Step 1 — Submission Received
Author submits manuscript, abstract, category selection, and required AI-disclosure level (Level 1, 2, or 3) through the submission form. Incomplete submissions (missing disclosure statement, no category selected) are returned to the author before triage begins.
Step 2 — AI-Assisted Triage
The submission goes through an AI-assisted first pass that produces:
- A category fit assessment (confirms or questions the author's selected category)
- A completeness check (required sections present, disclosure statement included)
- A structured brief: stated methodology, core claim, data/sample description, and any surface-level inconsistencies or unsupported claims worth the reviewer's attention
This step does not produce a recommendation, score, or lean toward any outcome. Its only output is a brief and a completeness flag.
Step 3 — Routing to Reviewer
The submission and triage brief route to the reviewer whose credentialed category matches the submission (see Current Reviewers on the Call for Papers page). Submissions spanning two categories — like program-evaluation work touching both Public Policy and Workplace Learning & OD — may be read by both reviewers, or by the reviewer with the closer primary fit at the editor's discretion.
Step 4 — Independent Review
Reviewer reads the full submission. For self-reviewed categories, the reviewer records a first-pass read (strongest point, weakest point, initial leaning) before opening the AI triage brief, per the Quality Assurance Statement's sequencing requirement.
The reviewer then completes the decision form:
- Two specific citations from the submission — strongest point, weakest point
- An independent restatement of the core claim and methodology, in the reviewer's own words
- A decision: accept, revise and resubmit, or reject
- A rationale referencing something specific to this piece
Step 5 — Decision Communicated
- Accept → proceeds to Step 6, Publication.
- Revise and resubmit — author receives the specific feedback tied to the citations from Step 4, with a reasonable resubmission window. Resubmission re-enters at Step 4 with the same reviewer.
- Reject — author receives a direct, respectful explanation. Undeterred Press does not use form-letter rejections; the citation-based rationale from the decision form is the basis for what the author receives, adapted for tone. This is the branch point for an appeal — see Step 5a below.
Step 5a — Appeal (branch, reject decisions only)
An author who receives a reject decision may appeal once. The appeal must identify a specific claimed error in the original review — a factual mistake, a misreading of the methodology, evidence the reviewer's rationale overlooked — not simply disagreement with the outcome.
Where the category has a second qualified reviewer, the appeal goes to them for an independent second read, blind to the first reviewer's decision. Where it doesn't (true for most categories today, given the current roster), the Founding Editor conducts the second read personally, after a cooling period of at least a few days, and documents the second read using the same decision-form requirements as any other review.
The appeal decision is final; there is no second appeal on the same submission. It resolves back into Step 5: upheld → decision stands, process ends; reversed → proceeds to Step 6, Publication; converted to revise-and-resubmit → re-enters at Step 4.
Step 6 — Publication
Accepted pieces are published with: full author byline, AI-disclosure tier stated plainly, reviewer category noted, and publication date. No piece is published without a completed decision form on file.
Timing Target
18 days average from submission to decision, tracked per the Quality Assurance Statement's timing-signal process. This is a target, not a guarantee — genuine review quality is not sacrificed to hit it.
Conflict of Interest
A reviewer recuses from any submission where they have a financial interest in the outcome, a close personal relationship with the author, or a business or investment relationship with the author or their organization. This is relevant given the Founding Editor's angel-investing history — a submission touching a company he holds a stake in is an automatic recusal, not a judgment call made in the moment.
Given the current two-person roster, one specific case needs its own rule: a submission from a close family member or personal connection of either reviewer. In that situation, the connected reviewer recuses. If the submission's category has no other qualified reviewer available, the Founding Editor discloses the relationship plainly in the piece's published record rather than proceeding as though independent review occurred. Undisclosed conflicts are the failure mode this exists to prevent — a disclosed one, handled openly, is not a scandal.
Recusal is noted on the decision form itself: which reviewer recused, why, and who reviewed instead.
What This Workflow Does Not Yet Cover
Co-reviewer disagreement resolution — what happens when two reviewers reach different conclusions on the same submission — is intentionally left undefined for now. With review volume this low, it's better to see how early submissions actually go and design this from a real case than to pre-build a process for a scenario that hasn't happened yet.
Companion document: see the Editorial Review Quality Assurance Statement for how adherence to this workflow is verified, audited, and corrected if a lapse occurs.