When the editors ask for changes
Updated August 24, 2026
A change request is not a rejection. It is an editor saying they will publish this once two paragraphs are fixed, which is a much better outcome than silence. This article covers what arrives, where to read it, and how to send the article back.
How you find out
A notification appears on the bell, filed under Action needed and labelled "Your article needs changes". The notification itself reads:
Changes were requested on {your title} Read their notes, revise the draft, and send it back.
Clicking it opens the article. The status on My Guest Posts changes to Changes requested at the same time.
Note
Rankbank does not email you about this or anything else. If you are waiting on a decision, the bell is the only place it will appear. See How Rankbank tells you things.
Where the feedback lives
On the article's page, a panel called Feedback from the editors holds every note, described as "If they want changes before publishing, their notes show up here." Before anyone has said anything it reads "No feedback yet" with "If the editors ask for changes, you'll see exactly what to fix here."
Above the panel you may also see an alert headed Review outcome: Changes requested, carrying the reviewer's overall note, or "No review note was provided." if they left it blank, plus the date it was reviewed.
There are two kinds of note and they read differently.
General comments are the reviewer's overall direction. Fix the intro, the piece is too promotional, that sort of thing.
Highlighted comments are tied to a specific piece of your article. The reviewer selected a sentence in your text, or clicked one of your images, and attached a note to it. The comment quotes the exact words they selected, so there is no ambiguity about which paragraph they mean.
Every comment shows who wrote it and a line reading Version {n} · {Type} · {Status}. The version number is the submitted version the comment belongs to, which matters once you have been round the loop more than once. The type is a category the reviewer picked: Overall, Grammar, Seo, Links, Formatting, Images, Brand or Other. The status tracks the conversation.

Tip
The comments you see are all of them. Reviewers can also leave internal notes for their own team, and those are built so they never reach you and never notify you. Nothing is being hidden from you in a way that affects your article.
Replying
Reply opens a modal headed Reply to review comment, described as "Author-visible conversation for {your title}." Type your reply and press Reply. An empty one is refused with "Reply body is required."
Reply when the request is unclear, when you disagree, or when the fix is not what they think it is. A reply notifies the reviewing side, so it is a real conversation and not a note in a bottle.
Do not reply to every comment just to acknowledge it. Use Mark Addressed for that.
Mark Addressed
Mark Addressed on a comment says: I have made this change. It moves the comment out of the open pile and tells the reviewer where you got to. The button greys out once a comment is addressed, and you can only do it to comments on your own article.
You cannot resolve or reopen a comment. Those are the reviewer's calls, deliberately, so that "done" is agreed by the person who asked rather than declared by the person who was asked.
The most useful pattern is boring: work through the comments in order, make the fix, press Mark Addressed, and reply only where you actually have something to say.
Resubmitting: two paths
This is the part that trips people up, because it depends on where your article was written.
If your article came from a Google Doc
While the article is in Changes requested an amber card appears under the feedback:
Make the changes in the same Google Doc, then resubmit — we'll pick up the new version and keep the old one on record.
Do exactly that. Open the same document you originally imported, make the changes there, and come back and press Resubmit my updated doc. Rankbank fetches the document again, files what it finds as a new version of the article, and tells the reviewing side that revisions were sent.

Two things to keep in mind. Do not make a copy of the document and share a new link, because Rankbank is holding a reference to the original. And leave the sharing setting as "Anyone with the link can view", because the resubmission fetch is anonymous just like the first import was. If it fails, the reasons are the same ones listed in Importing from Google Docs.
If you wrote your article in the Rankbank editor
Do not use the resubmit card. It is written for Google Doc articles, and it looks for a document your article does not have.
Your path is the ordinary one, reopened:
- Press Edit article, or open the editor from My Guest Posts. It is editable again, because Changes requested unlocks it.
- Make the changes.
- Press Next: review & send →.
- On Review & send, check the links table again. If you added or removed links, this is where you find out.
- Press Send for review, then Send it in the confirmation.
That creates the new version and notifies the editors, exactly as the first submission did.
Warning
The amber resubmit card appears for every article in Changes requested, including ones written in the editor, and its wording only fits Google Doc articles. If your article was written in Rankbank, ignore that card and use Next: review & send → instead. You can tell which kind you have from the Article details panel in the editor: From: Google Docs or Written in: Rankbank editor.
What each round leaves behind
Every resubmission creates a new version. The old ones are not replaced, overwritten, or deleted. They stay on record as immutable snapshots, along with every decision that was made about them.
The article itself keeps one identity throughout. Version 3 of your article is the same article as version 1, at the same address, with the same comment history and the same accepted guideline pinned to it. Nobody has to hunt for "the latest one".
Reviewers only ever decide on the latest submitted version. If a reviewer left a stale tab open on an older version and tries to act on it, Rankbank refuses rather than letting a decision land on the wrong package.
On the article page, the Where it's been panel is the summary: "Versions, the live link once it's published, and our link checks." Its Current source version tile tells you which version is live in the review at any moment.
If the answer was no
Rejected is different from Changes requested. The notification is labelled "Your article was declined" and reads "They passed on this one. Their notes explain why." There is no resubmit card, because the decision is final for that review. Read the note, and if you still want to write for that site, ask them directly before starting again.
Still stuck?
Open Rankbank and the page this article describes — most screens explain themselves as you go.