AI Overviews vs Featured Snippets: What Cardiff Businesses Need to Know
AI Overviews vs Featured Snippets: What Cardiff Businesses Need to Know
If you've searched anything on Google recently, you've probably noticed a large block of AI-generated text sitting right at the top of the results page, often before you even see a traditional list of blue links. This is Google's AI Overview feature, and it has changed the way people find and interact with information online. For years, businesses have chased "featured snippets," the highlighted answer boxes that used to sit at the top of results. Now there's a new, more dominant format to understand. This post breaks down the difference between the two, what the data tells us about how they affect traffic, and what it means for how you approach your website content.
What Are Featured Snippets?
Featured snippets have been around for years. They're the boxed answers you see at the top of some search results, usually pulling a short paragraph, a list, or a table directly from a specific web page, along with a link to that page. If you've ever searched "how long does a cold last" and seen a neat answer box with a link underneath, that's a featured snippet.
Crucially, featured snippets always credit and link to a single source page. Even though the answer appears directly in the search results, the business behind that content still gets visible attribution and, often, a decent number of clicks from curious searchers wanting more detail.
What Are AI Overviews?
AI Overviews are a newer, more complex feature. Rather than pulling a single answer from one page, Google's AI system generates a synthesised response by drawing on multiple sources at once, blending information together into a conversational summary. Sources are usually credited with small links, but they're often less prominent than a featured snippet's link, and the overview itself may fully answer the searcher's question without them needing to click through to any website at all.
This is the crucial difference. A featured snippet tends to draw a curious click. An AI Overview often satisfies the search entirely on the results page itself.
What the Data Actually Shows
This shift has real, measurable consequences for website traffic. Multiple studies tracking click-through rates have found that pages appearing in AI Overviews see noticeably lower click rates compared to when the same query previously triggered a traditional featured snippet or a standard set of blue links. In some cases, the reduction in clicks has been substantial, particularly for straightforward informational queries where the AI Overview genuinely answers the question in full.
This doesn't mean visibility in AI Overviews is worthless. Being cited as a source still carries real brand value and can build trust, even if the click doesn't always follow immediately. But it does mean businesses relying purely on informational content to drive direct website traffic need to rethink their expectations and strategy.
What This Means for Different Types of Queries
Not all searches are affected equally. Simple factual questions ("what time does the library close," "how many ounces in a pound") are the queries most likely to be fully answered by an AI Overview, with little reason for anyone to click further.
More complex, nuanced, or commercially-driven searches behave differently. Someone searching "best accountant for a small business in Cardiff" or "how much does a kitchen renovation cost" is usually looking for more than a quick fact. They want comparisons, local options, pricing detail, and reassurance, things an AI Overview typically can't fully satisfy on its own. These types of searches still send meaningful traffic to well-optimised business websites.
This is an important distinction for local businesses in particular. Purely informational blog content aimed at very simple questions may see reduced direct traffic value going forward. Content aimed at helping someone make a decision, compare options, or take an action still holds strong value, both for rankings and for genuine business results.
How to Adjust Your Content Strategy
1. Focus more on decision-stage content. Rather than writing content purely to answer simple factual questions, focus more energy on content that helps people choose, compare, and decide. Buying guides, comparison pages, case studies, and local service pages tend to hold up much better against AI Overview competition than purely factual explainer content.
2. Don't abandon informational content entirely. Being cited in an AI Overview still has genuine brand value, even without a guaranteed click. It builds recognition and trust over time, and for competitive topics, being one of the credited sources is still far better than not appearing at all.
3. Structure your content clearly for extraction. Both featured snippets and AI Overviews favour content that's clearly structured with headings, concise definitions, and well-organised lists. Writing with this clarity in mind gives you a better chance of being pulled into either format.
4. Strengthen your other traffic channels. If a portion of your informational search traffic is likely to decline over time, it makes sense to diversify. This might mean putting more effort into email marketing, local partnerships, social media, or paid search for your most commercially valuable terms.
5. Track the right metrics. Rather than only watching keyword rankings, keep an eye on actual click-through rates and conversions for your key pages. A drop in clicks for a ranking you still hold might well be explained by an AI Overview appearing above it, and understanding this helps you make sense of what your analytics are actually telling you.
The Bigger Shift Behind All of This
What's happening here reflects a broader change in how people use search engines. Increasingly, simple questions get answered instantly without a website visit, while more complex, high-value decisions still require genuine research, comparison, and trust-building, exactly the kind of work a well-built website and good content can support.
For Cardiff businesses, this means the winning strategy isn't to chase every keyword indiscriminately, but to focus resources where a website visit and a genuine connection with a potential customer still make a real difference.
Need Help Reworking Your Content Strategy?
Understanding exactly which of your pages are losing clicks to AI Overviews, and which topics are still worth investing in, takes proper analysis of your specific traffic data. If you'd like a clear, honest look at where your website stands and how to adjust, the team at OmniSEO would be happy to talk you through it.
How to Build Topic Clusters That Get Cited by AI
How to Build Topic Clusters That Get Cited by AI
If you've noticed AI tools like ChatGPT or Google's AI Overviews confidently answering questions and pointing to certain websites as sources, you've probably wondered how those sites earned that spot. It's rarely luck. More often than not, those sites have built what's called a topic cluster: a structured group of connected pages that together demonstrate real, thorough knowledge of a subject. This post explains what topic clusters are, why they matter more than ever for AI visibility, and how you can start building them for your own website.
What Is a Topic Cluster?
A topic cluster is a group of related pages built around one central subject, usually anchored by a broad "pillar" page and supported by several narrower pages that each dig into a specific aspect of that subject.
For example, imagine you run an SEO agency in Cardiff. Your pillar page might be a comprehensive guide titled "SEO for Cardiff Businesses: The Complete Guide." Around that, you'd build supporting pages covering specific pieces of the puzzle: local SEO, technical audits, content strategy, link building, and Google Business Profile optimisation. Each supporting page links back to the pillar page, and the pillar page links out to each supporting page, creating a clearly connected web of content.
This structure does two things. It helps human visitors navigate naturally from a broad overview to the specific detail they need, and it helps search engines and AI tools understand that your website has genuine, comprehensive expertise on the topic, rather than one isolated page written in a vacuum.
Why This Matters So Much for AI Search
Traditional search engines rank individual pages. AI tools work a little differently. When an AI system is generating an answer, it's often trying to piece together a complete, accurate response from multiple sources, and it favours sources that demonstrate depth and internal consistency across a subject rather than a single disconnected page.
Think about it from the AI's perspective. If it needs to answer a detailed question about local SEO, and it finds a website with one thin page briefly mentioning the topic, that's a weak source. But if it finds a website with a well-organised cluster of pages covering the topic from multiple angles, each one clearly linked and consistent with the others, that's a much stronger signal of genuine authority. It's the difference between asking a random person a question and asking someone who has clearly spent years studying and writing about the subject.
This is also why topic clusters tend to perform well for traditional rankings too. Google has long favoured websites that show real depth on a subject over those that scatter thin content across unrelated topics.
How to Plan Your First Topic Cluster
Building a topic cluster doesn't need to be overwhelming. Here's a practical way to approach it.
1. Choose a broad topic that matters to your business. This should be something genuinely central to what you do, not a tangential idea. If you're a local accountant, your pillar topic might be "Small Business Accounting in Cardiff," not something unrelated like general financial news.
2. Map out the subtopics people actually ask about. Use tools like Google's "People Also Ask" boxes, AnswerThePublic, or simply your own experience of customer questions to list out the specific areas your audience wants to know about. For an accountant, that might include tax deadlines, bookkeeping software, VAT registration, and payroll management.
3. Write your pillar page first, but keep it broad. The pillar page should give a solid overview of the whole topic without trying to cover every detail. Think of it as a well-organised table of contents that's also genuinely useful to read on its own.
4. Build out supporting pages one at a time. Rather than trying to write everything at once, tackle one supporting page per week or month, whatever's sustainable for your business. Each one should go into proper depth on its specific subtopic and link clearly back to the pillar page.
5. Link consistently and logically. Every supporting page should link to the pillar page and, where relevant, to other supporting pages that cover closely related ground. This is often the step people skip, but it's what actually creates the "cluster" effect rather than just a pile of separate blog posts.
6. Keep it updated. Topic clusters aren't a one-off project. As your industry changes or new questions emerge, add new supporting pages or update existing ones. This keeps the cluster current and shows both search engines and AI tools that the information is actively maintained.
Common Mistakes to Avoid
A few things can undermine an otherwise good topic cluster strategy.
Spreading yourself too thin across topics that don't connect. If your supporting pages don't clearly relate back to your central pillar topic, you end up with a scattered blog rather than a genuine cluster. Focus on depth in a few core areas rather than breadth across everything.
Forgetting internal links. This is the glue that holds a cluster together. Without it, you just have a collection of unconnected pages that happen to be about similar things.
Duplicating content across pages. Each page in the cluster should cover its specific angle without repeating large chunks of text found elsewhere on the site. Overlap confuses both readers and search engines about which page is the authority on a given point.
Neglecting the pillar page over time. It's common to build supporting content diligently while letting the pillar page go stale. Since it's often the most-visited page in the cluster, it deserves regular attention too.
A Realistic Timeline
Building a genuinely strong topic cluster from scratch usually takes a few months rather than a few weeks, especially if you're doing it alongside running a business. That's completely normal. The value comes from consistency and depth, not speed. A cluster of eight or ten genuinely thorough, well-linked pages will outperform fifty scattered, shallow blog posts almost every time, both for traditional rankings and for the kind of authority that gets you noticed by AI tools.
Where OmniSEO Can Help
Planning and building a topic cluster from the ground up takes time, research, and a fair amount of writing, which is often the part that gets pushed to the bottom of the to-do list when you're busy running a business. If you'd like help mapping out a cluster strategy for your website, or turning your existing scattered blog posts into a properly connected structure, our team at OmniSEO would be glad to walk you through what that could look like for your specific industry and goals.
What Is Semantic Depth and Why Does it Beat Keyword Stuffing?
What Is Semantic Depth and Why Does it Beat Keyword Stuffing?
If you've been running a website for more than a few years, there's a good chance you remember (or still practice) the old approach to SEO: pick your target keyword, repeat it as often as you can get away with, and hope Google notices. For a while, that actually worked. Those days are firmly behind us. Today, search engines are far more interested in something called semantic depth, and understanding what that means could be the difference between content that ranks and content that quietly sits on page four of Google forever.
This post will explain what semantic depth actually is, why keyword stuffing stopped working (and can even hurt you), and how to write content that genuinely satisfies both search engines and the humans reading it.
A Quick History Lesson: Why Keyword Stuffing Used to Work
In the early days of search engines, ranking algorithms were relatively simple. They counted how often a word or phrase appeared on a page and used that as a rough signal of relevance. If you wanted to rank for "wedding photographer Cardiff," the logic was straightforward: cram that exact phrase into your title, your headings, your image alt text, and your body copy as many times as you could without the page reading like nonsense (and sometimes even then).
This created an obvious problem. Content written for algorithms rather than people tends to read badly. It's repetitive, unnatural, and often unhelpful, because the writer's main goal was pleasing a piece of software rather than answering a real question.
Google noticed this too, eventually, and its algorithms evolved significantly to combat it. Updates over the years have specifically targeted thin, repetitive, keyword-stuffed content, pushing it further down the rankings rather than rewarding it.
So What Is Semantic Depth?
Semantic depth refers to how thoroughly and meaningfully a piece of content covers a topic, including all the related concepts, questions, and context a reader would reasonably want to understand it fully. Rather than asking "does this page mention the keyword enough times," modern search algorithms ask something closer to "does this page genuinely explain this topic in a way that would satisfy someone who searched for it?"
Think about the difference between these two approaches to a blog post about "how to choose an SEO agency":
A shallow, keyword-focused version might repeat "choose an SEO agency" a dozen times, list three generic tips, and call it done. It technically contains the keyword, but it doesn't actually help anyone make a decision.
A semantically deep version would cover the actual questions someone has when trying to choose an agency: what different pricing models look like, what red flags to watch for, what questions to ask in a first meeting, how to interpret case studies, what a realistic timeline for results looks like, and how local versus national agencies might differ. It naturally uses variations of the topic (like "SEO provider," "search marketing partner," "digital marketing agency") because that's how people actually write and speak about the subject, not because a keyword tool told it to.
Google's language models are now sophisticated enough to recognise the difference between these two approaches almost instantly. The second version demonstrates real understanding of the topic. The first one is just going through the motions.
Why This Matters Even More With AI Search
This shift towards rewarding depth over repetition has become even more important with the rise of AI-generated answers, like Google's AI Overviews or tools like ChatGPT. These systems are trying to synthesise accurate, comprehensive answers from the content they find. A page with genuine semantic depth gives them much more useful material to draw from and cite. A thin, keyword-stuffed page simply doesn't offer enough substance to be useful as a source.
In other words, if you want your content to have a chance of being referenced in an AI-generated answer, and increasingly this matters for visibility, it needs to actually contain the depth of information a comprehensive answer would require.
How to Build Semantic Depth Into Your Content
The good news is that writing with semantic depth doesn't require any special software or technical skill. It requires a genuine understanding of your topic and your audience's real questions. Here's how to approach it practically.
1. Start with the questions, not the keyword.
Before writing, list out every question a real person might have about the topic. If you're writing about "SEO audits," that might include: what does an audit actually check? How long does one take? What happens after you get the results? How much should it cost? What's the difference between a technical audit and a content audit? Answering these thoroughly naturally builds depth.
2. Cover related subtopics, not just the core one.
A genuinely deep piece of content acknowledges the topics that sit next to the main one. A post about local SEO for Cardiff businesses might naturally touch on Google Business Profile optimisation, customer reviews, and local link building, because these are all part of the same broader picture. You don't need to write a full guide to each, but acknowledging them and linking to further reading shows real command of the subject.
3. Use natural language variation.
Rather than repeating your exact target phrase, use the words and phrases real people would actually use. If your keyword is "SEO agency Cardiff," your content will naturally end up including phrases like "digital marketing company in Cardiff," "search engine optimisation experts," or "local SEO specialists." This isn't about tricking an algorithm, it's simply how genuine, well-informed writing sounds.
4. Answer the "so what" question.
Depth isn't just about listing more facts. It's about explaining why those facts matter. If you mention that page speed affects rankings, explain what a slow page actually costs a business in real terms (lost visitors, lower conversion rates, frustrated customers), rather than just stating the fact and moving on.
5. Update and expand over time.
Semantic depth doesn't have to be achieved in one sitting. Some of the best-performing content online has been updated repeatedly over months or years, with new sections added as new questions emerge or as the topic evolves. This also signals to Google that the page is actively maintained and current, which is another trust factor.
What This Means for Your Existing Content
If you've got older blog posts sitting on your website that were written with the old keyword-density approach in mind, it might be worth revisiting them rather than starting from scratch. Often, the core idea is sound, it just needs expanding with the kind of context and related information that gives it genuine depth. This is usually far more efficient than writing brand new content, and it can breathe new life into pages that have been quietly underperforming for a while.
The Bigger Picture
At its heart, semantic depth is really just a return to good, honest writing. Write like you're genuinely trying to help the person reading, covering the questions they actually have rather than the ones a keyword tool suggested. Search engines have simply gotten much better at recognising when that's happening, and rewarding it accordingly.
If you're not sure whether your current content has the depth it needs, or you'd like a second pair of eyes on where the gaps are, we're happy to take a look. At OmniSEO, we spend a lot of time helping Cardiff businesses turn thin, keyword-stuffed pages into genuinely useful resources that both readers and search engines respond well to. Get in touch whenever you're ready for an honest conversation about your content.
Entity-First SEO: How to Build Topical Authority Google Trusts
Entity-First SEO: How to Build Topical Authority Google Trusts
If you've spent any time reading about SEO recently, you've probably come across the phrase "entity-first SEO" and wondered what on earth it actually means. It sounds technical, and in some ways it is, but the idea behind it is simple once you strip away the jargon. This post will walk you through what entities are, why Google cares about them so much, and what you can practically do to build the kind of topical authority that search engines (and increasingly, AI tools) trust.
What Is an "Entity" in SEO Terms?
In the world of search, an entity is simply a distinct, well-defined thing: a person, a place, a business, a product, a concept. Cardiff is an entity. Your business is an entity. So is "SEO audit" as a concept, or "vegan restaurant."
Google doesn't just read your website as a string of words anymore. It tries to understand what your website is about by identifying the entities mentioned on it and working out how they relate to each other. This is done through something called the Knowledge Graph, which is essentially Google's enormous map of real-world things and how they connect.
So when Google crawls your site, it's not just looking for the keyword "plumber Cardiff" fifteen times on a page. It's trying to work out: is this business actually a plumber? Does it operate in Cardiff? What services does it offer? Who runs it? Is it connected to other trusted plumbing resources or directories?
Why This Shift Matters for Your Business
Keyword matching used to be the main game. If you wanted to rank for "best coffee shop Cardiff," you made sure that phrase appeared enough times on your homepage. That approach still has some value, but it's increasingly weak on its own.
Search engines are getting better at understanding meaning and context rather than just spotting repeated phrases. This matters even more now that AI tools like ChatGPT and Google's AI Overviews are pulling information to answer questions directly, often without the person even clicking through to a website. These AI systems lean heavily on entity relationships to decide which sources are credible enough to cite.
In plain terms: if your website clearly establishes who you are, what you do, where you operate, and how that connects to other trusted sources, you have a much better chance of being recognised as an authority, both by traditional search rankings and by AI-generated answers.
How to Start Building Entity Authority
The good news is that you don't need a computer science degree to work on this. Here are some practical, achievable steps.
1. Be consistent with your business information everywhere.
Your business name, address, and phone number (often shortened to NAP) should match exactly across your website, Google Business Profile, social media, and any directories you're listed on. Inconsistencies confuse search engines about whether they're looking at the same entity or several different ones.
2. Use schema markup.
This is a bit of code added to your website that explicitly tells search engines "this is a business, this is its name, this is its service area, these are its reviews." Think of it as handing Google a clearly labelled fact sheet instead of making it guess from paragraphs of text. If that sounds intimidating, don't worry, this is exactly the sort of technical groundwork an SEO agency can set up for you without you needing to touch a line of code.
3. Build genuine topical depth, not just one page per keyword.
Rather than having a single page trying to cover everything about "SEO services," search engines respond better to a cluster of connected pages: one covering local SEO, one covering technical audits, one covering content strategy, and so on, all linking sensibly to one another. This shows real depth of knowledge rather than a single thin attempt to cover a broad topic.
4. Get mentioned and linked to by other credible entities.
This is where things like local press coverage, industry directories, and partnerships come in. If other trusted websites mention your business alongside relevant topics (say, a Cardiff business directory listing you as a "digital marketing agency"), that reinforces Google's understanding of what you are and where you belong.
5. Make your "About" and author information clear and complete.
Who is writing your content? What are their qualifications or experience? This matters more than many business owners realise. Clear authorship signals help both Google and AI models understand that a real, credible source stands behind the information.
A Simple Way to Think About It
Imagine you're introducing your business to someone at a networking event, not with a rehearsed sales pitch, but by clearly explaining who you are, what you do, and how that connects to things they already know and trust. That's essentially what entity-first SEO is asking you to do, except the "person" you're introducing yourself to is a search algorithm that then relays that understanding to millions of people searching every day.
The businesses that get ahead here aren't necessarily the ones with the biggest budgets. They're the ones who take the time to be clear, consistent, and genuinely informative about who they are and what they offer.
Getting This Right Takes Time, But You Don't Have to Do It Alone
Building entity authority isn't a one-off task you tick off a list. It's an ongoing process of consistency, clarity, and connection building across your website and beyond it. For many small business owners in Cardiff, that's a lot to juggle alongside actually running the business.
If you'd like a clearer picture of how your website currently comes across to Google, and where the gaps in your entity signals might be, our team at OmniSEO would be glad to have a look and talk you through it. No jargon, no pressure, just an honest conversation about where you stand and what would genuinely move the needle.