Amr Workflow

AMR Theory-Building Workflow (amr-workflow)

Overview

This is the router. It does not replace any specialized skill — it tells you which amr- skill to use at the current stage* of an Academy of Management Review manuscript.

Default assumption: unless the user says otherwise, treat the target as AMR — the Academy of Management's premier theory-development journal. AMR publishes conceptual articles that build new theory; per its scope, submissions "must extend theory in ways that develop testable knowledge-based claims." It contains no datasets, no hypothesis tests, no results section — empirical hypothesis-testing is the lane of its sibling AMJ. The deliverable is a genuinely new theoretical contribution: new constructs, a new process model, or a reconceptualization — developed with rigorous logic, explicit assumptions, propositions, and boundary conditions. If the project has data and tests, it belongs at AMJ / ASQ / SMJ, not AMR.

The AMR contribution bar is set by two AMR-published editorials worth keeping in view at every stage: Whetten's "What Constitutes a Theoretical Contribution?" (1989, DOI 10.5465/amr.1989.4308371) — the What/How/Why/Who-Where-When rubric — and Suddaby's "Editor's Comments: Construct Clarity in Theories of Management and Organization" (2010, DOI 10.5465/amr.2010.0419). Pure-theory exemplars built by argument alone: Oliver, "Strategic Responses to Institutional Processes" (1991, DOI 10.5465/amr.1991.4279002); Dyer & Singh, "The Relational View" (1998, DOI 10.5465/amr.1998.1255632). References follow APA-style (AOM house style).

When to trigger

• The user asks "what should I do next?" on a conceptual paper
• A draft arrives and you must locate its bottleneck (puzzle? logic? contribution?)
• Work is thrashing between theorizing, figure-building, and writing
• An AMR decision letter (developmental review) has arrived and the work shifts to revision

Routing table

Default order

1. amr-topic-selection — lock the theoretical puzzle (the "why doesn't existing theory explain this?")
2. amr-literature-positioning — identify the conversation to challenge and extend
3. amr-theory-development — build constructs, relationships, propositions, boundary conditions
4. amr-methods — theory-construction craft: construct domains, mechanisms, assumptions
5. amr-data-analysis — argument development: logic checks, counterfactuals, alternative explanations
6. amr-contribution-framing — differentiate the new theory from prior work
7. amr-tables-figures — finalize the conceptual figure / typology / propositions table
8. amr-writing-style — AOM house style; argument-driven prose (polish)
9. amr-submission — ScholarOne preflight
10. amr-review-process — understand the developmental review you are about to enter
11. amr-rebuttal — after the R&R

amr-writing-style is a late-stage polish. Do not polish prose before the theory's logic stands up (amr-data-analysis) and the contribution is differentiated (amr-contribution-framing).

Decision shortcuts

• "I have an interesting phenomenon but no theory" → amr-topic-selection
• "I don't know whose theory I'm arguing with" → amr-literature-positioning
• "I have propositions but no logic connecting them" → amr-data-analysis
• "My P1–P5 read like assertions" → amr-theory-development then amr-data-analysis
• "A reviewer will ask 'what's new here?'" → amr-contribution-framing
• "My model figure is boxes and arrows with no mechanism" → amr-tables-figures
• "It reads like a review essay" → amr-writing-style
• "I'm about to hit submit" → amr-submission
• "I got a Reject & Resubmit / Major Revision" → amr-review-process then amr-rebuttal

Differences vs. AMJ / ASQ / SMJ skill stacks

If the manuscript has data, measures, and statistical tests, an empirical-management stack (AMJ / ASQ / SMJ) fits better. The core split:

• AMR: builds theory; the contribution IS the theory; no data; propositions, not hypotheses tested.
• AMJ / ASQ / SMJ: test theory; hypotheses, samples, estimation, results.

A common failure is sending an AMR draft that is really an under-powered empirical paper, or an AMJ draft whose "theory" is a literature summary. Pick the right stack early.

Stage ledger (paste at the top of your working file)

Keep one honest line per stage. AMR reviewers judge the theory, so the ledger tracks argument-state, not word count. Mark a stage DONE only when its gate is truly cleared.

AMR THEORY-BUILDING LEDGER  —  manuscript: ____________________
[ ] puzzle        why does existing theory FAIL to explain X?  ......... route: amr-topic-selection
[ ] conversation  whose theory am I extending / challenging? ........... route: amr-literature-positioning
[ ] constructs    each construct: name + domain + What/How/Why (Whetten) route: amr-theory-development
[ ] propositions  P1..Pn stated; each has an explicit causal logic ..... route: amr-data-analysis
[ ] boundaries    Who/Where/When conditions named for the theory ....... route: amr-theory-development
[ ] contribution  one sentence: "New vs. prior theory is ____" ......... route: amr-contribution-framing
[ ] figure        box-and-arrow shows a MECHANISM, not a taxonomy ...... route: amr-tables-figures
[ ] prose         reads as argument, not literature review ............. route: amr-writing-style
[ ] preflight     200-word abstract, APA refs, ScholarOne fields ....... route: amr-submission
GATE: no data, no hypothesis tests, no results section — if present, this is an AMJ paper.

Anti-patterns

• Do not skip amr-literature-positioning and jump to building — reviewers first ask whose conversation you are in.
• Do not let amr-tables-figures pretty up a model before the propositions and mechanisms exist.
• Do not let amr-rebuttal draft a response before the theory itself has actually been revised.
• Do not treat AMR as a venue for data; route empirical work elsewhere.

Submission specifics change. The source map now verifies the current editor, ScholarOne portal, no-fee policy, AOM style mechanics, 200-word abstract limit, and AI-disclosure policy; keep checking the official AMR / AOM author pages for manuscript length, portal prompts, and any newly published acceptance-rate or reviewer-count figures.