Author First: AI Assisted | Steve Castledine | Working v5, 8 October 2026 | Review date: 9 October 2026

How to use this review

This review offers Casover's editorial assessment of your manuscript. It identifies strengths and explains where, in our judgement, changes could improve the reading experience. Recommendations are supported by examples and reasons, but editorial judgement allows for different interpretations. You decide which suggestions serve the book you want to write. You may adopt, adapt or decline them.

1. The Manuscript in Brief

Quick overview

Your book teaches authors to use AI for specific writing and publishing jobs while retaining responsibility for the result. Its strongest feature is the connection between advice and decisions a reader can make: preparing a brief, judging an edit, checking a claim and inspecting the finished book.

The completed review finds the manuscript editorially close to production. Its chapter sequence is sound, the appendices contain the promised working material, and the voice is clear without making inflated promises. Extensive rewriting is not justified by the evidence.

The immediate work is small: reunite Chapter Seven's voice-profile explanation and clarify two references to outside guides described as “supplied”. Then confirm that consequential claims and the statements about checking them agree with the actual source record. The absence of that record here does not show that the claims are wrong.

Finally, test usable source access and navigation in the actual publication files. Preserve the existing method and structure while completing those tasks. This follows the reconciled Final Editorial Summary; it is not a new review or a release approval.

2. What Is Working Well

• Judgement is demonstrated. The worked edit distinguishes a useful cautious sentence from an exaggerated promise, and allows retention of good original prose.
• Readers have routes through the work. The Introduction supplies starting points and prerequisites; recaps help readers return to a task.
• The working material is present. Eighteen templates, checklists and longer briefs support the explanations rather than merely being advertised.
• AI contributions remain distinct. Generated drafting, help with author prose and ideas-only use are consistently treated as different choices.
• The final stages stay manageable. The smaller marketing plan and practical ending let the reader choose a next action without adopting every tool or channel.

3. Your Main Revision Priorities

Priority: Complete the voice lesson before the practice discussion.

Why it matters: Chapter Seven returns to voice profiling after three paragraphs on practising decisions unaided. The reader has to recover the earlier subject.

What to do: Move Keep Practising the Decisions, including all three paragraphs, after the completed voice discussion and before the handover section. Preserve its learning qualification.

Detailed report: Pacing Review, PA-01; Dialogue and Voice Review, DV-01. PA and DV identify findings in those reports.