Treat a duplicate-image report as a list of candidates for review. Establish what the check compares, locate each use in the manuscript, and determine whether the repetition is intentional. Remove or replace an image only after checking its role, caption, and references, because repeated file content and redundant editorial content are different questions.
Establish what the report calls a duplicate
Read the check's description and identify whether it compares identical files, rendered image content, or visual similarity. These methods can produce different candidate sets. Keep the finding tied to the method actually used instead of treating every reported pair as proof of an editorial mistake.
Record the files and version of the manuscript inspected. A report based on an asset folder may include unused alternatives that never entered the book. Conversely, a check of the exported pages may not identify how several placements connect to the same editable source asset.
Locate the uses in the accepted manuscript
Find every relevant placement and inspect it with its surrounding text. Record the chapter, figure label, caption, and purpose of each use. This distinguishes an accidentally repeated illustration from a deliberate reference to an earlier example or a recurring decorative element.
Check whether the same image is being presented as different evidence. Reusing a generic icon across sections may be intentional, while using one photograph to imply two distinct observations can create a misleading explanation. The editorial meaning depends on how the image is used, not merely on whether its pixels repeat.
Separate repeated assets from repeated teaching
Identify what each placement contributes. An image may be useful again when readers need it beside a later task, especially if returning to the earlier page would interrupt the activity. That decision should be deliberate and proportionate to the reading need.
If two placements repeat the same explanation without adding useful context, consider whether one can be removed or replaced with a clear reference. Do not make that choice solely to reduce the report's count. A lower duplicate count is not an independent measure of a better book.
Review visually similar alternatives carefully
Compare candidate images that look alike but may differ in a meaningful detail. A before-and-after pair, a corrected diagram, or a change in object position can be central to the instruction. A similarity finding should prompt inspection rather than automatic consolidation.
For example, two figures may show the same object with one component moved. If the report treats them as near duplicates, preserve the distinction when it teaches the intended action. Verify that the captions and sequence accurately describe the difference instead of relying on the author's familiarity with the images.
Trace any removal through dependent material
Before deleting a placement, identify references, captions, navigation, and layout relationships that depend on it. A sentence directing readers to the figure below can become incorrect even when the surrounding paragraph remains unchanged. Update the connected material as part of the same revision.
Keep the source asset if it is still used elsewhere or needed for the project's accepted history. Removing one placement does not necessarily mean deleting the underlying file. Distinguish manuscript cleanup from asset retirement so a local edit does not break another chapter's image link.
Verify the revised export and record exceptions
Inspect the affected pages after changes and rerun the relevant check where appropriate. Confirm that the intended correction reached the actual export and that any remaining repeated uses are understood. A report can remain nonzero while the manuscript's intentional reuse is sound.
Record the reason for retaining important exceptions, such as a recurring symbol or an instructional comparison. Keep the review connected to the current manuscript and asset versions. The final decision should establish accurate, useful image placement rather than claim that a technical similarity check can decide editorial redundancy on its own.
Frequently asked questions
Clear answers for this publishing decision.
Must every repeated image be removed?
No. Review the role of each placement. Intentional repetition can support navigation, comparison, or a task performed in a later section.
Can similar images contain an important difference?
Yes. Inspect the actual details and captions, especially for instructional sequences and before-and-after comparisons. Similarity alone does not establish redundancy.
Does deleting a placement mean I should delete the file?
Not automatically. Check other uses and the project record before retiring the asset. Placement and source-file management are separate decisions.
What proves a duplicate finding was handled?
Record the locations reviewed, the editorial decision, and any resulting changes in the accepted export. The goal is appropriate use, not necessarily a report with no repeated assets.
What must I check, disclose, and verify myself when AI helped produce the book?
Review facts, rights, and source material yourself, record which parts were AI-assisted or AI-generated so you can answer Amazon’s disclosure question accurately, and treat every automated check as a starting point rather than a verdict.
See what the compliance checks cover