What is the 'perfectly optimised' article for SEO?
It can be daunting to decide how to give your news article the best chance of performing in Google. Here I look at the seven most valuable optimisation elements.
Journalists are busy enough without having to worry about how to best optimise their stories for maximum Google traffic. Deciding what and how to optimise every aspect of an article creates extra workload that isn’t necessarily conducive to journalism.
In this newsletter I want to look at the key elements of an article’s search optimisation that make a genuine difference. Moreover, I will look at these elements in order of importance, highlighting what has the biggest impact on an article’s performance in Google’s news elements.
Keep in mind that the ranking elements I outline below are specific for news stories in Google’s news-specific ecosystem: Top Stories carousels, the Discover feed, Google News, and the News tab.
I will only look at the editorial aspects of an article. Site-wide authority factors are arguably more important for Google than an individual article’s optimisation, and I’ll dig into that in an upcoming newsletter Harry Clarkson-Bennett digs into those in his excellent piece here. I’ll also ignore technical SEO considerations for articles, as I’ve previously covered many of them.
1. Article Headline
The first and single most important ranking factor for an article in Google’s news ecosystem is its headline. Especially in Top Stories and the News tab, the topics contained in an article’s headline are used by Google to determine the article’s focus and what search terms it should rank for.
Simply put, the keywords you put in the headline are the keywords the article is most likely to rank for. If you write an article about the Superbowl, it really helps to have ‘Superbowl’ in your headline.
While article headlines can be quite long - up to 110 characters, including spaces - it’s beneficial to put your article’s keywords early in the headline. More weight is given to keywords that are part of the headline’s first 8 words. Craft your headline accordingly.
Sometimes you can use shortened phrases for a topic, for example ‘Trump’ rather than ‘Donald Trump’. Google will nearly always correctly understand the topic, as it also takes clues from other words in the headline and the rest of the article.
When you’re not entirely sure if Google will understand a keyword, just search it on Google and see what shows up. If the search result is clear and unambiguous, Google obviously has a good understanding of what that keyword means.
However, if the search result is a mixture of different interpretations of the keyword, you may need to help Google along a bit.
My favourite example of this is the keyword ‘Hamilton’. Depending on when and where you search for ‘Hamilton’ you’ll either get a search result focused on the Hamilton musical or the F1 driver Lewis Hamilton.
In such cases, it might be better to expand the headline to include the full topic (‘Hamilton musical’ or ‘Lewis Hamilton’, depending on what you’re writing about)
An article can actually have up to five different SEO-relevant headlines. If you have to pick just one to optimise, the main headline that’s visible for readers is most important.
2. Freshness
After the headline, the ranking factor Google cares most about is how fresh the article is. Google has a very strong preference for newer stories, especially in Top Stories boxes and carousels.
On many levels this makes perfect sense. A newer article on a developing news story is more likely to be up to date and contain the latest information. However, this also means that publishers that are first to break a story often don’t get the visibility they perhaps deserve in Google’s results.
I’ve written before about how breaking news isn’t always rewarded by Google. If you want to ‘own’ a story in Google, especially in Top Stories, it pays to do multiple stories on a news event or take a bit more time to prep and expand your story before publishing.
There are cheats for manipulating freshness signals (for example by changing an article’s URL), but I generally don’t recommend these unless absolutely necessary. Sooner or later, overuse of cheats will backfire when Google decides to alter its algorithms and punish websites that have been taking too many liberties.
3. Featured Image
In almost every context where a news story is shown in a Google list - be that a Top Stories box, the News Tab, the Google News vertical, or the Discover feed - it is accompanied by an image.
An article’s featured image isn’t a direct ranking factor. There are plenty of theories about Google’s image recognition capabilities, but generally we can assume Google doesn’t take the time to do proper image analysis for news articles. It’s too hard and time-consuming.
Article images aren’t a direct ranking factor. Yet they have a massive impact on a hugely important direct ranking factor: Click-through rate (CTR).
Google constantly measures where its users click, and various other click-related signals (click length, click quality, etc.). In news-specific results, especially Top Stories, this click data is used to tweak the articles shown with regular intervals.
As CTR is determined by a combination of headline, news brand, time stamp, and image, it’s worth putting effort into selecting the right image to accompany your story. As a rule, people will click on people, so an image of a person or group of people tends to be a good starting point.
Read more about optimising images for Google News in a previous newsletter.
4. Prominence
In Google’s own documentation on ranking in Google News, one of the factors listed is ‘prominence’. What does this mean?
Prominence is a factor derived from how prominently the article is displayed on the publisher’s website. If the article is top of the homepage and on important section pages, it’ll have a high prominence. This will help its ranking in Google’s news ecosystem.
If, on the other hand, the article is featured very low down where few users will go, is not featured on the homepage and/or it’ll take several clicks from the homepage to reach the piece, then it’ll have low prominence signals. This is likely to have a negative impact on the article’s ability to perform in Google.
The way Google measures prominence is, I suspect, in large part a measure of link value. A website’s homepage is almost always its most valuable page in terms of link value, and a highly visible placement on the homepage means the article will receive a lot of link value from that homepage.
Whereas if the article is several layers of clicks away from the homepage, and/or has a placement where users are not as likely to scroll to and click, then it’ll receive less link value.
This theory is strengthened by data discovered in the Google content API warehouse leak, which featured many link-based attributes that fit such a model.
To give an article the best chance of maximising its Google potential, prominent placement on the homepage and section page(s) is key.
5. Intro/Summary/Subhead/Dek
An often overlooked aspect of an article’s optimisation is the intro (or summary, or subhead, or dek, or whatever you choose to call it). The intro is the sentence right after the headline and before the article’s main text. This intro is usually marked up differently from regular copy, almost like a smaller second headline.
Intros help set the context for the rest of the article and often serve as a brief summary of the article’s main content.
In terms of SEO, the intro is an opportunity to add additional keywords and topics that may not have fitted into the article’s main headline. By including more keywords in the intro, you can broaden the scope of the article’s potential rankings and its visibility in the Discover feed.
Rather than using meaningless <span> or <div> HTML tags, marking up the intro with semantic HTML gives it extra emphasis. Use a <h2> heading tag or a <strong> tag helps Google to understand it should pay a bit more attention to the intro text.
6. Article Copy & Structure
The actual main body text of the article isn’t as important as many might assume. Google is notoriously bad at understanding content, so I generally advise publishers not to overthink the article text for SEO. Let the journalism speak for itself.
One thing that is worth keeping in mind is that Google emulates understanding of text through vector embeddings.
In practice this means that it helps to use words and phrases that provide the right context, like ‘Jaguar’ and ‘miles per hour’, so Google understands that the article’s focus is luxury cars and not feline jungle dwellers.
Using subheaders like <h2> and <h3> to segment your text can also help, especially if you phrase the subheadings as questions. Google will realise your text probably answers that question, so your article may be shown as an answer to that question when someone asks it in Google.
7. Internal Linking
While links contained in an article aren’t direct ranking factors, consistent internal linking to topic pages does help to improve the site’s overall topic authority signals.
This is one area where generative AI can actually help. LLMs are language models, and internal linking to topic pages is a language-based tactic. There are quite a few AI systems that automate internal linking using LLMs.
If you don’t want to manually add links to your articles you can rely on these AI systems to help you along the way with a very high degree of accuracy.
Much More To It
These seven basic aspects of optimising an article may already seem like a lot of work, but it really just scratches the surface. There’s a lot more you can do to optimise your visibility in Google’s news-specific surfaces. However, this can quickly degrade into great effort for minimal gains.
Consistently doing the basics right will get you most of the way there. The key word there is ‘consistently’. Success in SEO is not difficult, but it is hard work.
Upcoming Events
I’ve got a few events coming up in the next while, I might see some of you there!
19/20 February: WAN-IFRA Bangalore AI Forum - I’m delivering a talk (via Zoom) about the impact of AI on search and SEO.
13 March: Friends of Search, Amsterdam, NL - I will be moderating the main SEO stage.
27 March: SearchIRL, Dublin, IE - An informal meetup of SEOs in Ireland, and I’ll be there to talk tech SEO.
10/11 April: BrightonSEO, Brighton, UK - Just attending and networking.
5/6 May: WAN-IFRA World News Media Congress, Krakow, PL - I’ll be talking about SEO for publishers in 2025 and beyond.
15 May: Hive MCR, Manchester, UK - I’m part of a panel discussion on technical SEO.
Miscellanea
Here are some of the latest Google documentations, interesting articles from the last while, and cool stuff from the world of SEO.
Official Google Docs:
Latest in SEO:
Predictions for news SEO in 2025, part 1 - WTF is SEO?
Predictions for news SEO in 2025, part 2 - WTF is SEO?
2025 News SEO Survey: Industry Insights into Emerging Trends - NewzDash
Mastering EEAT for publishers - Harry Clarkson-Bennett
Google quietly updated the News and Discover manual action policies - SEL
Why is Google losing market share in the EU? - Growth Memo
The rise of the AI crawler - Vercel
Interesting Articles:
2025 Will Break Media’s Addiction to Ads - Matt Karolian
Le Monde CEO: Digital subscriber revenue will pay for entire newsroom within two years - Press Gazette
“Just give me the f***ing links!”—Cursing disables Google’s AI overviews - Ars Technica
Google owes UK news industry £2.2bn from 2023 alone, claims new research - Press Gazette
Knowing less about AI makes people more open to having it in their lives – new research - The Conversation
A tool for evaluating the audience relevance of a news article - Baekdal
At long last, a new edition of SEO for Google News! Work and travel have kept me very busy. No promises for the next edition, but I have a few topic ideas in my head so hopefully it won’t be too long.
Thanks as always for reading and subscribing. Please send this on to anyone you think may find it useful.
Strongly agree! BTW, I think the summary or intro sentence is worth applying on every informative piece of content, including blogposts.
Side note: one can be creative and experiment with the concept to the point you can turn it into whatever you see fits your page template the best, even if it's not a sentence anymore.
I agree with almost everything here as a starting point! I would definitely add in "subheads within articles where appropriate" because those are great for users and google to orient themselves.
I was under the impression that vector embeddings/Word2Vec were no longer in use since the incorporation of transformers into the ML model, but it doesn't really change the fundamental instructions for journalists.