Baffled by KDP Auto Reviewer's Rejection of No-Bleed 8.5 x 11 Interior

[Anklebuster]

Hi All,

After combing through this forum, I see that there are many permutations of issues arising from Operator (me)  / Software / KDP. From bugs in Affinity Publisher to bugs in the KDP automated reviewer to rookie mistakes I and other publishers have made, it's a tough challenge to navigate toward best practices. 

Having uploaded six revisions, cumulatively improving the manuscript after each iteration, I finally reached the point where the automated reviewer deigns to talk to me. Unfortunately, it's speaking gibberish! It's telling me that pages 2-21 are outside of the margins. Try as I might, I can't wrangle these pages.

What's maddening is that they appear to be within the margins, just as the other, non-flagged pages appear! Here are the steps I took:

• Check for SVG frames that were outside of Picture frame (hey experts, doesn't the picture frame de-facto crop image when exporting?)
• Changed the inner margin from .5 to .375 (though I fear that if KDP adds blank pages, the page count will exceed 150)
• Moved picture frames and text frames inward from the .75 blue border (Though what is the point of snapping, then?)

After doing this three times (rev 4, rev 5 and rev 6) this error message and rejection persists. I'm left to conclude one of two things must be happening:

1. The Automated Reviewer is being triggered by something
2. The pages in question are somehow corrupted

These conclusions are speculations and I'm leaning toward KDP being the culprit, after having seen YouTube videos of other people struggle with the automated reviewer. One veteran author had to submit 17 revisions before the automated reviewer accepted his manuscript! I modeled most of my changes after his steps.

I'm using no bleed, exporting with PDF/X-1a:2003 compatibility, .25 outer margins and .375 inner margin.

 

If it's actually possible for Affinity Publisher pages to become corrupted, how would I detect and, more importantly, correct such corruption?

 

[Callum]

Hi AnkleBuster,

I'm sorry to hear you are having issues uploading to KDP, it doesn't really sound like your document has corrupted pages at all please could you provide a screenshot showing the document imported into the KDP with your KDP settings visible so I can investigate this further with you? Have you contacted Amazon KDP regarding this?

Thanks
C

[lacerto]

This is just a guess, but it might be related to auto reviewer possibly trying to check the bleed setting assuming that there is a BleedBox metadata in the PDF., and ensuring that it is the same as the MediaBox When producing with e.g. InDesign, the BleedBox is there, even if the document has no bleeds (0 at all edges), and without specifying any print marks. In Affinity apps BleedBox is added in meta data only if the document has actually a non-zero bleed defined in document setup, then enabled in export settings (can be enabled only if the bleed setting is non-zero), and if printer marks are checked (e.g. "Document Info" adds the bleedbox, but that adds also document info, which is possibly not wanted, and this information is naturally positioned off the page boundaries, which was probably initially wanted to be avoided).

a) 8x11.5 in document without bleeds when produced from InDesign:

b) The same document produced from Affinity Publisher:

It is possible to use 3rd party app, e.g. Adobe Acrobat Pro, to add missing pageboxes in a PDF file.

[thomaso]

1 hour ago, lacerto said:

might be related to auto reviewer possibly trying to check the bleed setting assuming that there is a BleedBox metadata in the PDF.

The bleed setting affects all pages in a PDF, right? – While the OP mentioned the auto-generated error message as being limited to certain pages:

17 hours ago, Anklebuster said:

It's telling me that pages 2-21 are outside of the margins.

• So, what is different on page 1?

• Did you upload pages – or facing 2-page spreads?

• How many pages does the PDF have totally? Page 21 is hardly the last page, I assume either 20 or 24 pages would be required.

[Anklebuster]

3 hours ago, Callum said:

Hi AnkleBuster,

I'm sorry to hear you are having issues uploading to KDP, it doesn't really sound like your document has corrupted pages at all please could you provide a screenshot showing the document imported into the KDP with your KDP settings visible so I can investigate this further with you? Have you contacted Amazon KDP regarding this?

Thanks
C

Hi @Callum,

I have contacted Amazon KDP twice. First was during the past weekend, when their system seemed to be unavailable and I had assumed that my book's cover was no longer showing up. The response was very helpful: I was told that I had inadvertently uploaded the manuscript instead of the cover.

The second time, although I explicitly stated that the margins on the 20 rejected pages looked fine and, despite attaching my PDF to the message and asking someone to look at it, I got the canned response to fix my margins!

Anyway, here is the latest status in my KDP account:

Unfortunately, the launch previewer is unavailable:
 

If you have a Dropbox link I can use, I can share rev 6 file.

Thanks for your reply!

Cheers,

 

Mitch

 

 

[Anklebuster]

1 hour ago, thomaso said:

The bleed setting affects all pages in a PDF, right? – While the OP mentioned the auto-generated error message as being limited to certain pages:

• So, what is different on page 1?

• Did you upload pages – or facing 2-page spreads?

• How many pages does the PDF have totally? Page 21 is hardly the last page, I assume either 20 or 24 pages would be required.

Hi @thomaso,

I exported as pages and there are 148 pages in the document. The title page had no issues and the first ten spreads (pages 2-21) were rejected. The remaining pages were accepted (earlier revisions had margin errors in later pages so, after fixing them and uploading the new PDF, they were accepted.)

Interestingly, I did not apply a Master Page to the title page. All other pages have a Master Page applied.

Cheers,

 

Mitch

 

[Anklebuster]

2 hours ago, lacerto said:

This is just a guess, but it might be related to auto reviewer possibly trying to check the bleed setting assuming that there is a BleedBox metadata in the PDF., and ...

It is possible to use 3rd party app, e.g. Adobe Acrobat Pro, to add missing pageboxes in a PDF file.

Hi @lacerto,

This is good to know, in case it turns out to be the problem. One of the things I had to make a decision on was whether to choose bleed or no-bleed. I was thinking of using bleed as "insurance". If the missing metadata is problematic, I will definitely reformat the document, as I'm not about to shell out money for Adobe :).

Thanks for giving me something to check.

Cheers,

 

Mitch

 

[walt.farrell]

12 minutes ago, Anklebuster said:

The title page had no issues and the first ten spreads (pages 2-21) were rejected. The remaining pages were accepted

So, what is different about pages 20-21 (which fail) and 22-23 (which work)?

 

[thomaso]

10 minutes ago, Anklebuster said:

148 pages in the document. The title page had no issues and the first ten spreads (pages 2-21) were rejected. (...)

Interestingly, I did not apply a Master Page to the title page. All other pages have a Master Page applied.

So what is different on pg. 2–21? Do they have a certain master paster page applied which is not on pg. 22–148? Did you alter page dimensions at any time?

You could upload to the forum for a possible community check just a few pages, e.g. pages 1-4 or 1-22.

[Anklebuster]

11 minutes ago, walt.farrell said:

So, what is different about pages 20-21 (which fail) and 22-23 (which work)?

 

Hi @walt.farrell,

I haven't got a clue. LOL Hence the title.

 

Cheers,

 

Mitch

 

[walt.farrell]

42 minutes ago, Anklebuster said:

I haven't got a clue. LOL Hence the title.

Well, unless you can show us more, or share some sample pages, we probably won't have a clue, either

[Anklebuster]

@walt.farrell and @thomaso What is the best way to export pages? I'm assuming you want to see the .afpub, not the PDF?

[thomaso]

Both files might be useful unless no reason for the issue is known yet.

Since you mentioned file size already, and I assume caused by images: export as PDF with a low resolution (e.g. 10 dpi). You can one or more page ranges (e.g.: "1-4, 20-24")

For .afpub there is no export feature but Save-As instead. Therefore you need to save a separate copy + delete the unwanted pages. In case your images are embedded, go to the Resource Manager, select all and choose "Make Linked" to reduce the .afpub file size (don't upload the image folder). Before uploading use Save As again with a different file name to make sure the file gets rid of obsolete data.

[Anklebuster]

12 minutes ago, thomaso said:

Both files might be useful unless no reason for the issue is known yet.

Since you mentioned file size already, and I assume caused by images: export as PDF with a low resolution (e.g. 10 dpi). You can one or more page ranges (e.g.: "1-4, 20-24")

For .afpub there is no export feature but Save-As instead. Therefore you need to save a separate copy + delete the unwanted pages. In case your images are embedded, go to the Resource Manager, select all and choose "Make Linked" to reduce the .afpub file size (don't upload the image folder). Before uploading use Save As again with a different file name to make sure the file gets rid of obsolete data.

Okay, thanks. Both files are attached. For reference, pages ii-15 were rejected, The title page (the only page with no Master) and pages 16-20 were accepted.

 

Cheers,

 

Mitch

 

Forum Sample 2-27 Lo-res.pdf 80279789_ForumUploadSample.afpub

[thomaso]

4 hours ago, Anklebuster said:

For reference, pages ii-15 were rejected, The title page (the only page with no Master) and pages 16-20 were accepted.

Unfortunately I can not detect any reason for the rejection "pages 2-21 are outside of the margins". (I inspected the PDF only)

At first – because the accepted page 1 (i) has no page number in the layout and thus the largest outer margin – I wondered if the KPD software check might be too perfect or sensitive about its minimum outer margin, but that doesn't seem to be the case, at least I can't see a difference at the outer margin between rejected & accepted pages. I am sorry I can't help.

[Anklebuster]

31 minutes ago, thomaso said:

Unfortunately I can not detect any reason for the rejection "pages 2-21 are outside of the margins". (I inspected the PDF only)

At first – because the accepted page 1 (i) has no page number in the layout and thus the largest outer margin – I wondered if the KPD software check might be too perfect or sensitive about its minimum outer margin, but that doesn't seem to be the case, at least I can't see a difference at the outer margin between rejected & accepted pages. I am sorry I can't help.

No worries. This may actually be a good thing. Negative test clears AP and implicates KDP.

 

Thanks for checking!

 

Cheers,

 

Mitch

 

[Anklebuster]

10 minutes ago, N.P.M. said:

Perhaps the linked eps files in the pictureframe(s) are the culprit.
Try to rasterize&trim the pictureframe if you want to use it only as an background image.

Since I have a dozen eps files in the accepted pages, I'm not sure that's the culprit. Thanks for the suggestion.

 

Cheers,

 

Mitch

 

[Anklebuster]

[Update]

I think I found the culprit. As a test, I edited the Master footer by moving the page number|section name text frame up .02 points. I speculated that the vertical bar might be encroaching on the margin.

Now, only 4 pages are rejected and only one of them is from the original 20. Hopefully, I can figure out what's wrong with these pages.

Finally! The manuscript was approved!

Thanks everyone for your help!

 

Cheers,

 

Mitch

 

Edited June 17, 2022 by Anklebuster
Status update

[thomaso]

3 hours ago, Anklebuster said:

page number|section name text frame up .02 points.

Glad you solved it! – Can you tell whether the culprit indeed has been the page number|section? 
Although there IS a slight overhang, have you determined what distinguishes the accepted pages (such as p. 11 ) from the rejected ones? (title page 1 excepted)

[Anklebuster]

7 hours ago, thomaso said:

Glad you solved it! – Can you tell whether the culprit indeed has been the page number|section? 
Although there IS a slight overhang, have you determined what distinguishes the accepted pages (such as p. 11 ) from the rejected ones? (title page 1 excepted)

The weird thing is that, once I lifted the footer .02 points, 19 of the 20 problem pages resolved but three additional pages were rejected, further into the manuscript.

This makes me wonder if the automated reviewer has a failure threshold beyond which it stops evaluating pages.

Those other 3 pages, along with the final holdout from the original 20, all had various, easily remedied margin violations. Twice, the culprit was the letter J which, at the leftmost edge of a text frame that was snapped to the left margin. The tail, if you will, crossed the margin.

The other two errors involved the bounding boxes of images that had been rotated off the vertical axis.

All in all, a great learning experience.

Cheers,

 

Mitch