agentleFS
liyuanbo1024liyuanbo1024/marketing-science-writing/SKILL.md

marketing-science-writing

Skill6 starsChanged 5 months agoMITView on GitHub

Use when writing or revising marketing science papers targeting Marketing Science, JMR, JM, JCR, JAMS, IJRM, QME, Marketing Letters. Covers the full pipeline: substantive problem framing, consumer utility modeling, identification strategy design, structural estimation (BLP, GMM), field/lab experiments, conjoint analysis, counterfactual simulations, and marketing implications with actionable managerial recommendations.

3 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions3 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 3

    1. Aradotso/marketing-skills★ 10same name
    2. reason-machines/marketing-skills★ 10same name
    3. wonhp1/claude-skills★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    The Identification Narrative: Every marketing paper must clearly articulate:

    line 117
  2. Distinctive:

    Goal: Write the full paper with marketing implications.

    line 300

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Marketing Science Paper Writing

Overview

Marketing Science papers follow a distinct pipeline: Substantive Problem → Consumer + Firm Model → Identification Strategy → Estimation/Experiment → Counterfactual Analysis → Marketing Implications. Unlike MS&E papers that prioritize analytical derivation, marketing papers are evaluated on the credibility of the empirical identification and the managerial actionability of the counterfactual.

This skill provides journal-specific guidance for 8 flagship marketing journals, covering structural modeling, causal inference, field experiments, conjoint analysis, and consumer analytics.

Core Principle: A marketing paper is an empirical argument about consumer or firm behavior with testable implications. Every model, regression, and experiment must answer: "What should a Chief Marketing Officer do differently because of this finding?"


When to Use This Skill

  • Drafting a paper targeting Marketing Science, JMR, JM, JCR, JAMS, IJRM, QME, or Marketing Letters
  • Designing a structural model of consumer demand (BLP, dynamic structural, discrete choice)
  • Planning identification strategy (IV, DID, RDD, field experiment, lab experiment)
  • Writing up conjoint analysis, demand estimation, or field experiment results
  • Conducting counterfactual policy simulations (merger, tax, product introduction)
  • Formulating substantive marketing implications from empirical results
  • Addressing endogeneity concerns in observational marketing data

Journal Selection Quick Reference

Tier 1: Top Marketing Journals

JournalCore IdentityIdeal Paper Profile
Marketing Science (MktSci)Quantitative, model-driven, structuralStructural model + novel data + counterfactual showing substantive managerial impact. ~35 pages.
Journal of Marketing Research (JMR)Empirical rigor, identification credibilityWell-identified study (experiment/quasi-experiment) + clear managerial recommendations. ~40 pages.
Journal of Marketing (JM)Broad strategy, conceptual + empiricalConceptual framework + multi-method validation; strong on marketing strategy. ~40 pages.
Journal of Consumer Research (JCR)Consumer psychology, behavioralTheory-driven lab/field experiments testing psychological mechanisms. ~40 pages.

Tier 2: Strong Field Journals

JournalCore IdentityIdeal Paper Profile
JAMS (J. Academy of Marketing Science)Broad scope, reviews valuedConceptual + empirical; strong review papers; AMA-affiliated
IJRM (Int. J. Research in Marketing)European tradition, diverse methodsAll methods welcomed; strong special issue culture
QME (Quantitative Marketing & Economics)Structural IO approachRigorous structural econometrics; smaller volume, higher technical bar
Marketing LettersShort, sharp empiricalConcise empirical contributions (~20 pages); replication-friendly

Decision Flowchart

What is your primary method?
├── Structural econometric model (BLP, dynamic) → Marketing Science, QME
├── Field experiment (A/B test, RCT) → JMR, Marketing Science
├── Lab experiment (psychological mechanism) → JCR, JMR
├── Multi-method empirical (survey + secondary) → JM, JAMS
├── Conceptual/theoretical framework → JM, JAMS, JCR
└── Short empirical note → Marketing Letters

The Marketing Paper Structure

Unlike IMRAD or MS&E papers, marketing papers have a distinctive structure driven by the identification narrative:

SectionTypical PagesKey Content
Abstract150-200 wordsSubstantive problem, method, key finding, substantive marketing implication
S1 Introduction2-3 pagesMotivating marketing phenomenon → Research gap → Identification approach → Contributions
S2 Conceptual Framework / Model3-6 pagesConsumer utility specification, firm objective, market equilibrium, hypotheses
S3 Data & Identification4-8 pagesData sources, sample construction, identification strategy, institutional context
S4 Estimation / Experiment Design3-6 pagesEstimator specification, experimental protocol, diagnostic tests, main results
S5 Counterfactuals / Robustness3-5 pagesPolicy simulations, heterogeneity analysis, alternative specifications, placebo tests
S6 Discussion & Marketing Implications1-2 pagesWhat should marketers DO differently? Theoretical contributions. Limitations.
References—40-80 papers typical
AppendixUnlimitedAdditional robustness, alternative specifications, simulation details

Methodology Guide by Approach

1. Structural Econometric Modeling

See references/structural-modeling.md for comprehensive guidance.

The BLP Framework (Berry-Levinsohn-Pakes, 1995)—the workhorse for aggregate demand estimation:

$u_{ijt} = \alpha_i p_{jt} + \beta_i x_{jt} + \xi_{jt} + \varepsilon_{ijt}$

Key elements:

  • Consumer heterogeneity via random coefficients ($\alpha_i, \beta_i$)
  • Price endogeneity addressed via instruments (cost shifters, Hausman instruments, BLP differentiation IVs)
  • Demand inversion via contraction mapping: $\delta_{jt} \leftarrow \delta_{jt} + \ln s_{jt} - \ln s_{jt}(\delta, \theta)$
  • Supply side: Bertrand-Nash FOCs provide additional GMM moments
  • Counterfactuals: merger simulation, tax incidence, product entry/exit

Dynamic Structural Models:

  • State variables: brand loyalty, inventory, advertising stock, learning
  • Solution: nested fixed point (Rust) or CCP (Hotz-Miller)
  • Applications: brand choice dynamics, new product diffusion, customer relationship management

2. Causal Inference / Reduced-Form

See references/causal-inference.md for comprehensive guidance.

Primary Designs in Marketing:

DesignTypical ApplicationKey Reference
Difference-in-DifferencesAd campaign effects, store entry, policy changesCallaway & Sant'Anna (2021); Sun & Abraham (2021)
Regression DiscontinuityPrice thresholds, eligibility cutoffs, geographic boundariesCalonico et al. (2014)
Instrumental VariablesPrice endogeneity, advertising endogeneityAngrist & Krueger (2001)
Synthetic ControlSingle-market interventionsAbadie et al. (2010)

The Identification Narrative: Every marketing paper must clearly articulate:

  1. The causal relationship of interest
  2. The source of endogenous variation
  3. The source of exogenous variation (the "natural experiment")
  4. The identifying assumptions
  5. How assumptions are tested (balance tests, placebo tests, falsification)

3. Field Experiments

See references/field-experiments.md for comprehensive guidance.

Design Elements:

  • Randomization unit: individual customer, store, DMA (Designated Market Area)
  • Stratification and blocking for balance
  • Power analysis to justify sample size
  • Pre-registration (AEA Registry, AsPredicted, OSF)

Reporting Standards:

  • CONSORT-style flow diagram showing participant flow
  • Balance table comparing treatment/control on pre-treatment characteristics
  • Intent-to-treat (ITT) as primary estimate
  • Treatment-on-the-treated (TOT) via instrumental variables for non-compliance
  • Multiple testing correction (Bonferroni, Holm, or Benjamini-Hochberg)

4. Conjoint Analysis

See references/conjoint-analysis.md for comprehensive guidance.

Design: Fractional factorial, random designs; incentive-aligned via BDM mechanism Estimation: Hierarchical Bayes multinomial logit; WTP from part-worths Validation: Holdout prediction accuracy; face validity of attribute importances


Identification: The Make-or-Break Dimension

Marketing Science and JMR reviewers prioritize identification credibility above all else. A paper without a clearly articulated identification strategy is desk-rejected.

Identification Checklist

  • Source of exogenous variation is clearly described with institutional detail
  • First-stage is strong (F > 10 for IV; visual jump for RDD)
  • Parallel pre-trends are shown visually (event study plot for DID)
  • Balance table compares treatment/control on pre-treatment observables
  • Placebo/falsification tests: fake treatment at different time, different place, different outcome
  • Robustness to: alternative samples, measures, specifications, estimators
  • Threats to identification are explicitly discussed, not hidden
  • Oster (2019) bounds or Altonji et al. (2005) ratio test for selection on unobservables

Common Rejection Reasons in Marketing Journals

ReasonFrequencyPrevention
Weak identification / unconvincing instrumentsVery HighShow first-stage F, discuss excludability, multiple instruments
Endogeneity not adequately addressedVery HighExplicit identification section; don't rely on fixed effects alone
"So what?"—no substantive marketing implicationHighConcrete counterfactual; what changes in practice?
Incremental over existing literatureHighClear differentiation table; novel data or context
Model not connected to dataMediumModel should motivate specification; don't estimate a kitchen sink
Mechanical application of methodMediumMethod serves the question; contribution is insight, not technique

0-to-Draft Pipeline: Five-Stage Workflow

Stage 1: Substantive Problem & Positioning

Goal: Identify the marketing problem, articulate the gap, select target journal.

Agent actions:

1.1 Understand the marketing phenomenon
    - Ask: What is the substantive marketing problem? What decision does it inform?
    - Identify the core consumer/firm trade-off
    - Classify: Structural / Reduced-form / Experimental / Conceptual

1.2 Literature gap analysis
    - Search for 10-15 closest papers in target journals
    - Build a Gap Table: rows = key references, columns = dimensions (data, method, context, finding)
    - Identify the empty cell this paper fills
    - Output: draft gap table with explicit differentiation

1.3 Journal selection
    - Match paper profile to journal identity
    - Consider: MktSci (structural + novel data) vs. JMR (identification + relevance) vs. JM (strategy + multi-method)
    - Output: recommended journal + 1-2 fallback options

1.4 Contribution statement
    - One sentence: "Using [data/method], we study [phenomenon] and find [key insight]. For marketers, this means [implication]."
    - Draft 3-5 numbered contributions
    - Output: contribution list

1.5 Writing plan
    - Outline paper structure with estimated page allocation
    - Identify reference files to load at each stage

Reference to load: references/journal-characteristics.md

Stage 2: Model & Identification Design

Goal: Specify consumer/firm model and articulate empirical identification.

Agent actions:

2.1 Consumer utility specification
    - Specify indirect utility: u_ijt = f(price, characteristics, heterogeneity)
    - For structural: random coefficients, outside good, market definition
    - For reduced-form: outcome equation, treatment definition

2.2 Firm behavior (if applicable)
    - Pricing: Bertrand-Nash, uniform pricing, price discrimination
    - Advertising, entry, product line decisions

2.3 Identification strategy
    - Source of exogenous variation: what quasi-experiment or instrument?
    - Identifying assumptions and how they're tested
    - Institutional details that make the strategy credible
    - Output: identification narrative (~1 page)

2.4 Data description
    - Data sources, sample construction, summary statistics
    - Key variables: construction, sources, limitations
    - Output: §3 Data & Identification draft

Reference to load: references/structural-modeling.md or references/causal-inference.md

Stage 3: Estimation & Results

Goal: Implement estimator, present results with diagnostics.

Agent actions:

3.1 Estimator specification
    - Structural: GMM objective, instruments, weighting matrix, optimization
    - Reduced-form: regression specification, fixed effects, standard error clustering
    - Experimental: ITT specification, randomization check

3.2 Results presentation
    - Main results table with diagnostics
    - Economic significance: elasticities, WTP, treatment effect magnitude
    - Not just statistical significance (p-values)
    - Output: §4 Estimation / Results draft

3.3 Diagnostic tests
    - First-stage F, overidentification J-test, DWH endogeneity test
    - Balance table, attrition analysis, parallel trends

Stage 4: Counterfactuals & Robustness

Goal: Conduct policy simulations, demonstrate robustness, show mechanism.

Agent actions:

4.1 Counterfactual simulations (structural papers)
    - Merger: change ownership matrix, recompute equilibrium
    - Tax/subsidy: add per-unit tax, compute pass-through
    - Product introduction/removal: change choice set
    - Report: compensating variation, profit changes, bootstrap CIs
    - Output: §5 Counterfactuals draft

4.2 Robustness analysis (all papers)
    - Alternative DV, IV, sample, specification, estimator
    - Placebo tests
    - Oster bounds for selection on unobservables
    - Output: §5 Robustness draft

4.3 Heterogeneity / Mechanism
    - Who is most affected? (demographic, behavioral, contextual)
    - Through what channel? (mediation, moderation, process evidence)

Reference to load: references/structural-modeling.md (counterfactuals) or references/causal-inference.md (robustness)

Stage 5: Writing & Assembly

Goal: Write the full paper with marketing implications.

Agent actions:

5.1 Introduction drafting
    - P1: Motivating marketing phenomenon (real company, real numbers)
    - P2: Why existing literature/approaches fail
    - P3: Our approach and identification strategy
    - P4: Contributions (numbered, 3-5 items)
    - P5: Paper organization

5.2 Literature review
    - Organize by substantive domain, not paper-by-paper
    - Include Gap Table from Stage 1
    - End each stream: what's known + what's missing

5.3 Marketing implications
    - What should a CMO/marketing manager DO differently?
    - Connect to specific results (robust to X, not just significant)
    - Quantify where possible: "A 10% increase in X leads to Y% change in sales"
    - Distinguish: statistical significance vs. economic significance vs. managerial significance

5.4 Abstract
    - Write LAST
    - Substantive problem → Method → Key finding → Marketing implication
    - Max 200 words for MktSci, JMR, JM, JCR

5.5 Cross-section consistency check
    - Contributions list ↔ body sections (1:1)
    - All variables defined at first use
    - All tables/figures referenced in text

References to load: references/writing-patterns.md, references/reviewer-expectations.md


Writing Patterns

Pattern 1: The Managerial Motivating Example

Open with a concrete marketing decision faced by a real company. "In 2023, P&G allocated $7.1B to advertising across 40 brands, yet had no systematic way to measure cross-brand spillover effects..."

Pattern 2: The Identification Narrative

Devote an entire section to explaining the source of exogenous variation. The best marketing papers make the identification strategy feel like a detective story.

Pattern 3: The Elasticity Table

For demand estimation papers, the own- and cross-price elasticity matrix is the central result. Present a representative subset with standard errors.

Pattern 4: The Counterfactual Triplet

For each counterfactual: Baseline → Counterfactual → Difference. Always report consumer welfare (compensating variation) alongside firm profit.


Venue-Specific Page Budgets

JournalSubmission LengthReferencesAppendixAbstract
Marketing Science35 pagesUnlimitedUnlimited200 words
JMR40 pagesUnlimitedUnlimited200 words
JM40 pagesUnlimitedUnlimited200 words
JCR40 pagesUnlimitedUnlimited150 words
JAMS40 pagesUnlimitedUnlimited200 words
IJRM35 pagesUnlimitedUnlimited200 words
QME40 pagesUnlimitedUnlimited200 words
Marketing Letters20 pagesUnlimitedLimited150 words

Quick Checklist Before Submission

  • Target journal fits the paper's method-contribution profile
  • Identification strategy is clearly articulated and credible
  • First-stage (if IV) or parallel trends (if DID) convincingly shown
  • Results table: coefficients + SE + economic significance + diagnostics
  • Robustness: 8+ checks covering alternative measures, samples, specifications
  • Counterfactuals (if structural): baseline vs. counterfactual with welfare
  • Marketing implications: specific, actionable, connected to results
  • Introduction: motivating example + gap + numbered contributions
  • All citations verified programmatically (no hallucinated references)
  • Page budget within limits; long proofs/materials in appendix

References

  • references/journal-characteristics.md: Detailed per-journal characteristics, editorial philosophy, reviewer expectations
  • references/structural-modeling.md: BLP framework, demand estimation, supply side, counterfactuals, diagnostics
  • references/causal-inference.md: DID, RDD, IV, synthetic control, matching—implementation and reporting
  • references/field-experiments.md: Design, power analysis, pre-registration, CONSORT, analysis
  • references/conjoint-analysis.md: Fractional factorial design, HB estimation, WTP, validation
  • references/reviewer-expectations.md: What MktSci/JMR/JM/JCR reviewers look for, common criticisms, rebuttal strategies
  • references/writing-patterns.md: Reusable writing patterns from published marketing science papers

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.