Find new opportunities across AI search and SEO.

AI search and SEO.

Try free for 7 days
Outperform your search competitors.
Outperform your search competitors.Search now spans Google, ChatGPT, and beyond—and Semrush gives you the SEO and AI visibility tools to stay visible everywhere:
See how AI platforms talk about you
Discover valuable topics to cover
Get AI-powered strategy suggestions
Build keyword plans with personalized insights
Identify fixes to improve technical health
Benchmark yourself against competitors
Outperform your search competitors.
10M marketing professionals have already used Semrush
Outperform your search competitors.
Outperform your search competitors.
Outperform your search competitors.

I edit 115+ articles per year. Here are the 5 ways I edit drafts.

8 min read
September 17, 2026
Contributor: Alex Lindley

Semrush publishes hundreds of blog articles every year. Each one needs editorial review, and most need revisions before they’re ready for publication. As a Senior Editor, I make that process happen. 

What the editing for a given article actually entails is pretty varied. We cover a lot of topics, work with many writers, and balance competing priorities.

So, I simply can’t treat every piece the same way. But I also can’t invent a completely new approach whenever it’s time for me to edit content. (I did that initially, and it didn’t go well.) 

Instead, I use one of five editing approaches. And I’m about to walk you through each of them, how I decide which one to use, and some actual examples of what those editing approaches look like. I’m also sharing my editing best practices for any piece — no matter the approach.

My 5 editing approaches

Here’s how I define my five main editing approaches:

  • Overhaul: An edit that’s almost exclusively comments to get the article directionally on track. There are usually some high-level edits in Google Docs’ Editing mode, such as rearranging the order of sections.
  • Rework: A 30%-50% line-by-line edit in Editing mode with overarching comments for sections that need more substantial work. Alex Lindley, our Managing Editor, is the one who initially recommended a 30% edit for pieces that need a substantial number of revisions. I’m glad he did!
  • Deep Dive: A line-by-line edit in Editing mode, with comments and revision requests throughout the entire article. This is easily what most content I edit falls under.
  • Lighter Touch: An edit where most changes are straightforward. I apply changes either directly or in Editing mode, adding comments and revision requests as needed.
  • No Notes: An edit where I apply all changes directly and add no commentary. These pieces are extremely rare.

I’m not normally one to coin terms — I hadn’t even thought to do so before I started writing this article. But using concrete names seems important for making sure you know what I’m talking about throughout the coming sections.

That said, you can call these editing approaches anything you want. I’m just not that precious about it. 

How I decide which editing approach to use

I choose an editing approach for an article by considering four dimensions in combination with one another: 

  1. Article quality
  2. Article type
  3. Article importance
  4. Writer experience

I usually do this assessment in my head, but it translates pretty easily to a concrete framework that anyone who reviews content can use.

Here’s how it works: For the piece you’re working on, assign a score of 0 to 2 for each of the four dimensions based on the below framework. Add the scores for each column. Then add the column scores to get your total score. 

 

0

1

2

Article quality

Significant issues across factual inaccuracy, tone of voice, structural coherence, clarity, search optimization, product mentions, and style guide adherence. 

A few issues with factual inaccuracy, tone of voice, structural coherence, clarity, search optimization, product mentions, and style guide adherence. 

Few if any issues with factual inaccuracy, tone of voice, structural coherence, clarity, search optimization, product mentions, and style guide adherence. 

Article type

Fully rewritten version of an existing article

Refreshed version of an existing article

New URL

Article importance

Critical level

Medium level

Low or unsure

Writer experience

This is a new writer. Or this writer has yet to submit a piece that’s close to expectations.

The writer has a decent understanding of what their content should look like. This can come from general familiarity with the brand’s content or from feedback they’ve received. 

The writer has a clear understanding of how to create content that meets expectations. This can be from having substantial familiarity with the brand’s content or a longstanding relationship with the brand. 

Column scores

     

Total score

 

Your total score determines which editing approach to take. That breakdown looks like this:

Score of 0: Overhaul

Score of 1-2: Rework

Score of 3-4: Deep Dive

Score of 5-6: Lighter Touch

Scores of 7-8: No Notes

One thing to consider is that different brands can have different definitions of what makes a high-quality article. For example, not every company is as heavily focused as Semrush on content that drives search engine and AI visibility. So, make adjustments to that row to suit your needs. 

And if you just launched your website? Take out the row for article type completely to accommodate having only new URLs. Then, adjust the scale to be from 0 to 6 at your own discretion. 

The benefit of using a framework to choose an editing approach

Using a decision framework to determine an editing approach improves efficiency and reduces mental fatigue.

I can usually determine what approach I’m taking in just a few minutes, which means I automatically know what to do once I start. I don’t have to ruminate over the details of what I’ll say or adjust because that all falls into place.

And as I continue working with the same writers, their articles naturally start to have higher scores. Which means the editing goes faster. And that I don’t need to provide such extensive feedback. 

For example, I usually have to explain that we use the bottom line up front (BLUF) principle — stating the main takeaway at the beginning of every section — the first time I work with a writer. But by about their third article, they know what to do based on a comment that simply says, “Let’s BLUF this paragraph.”

What my 5 editing approaches actually look like

I frequently make copies of articles after I’ve finished reviewing them but before the writer has revised them. Most of the examples in this section come from those copies. 

Overhaul

When I’m using the Overhaul approach, I focus primarily on comments that will get the article directionally back on track.

One of the most common things I do in these sorts of articles is leave a comment over a few paragraphs or sections on how to shape the overall discussion. Like this:

A note by the editor on an article draft commenting on how to shape the overall discussion of the piece.

If the way the sections are ordered doesn’t flow logically, I’ll either copy and paste them to the appropriate place in Editing mode or ask the writer to move the sections. It depends on how complicated the restructuring is. 

Rework

My Rework approach is how I provide enough direction and feedback for a writer to make substantial revisions without the edit taking my entire day.

One of my go-to techniques when doing a Rework is modeling what I want to see. That means I’ll heavily edit a few sections in Editing mode, explain why I made the changes I did, and then ask the writer to apply that elsewhere in the article. That’s exactly what I did here:

A note by the editor on an article draft explaining why certain changes were made and asking the author to apply it elsewhere in the article.

And I sometimes leave a comment at the top letting the writer know that I only edited a certain percentage of the piece. That way, they understand that I expect them to make some heavy revisions before I’m able to do more. 

Deep Dive

Most of my edits are Deep Dives. They also take the most time because I’m going through the entire piece one line at a time.

When I’m working on a Deep Dive, I move beyond addressing the most substantial issues and provide suggestions on how to strengthen arguments, better incorporate actionable workflows, and improve readability. 

For example, I left this comment to explain how using a consistent sentence structure in a bulleted list improves readability:

A note by the editor on an article draft explaining how using a consistent sentence structure in a bulleted list improves readability.

Interestingly, some of the messiest-looking Deep Dives require fewer revisions than ones that look cleaner. Why? Because there are more visible edits when I’ve used Editing mode for something I was able to phrase more succinctly compared to just a comment asking for clarity.

Lighter Touch

Pieces where I’ve used the Lighter Touch approach can either look like I made no changes or just minimal changes — it depends on the writer and how much time I expect to have for accepting Editing mode changes later on.

In the example below, I wanted to show all changes (even minor ones) because the writer was still getting used to working with us.

A note by the editor on an article draft stating how a section is missing a clear BLUF statement at the start.

No Notes

When I use the No Notes approach, it looks like nothing in the draft has changed after editing.

I don’t have a meaningful example of a No Notes edit to show. It would just be the Google Doc version of an article published on our blog. 

But while there aren’t comments in the draft itself, I always leave comments in our project management system. Because there’s usually something the writer should be aware of, like in the example below.

A note by the editor on a Monday.com task giving the author general feedback along with a specific example on an article draft.

Best practices regardless of the editing approach

Fact-check everything

Fact-checking is a nonnegotiable part of editing for everything that comes across my desk.

Because content I first review as an Overhaul or Rework can change substantially by the second draft, I typically only do line-by-line fact checking once I’ve received the revision.

And even if the article is great and I trust the writer, fact-checking still has to happen. It’s easy to misstate the findings from a study or say something related to SEO that’s almost true but not quite true. And those sorts of mistakes can break trust and hurt our reputation. 

Common fact-checking tasks include:

  • Verifying the spellings of individuals’ and companies’ names
  • Ensuring specific numbers are supported and linking to the appropriate primary sources
  • Cross-referencing findings covered in one or our written studies with the raw data
  • Validating workflows and statements about tool features
  • Finding support for assertions about SEO and AI visibility that I’m not certain about

Leave more feedback than you think you should

Leaving a substantial amount of feedback clarifies your expectations and helps writers understand why you’re making changes. 

I’ll admit that I’ve felt uncomfortable at the number and length of comments I’ve left before. But those details make a huge difference in the quality of a writer’s future articles. Writers also appreciate feedback more than you might expect. Case in point:

A response by an author thanking the editor for their comments, suggestions, and edits on a Monday.com task.

Lastly, writers are able to respond when you actually provide your rationale. And they sometimes raise valid points that make you change your mind. That happens to me all the time. 

Point out what the author did well

Providing words of encouragement is great for morale and ensuring that writers know what they should keep doing.

For the article I was editing in the image below, I wanted to make sure the writer knew I liked the approach to going over the benefits of doing a Google Ads analysis:

A note by the editor on an article draft appreciating the author's approach of highlighting multiple benefits.

What about for “No notes” articles? Instead of cluttering the draft with comments, I either send a direct message to the writer or leave a comment in our project management system giving praise where it’s due.

Ask questions

Any time you encounter a section that isn’t clear, contains unsubstantiated claims, or seems out of place, ask the writer about it.

If you come from a background in editorial like I do, it’s easy to feel like you need to fix things yourself ASAP. But it’s not a good use of your time to spend 30 minutes tracking down the source for a stat. And assuming you know what a writer is trying to convey can lead to sending the wrong message.

Asking for clarity can be as simple as this:

A note by the editor on an article draft asking for clarification from the author.

Refine your own editing approaches

My editing approaches can benefit any content team. 

That said, take what you need and leave what you don’t. Maybe you don’t publish a huge amount of content, so three editing approaches is enough.

Ultimately, I hope you walk away from this article with a clearer understanding of how to maintain a thorough review process that doesn’t become a drain on your resources and energy.

Share
Author Photo

Christine Skopec has 10+ years of experience in content writing, editing, and SEO. She has developed content strategies in agency and in-house roles, defined brand style guides, led teams of staff and freelance writers, and developed content production processes. Christine’s blog work drove a 95% increase in keyword rankings for a client within a single year. She has been featured on the Content in the Kitchen Podcast and holds an MS in Journalism from Northwestern University.

Author Photo
Christine Skopec
Christine Skopec has 10+ years of experience in content writing, editing, and SEO. She has led teams of writers, defined brand style guides, and developed content production processes. Christine has an MS in Journalism from Northwestern University.
Share
Add Semrush as a preferred source in Google SearchTells Google to show you more from Semrush in AI, Search, and Discover.

Most popular pages