How to Optimise for AI Overviews

With AI Overviews (formerly Search Generative Experience or AI Overview) released for testing over the last six months, the question now arises for SEOs and business owners around the world..how do we rank for AI Overviews?

To rank your webpage in AI Overview results, first focus on achieving a first-page position in regular Google searches, target the keyword search with content, and use brief and clear sentences to summarise the whole topic of your webpage.

In this article, we’ll go through in more detail the current Search Generative Experience and show different techniques to optimise for this new search feature, so you’re ready for when it launches to the world!

Table Of Contents

1. What is Search Generative Experience?
2. Understanding AI Overviews & Its Impact
3. Types of AI Overview Rankings
4. Optimisation Techniques for AI Overviews
5. eCommerce SEO for AI Overviews
6. Critical Observations & Recommendations for AI Overviews
7. Will AI Overviews be Released?
8. Final Word

What Are AI Overviews?

The Search Generative Experience (AI Overview) was an AI-powered search result feature by Google, currently in testing and accessible through Google’s Search Labs in certain countries – initially in the US only, then expanded out to 120 other countries a few months back.

It’s designed to enhance search experiences by generating informative panels at the top of search results.

For the SEO community and wider business community at large it was met with excitement and confusion, this is a big change up for the future of SEO.

This is Google’s counter to OpenAI to keep up in the Artificial intelligence arms race. It was keeping Google relevant to the modern-day searcher.

Check out our video on preparing for AI Overviews (previously Search Generative Experience)

Understanding AI Overviews and Their Impact

AI Overviews is a significant change to the search experience. the 10 Blue Links people have been used to for the past 20 years are now being pushed down (literally), and a new generative result is at the top of the search results. It’s a part of the never-ending changes to be focused on on-the-SERP results.

As we know, whatever is at the top will get the most impressions, clicks, and, inevitably, conversions.

Experts predict up to 30% of traffic loss like any SERP feature change, which can syphon away traffic and impressions because of its presence in the search engine results page. (SERPs)

But if someone is losing traffic, surely someone is gaining on the other side?

Rankings on the typical Search Engine Results Page and the generated experience will be a massive opportunity, and Google will make it challenging for us to figure out.

Getting a grip on how to use your brands’ products and services in the search generative experience needs to be high on your list for the future of your organic traffic.

Types of AI Overview Rankings

From tests in demos, here’s the breakdown of the types of positioning you can get in AI Overviews:

1. Carousel Links

2. Article Links

Get your website listed under the dropdown of each statement in the AI Overview answer.

3. Product/Business Recommendations

Aim for your products or services to be recommended directly by AI Overviews.

4. Branded Searches and Alternatives

Leverage the opportunity where an AI Overview suggests your business as a follow-up question or alternative option.

You need a strong SEO foundation and brand.

First things first. You have to be already ranking well in the Google search results.

An indicator of this is if your website is ranking well for its target keywords and positioning, ideally on the first and second pages (ranking 1st – 20th, 30th at max) of the traditional search results.

Suppose a website about health supplements ranks well on Google for “best vitamin supplements for energy.”

This high ranking in the traditional search results provides a strong base for the site to be considered for AI Overview ranking.

Your brand authority must also be strong; you must be known to Google.

To check this, see if you have a knowledge panel for your brand:

Or if Google is pulling a correct and cited description of your brand:

Let’s say you don’t have this.

Your socials and backlink profile are the signals to increase your brand in the search results.

Get all your social media platforms up and consistently post on LinkedIn, X, Instagram, TikTok, and YouTube.

Example – AI Overview does use LinkedIn articles for citations.

When building up your link profile you have to have quality, relevant, non-spammy backlinks.

Consider investing in digital PR for your brand to give you a base of backlinks, and help to grow your brand visibility and showcase your expertise.

Get your brand covered by other comparisons and review websites as AI Overview will generate off-top pages only. It will present information at something different than the 100th result.

This will all depend on the competitiveness of your niche. If you’re competing against the biggest household brands in the world, this will be a nearly impossible feat without an infinite marketing budget.

For niches that have no brand dominance, you have a better shot at this.

AI Overviews doesn’t disallow you from succeeding in the search results; it’ll just make it more challenging.

By already having credibility and visibility in Google’s eyes, the website stands a better chance of being recognised and featured in AI Overview’s AI-generated panels.

If you need rankings or are starting fresh, don’t aim for AI Overviews.

Aspire to achieve the proper website structure and topical content to get Google to understand your relevance on a topic.

Optimisation Techniques for AI Overviews

Here’s a breakdown of the different experimental techniques you can use to rank in AI Overviews:

Targeted Content for AI Overview

Create content that explicitly addresses the search query.

I know I’m preaching to the choir; it just needs to be taken more seriously than ever.

For example, when targeting a query like “how to start a keto diet,” make an article that precisely addresses this question with a step-by-step guide rather than some general information about keto diets. This is more likely to be favoured by AI Overviews.

This means the content should focus exclusively on the initiation process of a keto diet, avoiding unrelated details.

How do I target terms in AI Overviews?

SERP analysis.

Try to understand the “Question” the user is asking on the search result and the “need” they are trying to fulfill.

This may seem obvious.

People can continuously pursue what an intent is – the only way to fully understand what a user may want in real-time is by reviewing the AI Overviews and the typical search results.

If the search query is “best yoga poses for lower back pain,” the targeted content would be a blog post titled precisely the same or very similar, focusing solely on yoga poses that alleviate lower back pain rather than yoga in general, or other types of back pain.

Or take a keyword like: “SEO audit.” Is the person looking to understand what an SEO audit is, or are they looking to purchase an SEO audit from a trusted provider?

AI Overview Result:

Normal Search Result:

It’s a mix of both commercial investigation and transactional intent.

So, in the process, you can do this by analysing:

1. The keyword
2. The question
3. The need
4. Then validate this by checking the top 10 search results for what is ranking already.
5. Iterate the question and need based on the SERP.

I’ve created a chatbot that can help guide you through thinking deeper about a keyword and less on its face value called Query Insight.

Once you’ve done this in your keyword research stage, develop blogs that match the search topic directly.

Here’s an example from one of our clients.

This section is pulled from the second heading after the introduction, answers the question, the need, and matches what the AI Overview result is after:

Analyse The AI Overviews Result

Here’s the secret sauce to optimise for AI Overviews:

Research the AI Overviews result and the information it’s presenting in the content, section by section.

The AI snapshots are giving us hints as to what Google thinks users want to read and the pages that it believes have the most relevant content.

They’ve actually made it quite easy to confirm what content is required on your page to get into the carousel and article links!

In AI Overviews, when you select a source, the text Google has analysed is marked in purple and usually has the element of being short, simple, and direct.

This is the key to AI Overviews. Give them what they’re generating.

As an example, for a keyword search like “tips for first-time dog owners”, a concise list of practical, straightforward tips specifically targeting first-time owners would be more effective than an exhaustive guide on dog ownership.

This is what is presented when checking the AI Overview result:

This could include direct, easy-to-understand advice like choosing the right breed, basic training techniques, and essential supplies.

Re-write and optimise your content to match what AI Overview is presenting:

Check the corresponding pages and note which content is on the page and how it’s related to the AI Overview result.

If AI Overviews features a paragraph from your page on “the benefits of sleep,” and another paragraph from a different website with the heading “Tips for Getting Better Sleep”, (that might be more concise, or list key methods, etc…), see what could be replicated from that website in your own content to align with AI Overview’s format.

Clicking through →

If there’s any additional information, (we’ll talk about ‘information gain’ later), that you don’t have on your page, then add in the new information to bolster the relevance of what AI Overview wants out of that citation.

Be Short, Specific, and Direct

It’ll likely be short and direct with a range of points.

From what we know of the modern day search result, always keep the content brief and to the point. Directly address the search query while keeping it easy to read by any user.

Google employs a method called ‘semantic similarity’ to comprehend your content. It’s essential to present your information clearly and concisely to increase the likelihood of Google referencing your site.

Focus on enhancing the ease of reading by:

• Providing straightforward answers
• Using clear and distinct headings
• Offering detailed FAQs

Optimise Content with Concise Summaries

Add short, clear paragraphs at the beginning of the content to summarise the topic.

A trend in AI Overview results is short summaries, and as we know, Google needs to be able to scan content. Having a synopsis at the top of the content while directly and concisely answering the keyword’s question will make your content easier to read.

You see this in many of the top blogs as “Key Takeaways” or “A Quick Summary.”

In an article about “energy-efficient home appliances,” the first paragraph could succinctly summarise the key points about why these appliances are energy-efficient, their benefits, and a brief overview of the top picks.

This aligns with AI Overview’s preference for content that quickly presents the information sought in the search query.

Elevate Highlighted Paragraphs

This technique involves identifying the content or paragraphs from your website that AI Overview uses in its results and moving it up on your webpage.

By placing this content prominently at the beginning of your page, you reinforce its importance and relevance to the search query. This aligns more closely with what AI Overview deems valuable.

When analysing the search result of a keyword you’re targeting, you’ll notice that the AI Overview results pull paragraphs from different types of content to generate back to the user.

Logically, this should be a sound snippet of content that Google has validated as relevant and worthy of putting into their result.

Imagine you run a cooking blog, and one of your pages is about “Quick and Easy Vegan Dinner Recipes.” Upon checking the AI Overview results for this query, you may notice that AI Overview is pulling a paragraph from midway through your article that succinctly lists three popular vegan recipes.

Here are some steps to take to rank for AI Overviews

1. Recognise which part of your content AI Overview is using. In this case, it’s a specific paragraph listing the three vegan recipes.
2. Edit your page so that this paragraph is now at the top of the article, preferably in the introductory section. This makes it immediately visible to users and search engines.
3. Ensure this paragraph is not just a list, but includes a brief introduction to the article. For example, “This guide will explore three easy-to-make vegan dinners that are perfect for a quick and nutritious meal.”
4. Beyond moving the paragraph, consider enhancing it with additional useful information like estimated cooking times or why these recipes are great for busy weeknights. This provides more value to the reader and may further align with AI Overview’s focus on delivering comprehensive answers.
5. After making these changes, use Google Search Console to request a reindexing of the page, then do a re-post on your social media channels. This informs Google about the updated content and prompts a reevaluation of the page for AI Overview results.

This can also help you optimise your existing blogs from a traditional search user perspective.

eCommerce SEO for AI Overview

Adapting to the new AI Overview shopping experience, consider the following in terms of your product pages:

• Build out some strong copy on the pages that have specific product features and benefits on the page itself. Make it unique so it’s not duplicated from other websites.
• Make sure the product and review schema markup in the code is clean and shows no errors.
  Use high-quality and unique imagery. Consider adding video content that demonstrates the product for a more profound user experience.
• Ensure your Google Merchant Center product feed is up to date.

This isn’t rocket science.

These are all common best practices for ecommerce SEO. If your brand still needs to action this, I recommend making this an SEO priority for your future marketing efforts.

SERPs are becoming category pages

Google is focusing on making their search results (SERPs) to be more like category pages you see on shopping sites. This includes adding features like navigation options and highlighting products directly in the search results.

This is mainly because they want to compete better with big online stores like Amazon. A lot of people start their shopping searches on Amazon, and not on Google. This is a challenge for Google, so AI Overview is a perfect opportunity to push this strategy forward.

Product pages on websites are becoming even more important and are almost like the new “category pages.”

Diversify Your Content

AI Overview pulls from written content. In search results themselves, they can also use resources like YouTube videos and LinkedIn videos. Google Perspectives may also have a significant role in the content used for AI Overview in the future.

With any SEO-written content, having the correctly targeted term, heading structure, and relevant content is considered the baseline.

Amplify and repurpose that content into non-SEO-focused assets like podcasts, webinars, whitepapers, or infographics.

In the future, AI Overview may lean on more of this audio/video content, so having this as a standard practice in your content workflow will only increase your content engagement.

Add Schema

Pages that appear in AI Overview have a solid structured data setup.

It’s crucial to have your page’s schema markup, like “Organisation,” “BlogPosting,” “Person,” “Reviews,” and others, in good shape as these are frequently included in search results.

Google needs shortcuts to understand what type of content you serve – a blog post, recipe, page, or video. You will need to know the difference faster, and having a suitable schema in place will assist.

With better categorising of your content via schema, it might end up getting picked for these AI Overview snapshots more often.

Focus on Information Gain

In 2020, Google introduced a patent for information gain scoring; a method that assesses the value of search results based not only on relevance, but on additional factors.

In simple terms, information gain measures how much additional, valuable knowledge you get from new data compared to what you already know. It’s like reading a series of articles on the same topic; if each article offers new facts or insights you didn’t get from the previous one, then each has a high information gain. On the other hand, if they all repeat the same information, their information gain is low.

For search engines or recommendation systems, understanding information gain allows these systems to provide users with content that is more likely to be new, relevant, and informative – enhancing the overall user experience.

With AI Overview, Google will take sources from different websites; if your content is the same as other pages, it’s a copycat of the rest.

If you are serving new, helpful, fresh, and unique perspectives on a topic – you’re setting up your page to hit information gain and get served in AI Overview results.

For example, you add an early trend on something before it hits critical mass.

You wrote a blog post on improving sleeping habits, and all the top 10 articles mention the same technique. You added a unique tip, “mouth taping,” before it took off all over the social media health space.

Add your unique opinions and perspectives

Creating content that includes unique insights, opinions, and expert commentary will benefit your business – something that AI snapshots can’t easily replicate.

People are looking for trusted opinions, and Google is aware. This is why they introduced EEAT. This is also why many users add “Reddit” to their search queries – to seek natural human perspectives instead of just a collection of facts.

If your content is aimed at engaging potential customers, (middle and bottom-of-funnel content), and is detailed, opinionated, includes original data, incorporates rich media, and features viewpoints from customers, employees, and other real individuals – you’re moving in the right direction.

You can target longer tail terms in the follow-up feature that have minimal search volume initially, but will grow as users start interacting with this new search style.

Critical Observations and Recommendations for AI Overview

Here are a couple of critical observations from AI Overview and why it’ll be interesting to see how Google rolls it out in the long-term:

• The Search Generative Experience has only had six months of testing and is still an experiment. The caveat is that with an experiment, all the techniques you can use are experimental in approach as well; no one can say they’ve cracked the code on AI Overview just yet.
• Most average users of Google and business owners need a deeper awareness of AI Overview at this stage. When more releases and updates come out, we’ll learn more about how users interpret this change.
• Google has also been criticised for using the website’s original content in these AI snapshots. They’ve resolved this by adding citations for each section they’ve used. Content creators may lose click-throughs if people get what they need out of AI snaps, decreasing the incentive to create high-quality content, which is the basis of the search.
• I’ve noticed that Google will sometimes use outdated content for their generated results. This also happens in the typical search results, so it will likely will also occur with AI Overview.

Will AI Overviews be released?

My prediction is some of it will, but not the entire feature.

It’s way too different from the normal search format for the normal audience to be properly informed about. It’s still quite laggy and misfires at times, plus it’s an experiment – and sometimes experiments are retired. I could be completely wrong here, but hey – that’s all coming from working in SEO my friends!

Update:

In May at Google I/O, Google introduced the first steps of a new chapter for Search with AI Overviews, initially launched in the US and later expanded to additional countries in August.

Last month, the feature underwent local testing, and starting from now, it is being rolled out across Australia as we speak.

Final Word: Prepare for the AI Overviews

Having a strong SEO foundation and being seen as an authority in your field is essential to making the most of AI Overviews. This means creating content that directly answers people’s questions and needs.

The key is always to focus on creating content that’s good for search engines but beneficial and exciting for people.

While it brings some challenges, it’s also full of opportunities. By using these techniques, you’re getting ahead of the game.