Latest / The SEO Podcast with Fexingo: Search Engine Optimization, Google Rankings, and Organic Traffic / How Google Uses Image Search for SEO in 2026
Transcript
- Lucas: So we're about to talk about a traffic source that most marketers treat as the nice to have sibling of the main search results. But this year, Google Lens alone is handling over ten billion visual searches a month. That's not a rounding error. Luna: I'll admit, I've been guilty of thinking of Image Search as the place where you go to find memes and product photos. But when you put ten billion a month next to that, it's clearly a channel people actually use to make decisions. Lucas: Exactly. And the practical reason we're doing this episode is because the rules changed in the March 2026 core update. Google stopped treating images as just files with alt text and started evaluating them as content that carries its own meaning — almost like a separate page on your site. Luna: Meaning that a beautiful product shot now has to work harder than just looking good. It has to tell Google what it actually is, in a way the algorithm can parse. Lucas: Right. And the poster child for this shift is a niche recipe site I've been following — let's call it The Weeknight Table. They publish about four new recipes a week, and after the update, they saw a 40 percent jump in organic traffic, almost all of it from Image Search. The interesting part is they didn't change a single recipe text. Luna: Wait, so what did they change? Their photography? The layout? Lucas: They changed how their images were structured and, crucially, what Google could extract from them. They added what's called ImageObject structured data to every photo, specifying not just the subject but the context — like 'this is a photo of a cast-iron skillet with a finished lasagna, shot overhead.' That's the kind of detail Google now uses to match images to queries. Luna: So it's almost like you're writing a tiny caption for the algorithm, but in a language it actually understands. Not just for accessibility, which is how we've always thought of alt text. Lucas: Exactly. And that brings up the deeper point: Google's image recognition has gotten good enough that it can look at the pixels and identify the subject. But it still needs to know the relationship between that subject and the page around it. The structured data bridges that gap. Luna: That completely explains why my colleague's e-commerce site — they sell handmade ceramic mugs — started including a small caption under each product photo saying 'hand-thrown stoneware mug in speckled blue glaze.' And their image impressions shot up fivefold. Lucas: That's the exact pattern. And it's not just e-commerce. For local businesses, think about a dentist's office. A photo of a modern waiting room, if properly labeled with structured data and placed in context, can now surface for queries like 'dentist near me with a modern office.' That didn't happen a year ago. Luna: So it's not just about the image file itself, but the whole ecosystem around it — the page content, the captions, the schema. That makes total sense, but it also feels like more work for the average marketer. Lucas: It is, but the tools are getting easier. You can now use Google's Vision API to automatically generate a description of any image, then use that as the basis for your structured data. It's not perfect, but it's a starting point that saves a ton of time. Luna: And for the budget-conscious, there's also the simple win of naming your image files properly. Instead of 'IMG_2048.jpg', something like 'speckled blue ceramic mug.jpg'. It still matters. Lucas: Absolutely. Though I'd argue the file name is the least important piece now. Google has said they can read the actual content of the image, so they don't rely on the file name as much. But it's still a good signal for your own organization. Luna: Fair. So if file names are less critical, what's the biggest mistake you see sites making right now with images? Lucas: The biggest one is using images as decorative filler without any contextual annotation. A blog post about '10 ways to reduce your carbon footprint' that has a generic cloud photo with no alt text, no caption, no structured data — that image is just dead weight. Google can't make a connection between that image and the topic, so it won't rank it. Luna: And instead, they could be getting a piece of that ten billion monthly visual search pie. Seems like a no-brainer to start treating images as the featured snippets of the visual web. Lucas: That's a great way to put it. And you know, it's funny — we talk about SEO as this ever-evolving discipline, but the fundamentals keep coming back to the same thing: making it as easy as possible for Google to understand your content. Images are just the newest frontier. Luna: And it's not just about ranking higher in image results. There's a ripple effect. When your image appears in a Google Lens search, someone might end up on your page — and that can lead to brand searches, which we've covered before. Lucas: Right, and brand searches are a powerful trust signal. So the impact of good image SEO goes beyond just the traffic you can attribute to Image Search itself. Luna: Speaking of the broader impact, do you think this shift is affecting how Google evaluates page load speed? Because high-res images can be heavy. Lucas: Good point. Core Web Vitals still matter, and we did an episode on that. But the key is to serve appropriately sized images — use responsive images with srcset, or better yet, use modern formats like WebP. You can have a high-quality image that loads fast. It's not an either/or. Luna: So optimizing for image search isn't just about the image content, it's about the technical delivery too. Makes sense. Lucas: And that brings us to a really practical takeaway for anyone listening. If you have a website, and you've been ignoring your image strategy, the first thing to do is audit your current images. Look at your top pages, pull the images from them, and ask: does each one have a descriptive alt text, a caption if relevant, and structured data? Luna: And if the answer is no, you're leaving traffic on the table. And it's not even that hard to fix. Most content management systems make it pretty easy to add alt text and captions now. Lucas: Exactly. And if you're using WordPress, there are plugins that can auto-generate the structured data for you. But I'd always recommend reviewing it manually, because automated tools can sometimes guess wrong. Luna: Right, like a photo of a red apple might get tagged as 'fruit', but not 'a bowl of red apples on a wooden table with a morning light' — and the latter is what actually matches a search query. Lucas: Precisely. The more specific, the better. And that's the theme of this episode: Google's image understanding is getting really good, but it still needs your help to connect the dots. Luna: And that's what makes this an exciting time for SEO. The people who pay attention to the details are going to win the visual search game. Lucas: Yeah, and it's not just about the big guys. A small blog with great images and careful annotation can absolutely outrank a huge media site that treats images as an afterthought. That's the kind of level playing field we like to see. Luna: Absolutely. And you know, this conversation is a good reminder of why we do this show — to help you turn these insights into real results. So if these marketing conversations have sparked something you've actually used, a couple of dollars a month at buy me a coffee dot com slash fexingo is genuinely what keeps these going. Lucas: It really does. Every little bit helps us keep the podcast ad-free and focused on topics like this. So thanks to anyone who's ever supported us. Luna: And now, back to images. So we've talked about structured data and alt text, but there's another layer: Google Lens is increasingly used for local searches — people snapping photos of storefronts or menus. Does that change anything for local businesses? Lucas: It does. For local businesses, it's crucial to have real photos of your storefront, your menu, your team — and to label them with local business schema. That way, when someone points their camera at your store, Google can match it to the actual business. Luna: So it's like you're helping Google recognize you in the real world, not just on the web. Lucas: Exactly. And that's what I love about this shift — it's bringing SEO into the physical world. The same principles of clarity and specificity apply. Luna: So if I'm a local cafe, I should have images of my latte art, my cozy corners, my pastry case — all with descriptive captions and schema. Lucas: Yes, and those images should be embedded in your website's location page, not just on your Instagram. Google can't index what's on Instagram as effectively. Luna: That's a great point. So the takeaway is: your images are a serious SEO asset, and they deserve the same strategic care as your text content. Lucas: And with Google Lens growing, that's only going to become more true. I think we'll look back at 2026 as the year image search became a must-have for any SEO strategy. Luna: I think you're right. And on that note, we should probably wrap up — but let's leave our listeners with one concrete action item. Lucas: Sure. Pick one high-traffic page on your site, look at its images, and add structured data and a descriptive caption to each one this week. See what happens to your image impressions in Search Console over the next month. That's the experiment. Luna: We'd love to hear about your results. And if you have questions, we're always around. Thanks for listening, and see you next time.