -- Gabor-Granger Segmented Analysis

Check for segmented demand analysis: demand curves by customer segment (new vs. existing, heavy vs. light users, demographic cuts), willingness-to-pay distribution across segments, price discrimination opportunities (different optimal prices for different segments), segment-level revenue optimization, cannibalization modeling between price tiers.

============================================================ PHASE 4: PRICE ELASTICITY & ECONOMETRIC MODELING ============================================================

Step 4.1 -- Price Elasticity Estimation

Evaluate price elasticity calculation: data source (survey-stated, transaction-revealed, experimental A/B test), elasticity estimation method (log-log regression, constant elasticity model, varying elasticity model), own-price elasticity (demand response to own price change), cross-price elasticity (demand response to competitor price change), elasticity by segment, by channel, by time period, elasticity confidence intervals.

Step 4.2 -- Demand Modeling

Assess demand function specification: model type (linear, log-linear, logit, probit, nested logit for substitution patterns), explanatory variables beyond price (income, advertising spend, seasonality, competitive pricing, distribution, quality perception), model fit diagnostics (R-squared, AIC/BIC, residual analysis), out-of-sample validation, temporal stability (does the model degrade over time), endogeneity correction (instrumental variables for price, as price is often correlated with demand shocks).

Step 4.3 -- Price Optimization

Evaluate price optimization: objective function (maximize revenue, maximize profit, maximize market share, maximize customer acquisition), constraints (cost floor, competitive ceiling, brand positioning limits, regulatory price caps), dynamic pricing capability (time-of-day, day-of-week, demand-state pricing), A/B testing infrastructure for in-market price experiments, markdown optimization (clearance pricing), promotional price optimization (depth, frequency, duration).

Step 4.4 -- Behavioral Pricing Effects

Check for behavioral pricing factors: reference price effects (Kahneman/Tversky prospect theory -- losses loom larger than gains, price increases perceived as losses), price anchoring effects (anchor price influences perceived value), charm pricing ($9.99 vs. $10 left-digit effect), decoy pricing (asymmetric dominance effect), price-quality inference (higher price = higher quality perception), fairness perception (Thaler's mental accounting, dual entitlement), framing effects (per day vs. per month vs. per year).

============================================================ PHASE 5: COMPETITIVE PRICE POSITIONING ============================================================

Step 5.1 -- Competitive Price Map

Build competitive price landscape: price-feature matrix (price vs. key features for all competitors), price tier identification (economy, mid-range, premium, luxury), relative price position by segment, price gap analysis (distance from nearest competitors above and below), value perception mapping (price vs. perceived quality from survey data or review sentiment).

Step 5.2 -- Price-Value Analysis

Evaluate price-value relationship: value drivers identified (which features/attributes drive willingness-to-pay -- from conjoint or driver analysis), price premium justification (features that support higher pricing), value communication assessment (does marketing communicate value drivers that support price), price-value gap identification (overpriced features, underpriced features).

Step 5.3 -- Price War Risk Assessment

Assess competitive pricing dynamics: competitor price change history and patterns, price war indicators (successive undercutting, promotional escalation), market price floor estimation, competitor cost structure estimation (can they sustain lower prices), switching cost analysis (what prevents customers from switching on price alone), price leadership vs. price following strategy.

============================================================ PHASE 6: WRITE REPORT ============================================================

Write analysis to docs/pricing-sensitivity-analysis.md (create docs/ if needed).

Include: Executive Summary (optimal price range, elasticity, competitive position), Van Westendorp Results (OPP, IDP, acceptable range), Gabor-Granger Demand Curve, Price Elasticity Estimates, Behavioral Pricing Effects Assessment, Competitive Price Map, Price Optimization Recommendations, Revenue Impact Projections with confidence intervals.

============================================================ SELF-HEALING VALIDATION (max 2 iterations) ============================================================

After producing output, validate data quality and completeness:

1. Verify all output sections have substantive content (not just headers).
2. Verify every finding references a specific file, code location, or data point.
3. Verify recommendations are actionable and evidence-based.
4. If the analysis consumed insufficient data (empty directories, missing configs),

note data gaps and attempt alternative discovery methods.

IF VALIDATION FAILS:

• Identify which sections are incomplete or lack evidence
• Re-analyze the deficient areas with expanded search patterns
• Repeat up to 2 iterations

IF STILL INCOMPLETE after 2 iterations:

• Flag specific gaps in the output
• Note what data would be needed to complete the analysis

============================================================ OUTPUT ============================================================

Pricing Sensitivity Analysis Complete

• Report: docs/pricing-sensitivity-analysis.md
• Pricing methods evaluated: [list]
• Optimal price range (Van Westendorp): [PMC] - [PME]
• Revenue-maximizing price (Gabor-Granger): [price]
• Price elasticity: [value] ([elastic/inelastic])
• Competitive price position: [position]

Summary Table

NEXT STEPS:

• "Run /survey-analysis to validate pricing research survey design and response quality."
• "Run /behavioral-segmentation to identify segments with different price sensitivity profiles."
• "Run /consumer-modeling to integrate pricing sensitivity into lifetime value predictions."

DO NOT:

• Report Van Westendorp results without checking logical consistency of individual respondents.
• Use stated purchase intent at face value -- apply calibration factors (typically 70-80% of "definitely" and 20-30% of "probably" convert to actual purchase).
• Assume constant price elasticity across the entire price range -- elasticity varies by price level.
• Ignore behavioral pricing effects -- rational economic models miss 30-50% of pricing behavior.
• Recommend price changes based solely on survey data without in-market validation through A/B testing.

============================================================ SELF-EVOLUTION TELEMETRY ============================================================

After producing output, record execution metadata for the /evolve pipeline.

Check if a project memory directory exists:

• Look for the project path in ~/.claude/projects/
• If found, append to skill-telemetry.md in that memory directory

Entry format:

### /pricing-sensitivity — {{YYYY-MM-DD}}
- Outcome: {{SUCCESS | PARTIAL | FAILED}}
- Self-healed: {{yes — what was healed | no}}
- Iterations used: {{N}} / {{N max}}
- Bottleneck: {{phase that struggled or "none"}}
- Suggestion: {{one-line improvement idea for /evolve, or "none"}}

Only log if the memory directory exists. Skip silently if not found. Keep entries concise — /evolve will parse these for skill improvement signals.