Masters of SearchPresented byRadyant
← All episodes
Masters of Search episode 19: Enterprise SEO | Roberto Grasiano, ex Head of Organic Growth & SEO @ Shutterstock | Masters of Search #19 cover art
EP 19·Nov 26, 2025

Enterprise SEO | Roberto Grasiano, ex Head of Organic Growth & SEO @ Shutterstock | Masters of Search #19

SpotifyApple PodcastsYouTube

Show notes

24 BILLION pages. That's the sheer scale at which Roberto Grasiano led SEO and organic growth at Shutterstock. I'm so excited that I could convince him to come on the Masters of Search to bring you 73 minutes of Enterprise SEO masterclass.

Here are my key takeaways, both for Enterprise SEO and for smaller sites:

  1. At enterprise scale, you can't crawl the site

Roberto couldn't audit individual pages at Shutterstock (24B pages, 500M indexed). Instead, he built comprehensive taxonomy mapping the entire site structure into clusters: broad categories (animals, cars, people), then subcategories (felines, SUVs), with exact page counts for each. This taxonomy determined which optimizations to test and how to measure impact. Even at 10,000 pages, taxonomy thinking beats keyword thinking. Group pages by intent clusters and user journey stages, not just traffic volume.

  1. Content optimized for bots actually costs you money

Roberto moved SEO content from top of category pages (above products) to below the fold. Theory said top placement ranks better. Result? 15% traffic increase. Users wanting specific products don't read text blocks. They want to see products immediately. When you optimize for bots over humans, you hurt user experience, which ultimately hurts rankings. Plus, bots crawling content cost server resources. His rule: content should be made for humans. If it's good enough, it'll show to the people who need it.

  1. ChatGPT commerce will be big

When I pushed back on instant checkout in ChatGPT (pointing to Meta's failed checkout), Roberto made a psychological argument: OpenAI builds something that feels like a friend who knows you. Remember when everyone said people would never buy things online? "No one will trust this, we need to touch products." Now our parents spend hours on Amazon. Shopping behavior changes over decades. People will buy through AI interfaces. It won't be fast, but it'll happen.

  1. SEOs are underpaid

Roberto's advice: learn your data, understand attribution, connect organic to revenue. Don't ask your boss to track revenue if you can't answer "how much money did you make last quarter?" Moving from vanity metrics (rankings, traffic) to commercial metrics (pipeline, revenue) gives SEOs negotiating power. We have the tools and data available. We just need to use them.

▶ Let's connect! 🔗 Niklas on LinkedIn: https://www.linkedin.com/in/niklas-buschner/ Radyant on LinkedIn: https://www.linkedin.com/company/radyant/ Roberto on LinkedIn: https://www.linkedin.com/in/robertograsiano/ Lottie on LinkedIn: https://www.linkedin.com/company/thankslottie/ Shutterstock on LinkedIn: https://www.linkedin.com/company/shutterstock/

Transcript

Full conversation

via podigee
  • 00:00:00Twenty four billion pages, approximately five hundred million indexed, when most SEOs think in terms of hundreds of pages, today's guest thinks in batches of tens of thousands.
  • 00:00:12So welcome to Enterprise SEO at scale.
  • 00:00:16My guest today is Roberto Graciano, who until recently was head of organic growth and SEO at Shutterstock, one of the world's largest visual content platforms.
  • 00:00:27He was operating at a scale comparable to giants like Canva, Booking.com and Etsy.
  • 00:00:33With over twelve years of experience across high-scale platforms like Farfetch, List and now Shutterstock, Roberto brings a rare combination of technical depth and strategic thinking to the table.
  • 00:00:46Roberto is now joining a UK-based startup called Lottie.
  • 00:00:50as his next chapter, which makes this the perfect time to extract everything he's learned, managing SEO for a site with twenty four billion pages and around five hundred million indexed URLs.
  • 00:01:03So before we dive in, Roberto, thanks so much for coming on.
  • 00:01:07Thank you, Nicholas.
  • 00:01:08Thank you for introduction.
  • 00:01:09Really happy to talk to you today.
  • 00:01:12And let's see if I can share some some knowledge and help our fellows search in their own world.
  • 00:01:19Yeah, I was already blown away, but by our like, ten, fifteen minutes get to no call.
  • 00:01:24We did a couple of weeks ago.
  • 00:01:26You just casually dropped some things.
  • 00:01:28So I think it will be a good talk for a lot of people.
  • 00:01:33But let's start maybe by going back a couple of years, because I'm always interested in how people got into SEO.
  • 00:01:40So how did you end up in SEO?
  • 00:01:44It was sheer luck.
  • 00:01:46I would say I used to work in a very different area before SEO.
  • 00:01:54After some time, I decided that I needed a break.
  • 00:01:57I took a six-month break when I was twenty-four and five years old and I didn't know what to do.
  • 00:02:05I did a very short PPC course at the time, thirteen years ago.
  • 00:02:11Then I got a job as a link builder in an SEO agency.
  • 00:02:16And I didn't know what SEO was and started from there.
  • 00:02:20So my first job was as a link builder.
  • 00:02:24Awesome.
  • 00:02:24And how did you like it?
  • 00:02:25Like link building, I mean, I think back in the day, like when was that?
  • 00:02:29What year was it?
  • 00:02:34Well, I think was when I started.
  • 00:02:39How can I reply to that?
  • 00:02:40I will say that very quickly.
  • 00:02:42I became a technical SEO.
  • 00:02:45I think backlink and link building at the time was very impactful, but also something that I did not enjoy myself.
  • 00:02:54So I fell in love with technical SEO shortly after.
  • 00:02:58Okay.
  • 00:02:58I was just asking because I mean, link building somehow has this bad reputation, right?
  • 00:03:03Of just, especially back in the day, spamming forums and like.
  • 00:03:08not always going for high authority domains, but also taking every cheap link you can get.
  • 00:03:16Yeah, probably a good move you made.
  • 00:03:19Yeah, it was fun at the beginning, but I only realized that I liked the code partying more, the backend and the challenges.
  • 00:03:31But it was a great introduction to SEO because it allows me to understand SEO fully.
  • 00:03:36not only one part of it, but be a generalist.
  • 00:03:40Got it.
  • 00:03:40But I mean, there is Tech SEO and there is Tech SEO at the scale that I described already in the introduction.
  • 00:03:47So I'm curious, was it an active decision to move into this enterprise SEO environment with the big sites I named in the introduction or was it also by accident?
  • 00:04:02A bit of both, I guess.
  • 00:04:05Technical SEO for smaller websites, they had a window of operation, right?
  • 00:04:12You have only so much to do until the website is in a proper shape to scale from there.
  • 00:04:18And then you move to content, to growth strategies, which at the time I wasn't very comfortable with.
  • 00:04:26I really wanted to stay in Technical SEO.
  • 00:04:29So after a couple of years in Technical SEO, I got offered from List.
  • 00:04:33which was my first real huge marketplace.
  • 00:04:40When I took the chance, I knew that it would be a challenge.
  • 00:04:45I got hooked to the marketplace to the scale because you learn so much things that you don't know on the smaller websites.
  • 00:04:56When you talk about technical SEO for smaller websites, you're going to start to think.
  • 00:04:59and canonicalization strategies and dexation, internal linking, page speeds and so forth, but when you go to billions of pages in marketplace, everything becomes a bit more challenging and a bit more complex, which requires regular interactions or regular optimizations to achieve a compounding growth that we expect from SEO.
  • 00:05:30So it was a bit of both.
  • 00:05:33I was getting bored of these smaller websites and then List came in and was like, oh, hell yeah.
  • 00:05:40I think this is going to be interesting.
  • 00:05:41So since List, I tried to stay on enterprise sites.
  • 00:05:48I can imagine that a lot of people listening and viewing are familiar with List and like what they do, but can you just give a very short a rundown of what List does and the scale they were at when you joined them back then?
  • 00:06:03and like just in terms of maybe number of pages, in terms of the people working there, just whatever you feel is relevant.
  • 00:06:13Yeah, it was six years ago, seven years ago, so pardon my memory.
  • 00:06:19But List is a marketplace that works in an affiliate model.
  • 00:06:23So it's a fashion platform.
  • 00:06:27And we acquire users, when the users click on a product, we redirect them to the seller.
  • 00:06:35So it doesn't hold any stock of products.
  • 00:06:40They try to focus on a luxury fashion side of things.
  • 00:06:46So think of your Gucci's or Prada's.
  • 00:06:48And when I joined, I believe that they had around four hundred million pages.
  • 00:06:56at the time, spent across different languages.
  • 00:07:00So I think it was my first time that I joined a company and say like, oh yeah, how many pages do we have?
  • 00:07:06Like around four hundred million.
  • 00:07:09I was like, wait, are you sure you saved the right number?
  • 00:07:13It's not four hundred thousand.
  • 00:07:15No, no, no, four hundred million.
  • 00:07:15Like, okay, okay, what I'm gonna do here now.
  • 00:07:20But yeah, it was my first impressions and shock on the size and scale of the operation.
  • 00:07:26But what did you do to like get familiar with the scale?
  • 00:07:30because I can imagine like for people that.
  • 00:07:34Either work with smaller sites or work at a company that is smaller and maybe this company is in the face of scaling up.
  • 00:07:41so there are always interesting.
  • 00:07:45Business models where you have a lot of pages like in your index or people that maybe transition.
  • 00:07:51from one job to the other.
  • 00:07:53So basically doing the opposite way that you did, not from Shutterstock to Lottie, but maybe the other way around.
  • 00:08:01What would you say to those?
  • 00:08:03How can you get familiar with this scale to not be completely overwhelmed?
  • 00:08:10Google Search Console is your friend.
  • 00:08:12I think it's a very good starting point to start to understand how the website operates.
  • 00:08:20And I know that a lot of people out there will say like, oh, yeah, but we use crimson frog or sight bulb or crawlers to understand.
  • 00:08:29Unfortunately for sites of this size is really rare to have a crawler that can actually give you a full picture.
  • 00:08:37So you need to think on the templates and site structure mindset.
  • 00:08:44So you start to understand how the... folders work, how the site is distributed, how the categories work, and how the products are presented, and then you audit smaller portions of those templates.
  • 00:08:59When I say smaller, it's still not small.
  • 00:09:01We're talking about twenty, thirty million pages, but when you see the grand scheme of things, it's less than ten percent of the site structure.
  • 00:09:11So you pretty much have technical data for ten percent of the domain.
  • 00:09:17And you hope that what you do actually scales correctly to the remainder that you don't have visibility.
  • 00:09:25So Google Search Console is a friend.
  • 00:09:27And of course, if you have capacity, you build your own tools in-house.
  • 00:09:33I think this is the best way that enterprise websites works.
  • 00:09:37Because we know our website, so we are the best ones to build actually solutions.
  • 00:09:43Awesome.
  • 00:09:45It sounds like the perfect career progression list.
  • 00:09:50then far-fetch obviously also luxury fashion from my understanding.
  • 00:09:56similar model to list.
  • 00:09:57then shutter stock.
  • 00:09:58i think everybody knows shutter stock probably one of the most well-known like digital brands worldwide.
  • 00:10:05but now you're making like a very interesting move or you already made the move from shutter stock to lotty.
  • 00:10:14From my understanding, a UK-based startup that helps people find the best care homes and retirement communities.
  • 00:10:22Why the switch?
  • 00:10:25Very good question.
  • 00:10:27I guess I've been very deep in the corporation enterprise side of SEO, which is great.
  • 00:10:35When you are growing your career, you want to have a more exposure to decision making.
  • 00:10:41And when you want to become a leader or want to become part of the leadership, and you want to make those decisions based on what you do.
  • 00:10:52But after some time, I started to feel that I wanted to be closer to the action.
  • 00:10:58And most importantly, I wanted to use my skills in a mission-driven company in a mission-driven project.
  • 00:11:09I've been working with fashion and then media, which is, of course, really fun.
  • 00:11:15I love working with fashion and e-commerce, something that I miss from time to time.
  • 00:11:22But for me, it was time to look for something more meaningful.
  • 00:11:27outside from only conversions and revenue, but also make a real impact in people's lives.
  • 00:11:34And lotty works in a very sensitive topic of every family.
  • 00:11:44When you get to the part that a loved one needs caring and you can't cope anymore, it's really difficult conversation to have a moment to be.
  • 00:11:55Lotty tries to make this easier.
  • 00:11:58by providing information, by providing assistance.
  • 00:12:03Of course, it's still a business.
  • 00:12:04Of course, we still need to generate revenue.
  • 00:12:08But the core of the company is very mission-driven.
  • 00:12:12And for me, it was something that I really wanted.
  • 00:12:16And also, get my hands dirty again with SEO.
  • 00:12:20I was in a very top-level decision-making process, which is fine.
  • 00:12:25But when you start to have six hours per day in meetings, just to, you know, a yes or no or to help of something, you start, oh man, this is my life.
  • 00:12:35This is what I'm going to do for the rest of my existence.
  • 00:12:37I don't want that.
  • 00:12:38So yeah, moving to Lotte takes a lot of boxes.
  • 00:12:46And yeah, it will be fun.
  • 00:12:48It's been fun so far.
  • 00:12:50And I think it's going to continue to be fun.
  • 00:12:53But is an adjustment.
  • 00:12:57And I know that a lot of people will look into Lottie, since they moved there, they're going to keep a close eye on it.
  • 00:13:03So there's a lot of pressure to make it work, but we'll get there.
  • 00:13:07Cool.
  • 00:13:08What will be your role or what is your role at Lottie and how can we imagine your... How does it look like if Roberto now gets his hands dirty again with SEO at Lottie?
  • 00:13:22Well, luckily enough, this time I'm going to be able to crawl a full website.
  • 00:13:26It's been years since I did that.
  • 00:13:31I guess a lot easier marketplace is still continuing the marketplace, but there's a very small marketplace.
  • 00:13:40What you're going to see is me working in the foundation of the website, how the website works and operates, how it presented back to Google and to users.
  • 00:13:49and preparing the websites to future expansion by meaning European expansion or US expansion or whatever the company decides to do.
  • 00:14:01So my job is to make sure that the foundations are correct and that we have the correct growth mindset of the representative information to capture the user in a very stressful moment.
  • 00:14:15we don't want to direct the user to a content that is not useful, so is looking to the entire journey from discovery to acquisition and to support.
  • 00:14:29What you're going to see is me deep diving in the care sector, talking about the carers that we have in-house, talking about different people in the business, hopefully visiting some care homes to understand how is actually operating in real life and really know their pain and hopefully this will help me to ease that pain in the moment with a CEO and discoverability.
  • 00:14:57Super interesting.
  • 00:15:00Now let's take a step back and look a little bit at your last tenure again, which was obviously at Shutterstock.
  • 00:15:10because So when we had our get to know call and you dropped this number of twenty four billion pages, I think you remember I was a little bit shocked, like positively speaking.
  • 00:15:22Since I have never managed a site at that scale, I can only imagine that it's probably like in the top fifty pages worldwide.
  • 00:15:31So I can imagine probably an Amazon, maybe a booking dot com, maybe like pages like that are at the same size.
  • 00:15:40But can you give me and also people watching and listening that I would say, ninety nine percent of them have probably also never seen an SEO operation at that scale from the insights?
  • 00:15:55What does it look like?
  • 00:15:56Or how does it feel like to manage SEO day to day on that scale?
  • 00:16:03I don't know if I know how to describe it, I guess.
  • 00:16:08The shock that you had when I told you was the same shock that I had when I joined.
  • 00:16:14It took me a couple months to digest the project that I had in my hands when I joined Shutterstock.
  • 00:16:23Not going to say that it was, oh no, easy peasy, I got in and I knew what to do, what to go, I didn't.
  • 00:16:31You really need to spend a lot of time looking into data points and different sources.
  • 00:16:39And when I joined, Shutterstock was still, and this is fun to say, but it's still on this infancy on the SEO strategy.
  • 00:16:49So you think like it's twenty years website, they've done SEO for the past twenty years, but it's usually not the reality for many, many of these companies.
  • 00:17:01So when I joined, I helped to build the SEO foundation work, meaning how can we see the server log files at this scale?
  • 00:17:13I mean, can you imagine how many bots are hitting Shutterstock every day?
  • 00:17:19And we have some type of visibility on the site structure.
  • 00:17:25I mean, I believe at the time, Shutterstock had around two hundred and fifty million category pages.
  • 00:17:31So, how do you see this structure?
  • 00:17:33How those pages are created?
  • 00:17:35So, it was two months of learning in Clip List's night, I guess, thinking about what I'm going to do tomorrow.
  • 00:17:47How am I going to see these data points?
  • 00:17:50But I was fortunate enough to have an amazing team with me.
  • 00:17:56And when I got unworded after two months, I found a bit less scared.
  • 00:18:04I'm not going to say the word that I used to my manager at the time, but I was very scared when I joined.
  • 00:18:10And after two months, I was more confident that I knew what to do.
  • 00:18:18You cannot crawl the websites for reasons.
  • 00:18:22So you rely a lot on log files.
  • 00:18:25You really understand.
  • 00:18:27And you build your own log files and answers.
  • 00:18:31we have amazing tools in the market like BotFi, JetOctopus, Conductor.
  • 00:18:37But I don't think there's a company in the world at that side that has the budget on board these tools.
  • 00:18:47You learn to do everything by yourself.
  • 00:18:50My SQL skills are very top notch now.
  • 00:18:54You can give me a snowflake dashboard and I can query away.
  • 00:19:00But that was my onboarding experience.
  • 00:19:03I was also scared.
  • 00:19:05Crazy.
  • 00:19:06The thing that I was really pondering is when you started thinking about optimizations and like how do I approach the different clusters.
  • 00:19:20So for, I would say, regular SEO people, they think about, hey, okay, we have this category there the category here.
  • 00:19:28we maybe have these like twenty to fifty block posts and it's something that they can like still comprehend.
  • 00:19:35so i will work on that.
  • 00:19:36it will probably move the needle if done right.
  • 00:19:39so and i was thinking with twenty four billion pages approximately five hundred million indexed.
  • 00:19:44so if you think about a thousand pages like a cluster of a thousand pages.
  • 00:19:49It feels too small.
  • 00:19:50if you think about ten thousand pages it feels too small.
  • 00:19:53even if you think about a hundred thousand pages it still feels too small.
  • 00:19:57and if you're looking at a million pages it's already less than one percent of whole index pages.
  • 00:20:03so.
  • 00:20:05How did you approach like I mean the lock files and this?
  • 00:20:10I think I get an idea of that.
  • 00:20:12but if you then.
  • 00:20:14like try to go into some optimizations like in terms of, okay, let's work on this category, let's work on that page type.
  • 00:20:21What was a number of pages where you felt like, okay, this is not too small, this is okay to focus on?
  • 00:20:29Taxonomy is the keyword that you're looking here.
  • 00:20:33We worked tirelessly to create a comprehensive taxonomy of the website, meaning we wanted to.
  • 00:20:45It's not decreased, but we wanted to understand the site structure by having less data points.
  • 00:20:53I know it's weird to hear, but I'm going to try to explain.
  • 00:20:58If we have three hundred million category pages, for example, how do you categorize them in clusters?
  • 00:21:06There's really difficult.
  • 00:21:08If you think about cars, you have hundreds and thousands of brands and types, et cetera, to understand.
  • 00:21:17if you think about animals, thousands and millions of animals.
  • 00:21:22So we really tried to create a very concise taxonomy that helped us to map the websites from homepage to top level categories, which is very broad words, animals, cars, people, businesses, and so forth.
  • 00:21:42And then from there, create the subcategories for those type of very broad words.
  • 00:21:49So from animal, we go to fill lines, memos, et cetera, for cars, sports cars, SUVs.
  • 00:21:58We had a very good data team that helped us to understand the data that we had in the backend, because when a contributor upload an image or a video to Shutterstock, we ask a lot of these questions.
  • 00:22:13And then this help us to map in the back end, this structure.
  • 00:22:17So create a very interesting taxonomy.
  • 00:22:22And from there, we had the number, right?
  • 00:22:25So we knew, for example, that's not a real number, by the way, but it's just an example.
  • 00:22:29We know that we have a hundred million category pages from animals.
  • 00:22:35And inside this, we have ten million that are for felines.
  • 00:22:38And inside these ten million category pages, we have thirty million assets.
  • 00:22:45And then we categorized them from A to Z. So we had this whole list, and when you want to do a test, for example, a canonicalization strategy for pagination, we would divide the control and testing by alphabets.
  • 00:23:05So we're going to say we're going to test an accusation from a category starting from A to C. And then we're going to measure against D, E, and F, and see how it works.
  • 00:23:16So it was never granular to a page level.
  • 00:23:23But we had enough data to move the needle at least a bit to make the decisions.
  • 00:23:29I hope that answered the question.
  • 00:23:31Yeah, it makes sense.
  • 00:23:32But I think maybe let's try to dissect it a little bit more.
  • 00:23:36So are you saying that at this scale basically all SEO work or like ninety five percent of SEO work is on this structural level, like thinking about how to categorize your inventory, like in terms of taxonomy, categories, types, attributes,
  • 00:23:55etc.?
  • 00:23:56Yes.
  • 00:23:58It's the first step to understand what to do next.
  • 00:24:02So let's say that we want to introduce a new internal link component in a page that has a specific algorithm that we worked on.
  • 00:24:12We used this structure to identify the pages that we want to test and to measure the impact.
  • 00:24:18It was very controlled because we knew exactly what pages, not in the granular level, so not every URL, but what category of pages and what type of pages we were testing.
  • 00:24:31and we could measure internally.
  • 00:24:34So you start to granularly, like, okay, I want to introduce internal links.
  • 00:24:39I have this new algorithm idea, and I want to do that.
  • 00:24:43So you start in the base, and then you start to think, okay, where I want to test this?
  • 00:24:47So you go a level on the subcategory level, and then, okay, I want to test this, but against what?
  • 00:24:54So you have this whole taxonomy on the flat line that you can really say, okay, I'm going to test on the subcategory for some lines.
  • 00:25:01against the subcategory for sports car.
  • 00:25:05So we had this level of data.
  • 00:25:07Now, it's not impossible.
  • 00:25:11Impossible is not a real word for SEO.
  • 00:25:13It's really hard to make a test in one page or ten pages because you don't know when or if Google will ever crawl that page.
  • 00:25:27you can force it by going to Google Search Console and ask Google to index that page or to re-index that page.
  • 00:25:35But you don't know when it's going to come back.
  • 00:25:38So it's really difficult to control very granular tests.
  • 00:25:44That's why usually a test for us at Shutterstock, it wasn't a millions of pages or so.
  • 00:25:53Because then we had a chance in a bucket to Google to actually see the changes.
  • 00:25:58at scale and provided the results.
  • 00:26:01I hope that answers the question as well.
  • 00:26:04Yeah, makes total sense.
  • 00:26:06And if we think about traffic, how can I imagine this based on the distribution of pages?
  • 00:26:15So is there a very, very small amount of pages that bring in like, ten percent of the total traffic, but then ninety percent of the traffic is a She early endless long tail.
  • 00:26:28or is the distribution like more even
  • 00:26:31is more even is more even.
  • 00:26:35I don't know if I can go too much on details here, but the distribution is even which is surprising.
  • 00:26:43and the website of the size because we expect to have like the famous twenty eighty.
  • 00:26:48so twenty percent bring all the revenue in the eighties.
  • 00:26:51just noise but no is is even which is good.
  • 00:26:59that gave us a lot of room for testing and improvements, right?
  • 00:27:03We could test in different parts of the website, we can test in different formats.
  • 00:27:07So yeah, it's not focused in only one.
  • 00:27:11One thing that I would like to add though, which is interesting, has been happening for years and is not only related to Shutterstock, I think all marketplaces have seen this, is the Google shifting.
  • 00:27:24the priority of marketplaces from PDPs, so individual pages to category pages.
  • 00:27:31I think that happens across the board.
  • 00:27:33In all the marketplace that I worked on, Google gives more visibility to category pages because Google itself is becoming a category page depending on what you're looking for.
  • 00:27:46So it is a really interesting move from Google that has been happening for the past few years.
  • 00:27:53And for Shutterstock was kind of the same for Farfetch and for List.
  • 00:27:58Kind of the same, like, category pages are always more visible than PDPs, which for other CEOs that report against revenue is always more difficult on version rates on PDPs or way higher than category pages, of course.
  • 00:28:14But Google don't like SEO too much, so they don't want you to show our products directly to the user.
  • 00:28:19They want to make it harder for them.
  • 00:28:24Got it.
  • 00:28:25And do you think this is also valid for smaller pages?
  • 00:28:29or is the whole shift from PDP to category pages primarily valid for pages of the size of list for Fetch Sutterstock?
  • 00:28:38So if a, for example, smaller e-commerce owner or e-commerce team is listening, so smaller, let's say they might even have like a hundred K pages in their index.
  • 00:28:48Is it also true for them?
  • 00:28:50So should they?
  • 00:28:52focus more on category pages than PDPs?
  • 00:28:56Looking through your data is the best answer that I can give.
  • 00:28:59I know for the marketplaces is a very clear move from Google trying to expose the PDPs in their own SERP enhancements like car cells and top of the SERP, etc.
  • 00:29:16and removing the visibility for the marketplaces because Google actually trying to compete with Amazon, in a sense, like creating a marketplace inside the SERP.
  • 00:29:27So for us, the CEOs at the product level, we had to set up a machine center to be available to go products in the carousel.
  • 00:29:36But when you go looking for specific keywords like Gucci black dresses, you will more often find category pages ranking organically.
  • 00:29:48And the products that are ranking on the server are usually paid.
  • 00:29:53And this more like the ship that we've seen is rarely we would see our organic PDP page ranking for a very broad keyword, like good dresses and so forth.
  • 00:30:05For smaller websites, look into your data.
  • 00:30:10I would say personally that probably category pages have more visibility, but they have worse.
  • 00:30:17PTR or convention rates, which I think is the norm now.
  • 00:30:21And then with the rise of the AI modes and how people are changing the behavior of search, clicks are becoming less and less and more rare.
  • 00:30:31So try to improve your convention rates on the category pages.
  • 00:30:34It's a very smart move.
  • 00:30:36And did you ever engage in editorial SEO at Shutterstock?
  • 00:30:40Because I mean, if you look at Some, for example, also pretty successful B to B SaaS startups, for example, editorial SEO has been a key growth for them over the last years.
  • 00:30:53So some of them are obviously struggling with AI search, but depending on how they structure their playbook, they might even be benefiting from AI search.
  • 00:31:04But so, did Shutterstock ever publish a classic, so to say, blog post?
  • 00:31:09Yeah, yeah.
  • 00:31:11The Sharestocks does have a blog, and it does have an editorial lead that handles the blog and handles special landing pages that we created along my tenure there.
  • 00:31:28I myself, I am not a big fan of editorial SEO, not because I don't think it's useful.
  • 00:31:37I think is really important for the SEO growth, organic growth and winnings editorial SEO.
  • 00:31:45I'm just not very good at it.
  • 00:31:47So I try to let the people that know editorial SEO better than me to take the lead.
  • 00:31:55I can help with the technical side, with the strategy, but the execution of editorial is not something that I did in my career.
  • 00:32:03As you can see for our chat, way more technical.
  • 00:32:09Editorial SEO is not my cup of tea.
  • 00:32:12Okay, got it.
  • 00:32:12So you are personal as Roberto, you are not a fan, but you see the value in the bigger SEO picture for a company.
  • 00:32:23Yes, I think it's crucial to have content that is not optimized for bots.
  • 00:32:30That's the catch.
  • 00:32:32I think with the AI rise in the past few months or years, everyone started to do listicles.
  • 00:32:41So any content that you read right now online is on bullet points.
  • 00:32:45It's just infuriating because you see that they are trying to optimize for bots for LLMs because they want to have more visibility.
  • 00:32:55But we've learned along the years that content should be made for humans.
  • 00:33:01And if they are good enough, they will show to the people that actually need them.
  • 00:33:08But I am a very white-headed CEO.
  • 00:33:11So probably someone that listens to this that is more in the gray area or maybe a bit like issues like, oh, this guy.
  • 00:33:16Oh, no.
  • 00:33:18He doesn't know what he's talking about.
  • 00:33:20And I can assure you that I have my own personal websites that I do the dodgy things to test it.
  • 00:33:28And they didn't work for a couple of months.
  • 00:33:30They didn't die.
  • 00:33:31But when you have so much responsibility in your hands for companies like Farfetch, Shutterstock, now Lottie, you don't have the luxury to try quick wins and lovers.
  • 00:33:45And I think everyone that is listening here, thinking about longevity and long term thinking, often is made for humans.
  • 00:33:58everything that comes after this is a bonus.
  • 00:34:00Of course, you want visibility, you want traffic, you want conversion, but bots don't convert.
  • 00:34:08Actually, bots cost you a lot of money because they need to crawl your website, they need to use your resources to render page.
  • 00:34:16So they actually optimize into spend money, which is contraintuitive, right?
  • 00:34:22So yeah, content is king, as they used to say.
  • 00:34:29but it needs to rule correctly.
  • 00:34:32It's not only listicles to show on chat.
  • 00:34:34DPD is deeper than that.
  • 00:34:37Sure.
  • 00:34:38Let's look a little bit at this intersection of tech and marketplace SEO and editorial SEO because I would like to get your point of view on people trying to enrich.
  • 00:34:55For example a category page with like a additional editorial content and there are obviously multiple ways to do it.
  • 00:35:02there are the classic like ecom category page and you just put like two paragraphs below the product and then you feel like wow.
  • 00:35:13Massive value add.
  • 00:35:15how do you think about that?
  • 00:35:16did you for example and rich.
  • 00:35:20key category pages at Shutterstock with additional editorial content alongside the images or videos or whatever asset was presented there?
  • 00:35:30Yes, there are a couple of tricks that you can do to do that.
  • 00:35:35I have a couple of examples that are really, really interesting.
  • 00:35:38I'm not going to say the name of the companies that I did the test,
  • 00:35:41but
  • 00:35:42in one of them, we used to have content in the top of the page of the category page, SEO content.
  • 00:35:49which was behind the read more button or execution.
  • 00:35:54So I was looking to the category period like, oh, this is a bit rubbish, right?
  • 00:35:59I mean, when you open in a mobile and you click on the page is on block of text, and then you see the product be lower, like, it doesn't make any sense.
  • 00:36:10So I suggested to move the contents to below the fold to see what would happen.
  • 00:36:14I mean, The theory is content at the top of the page is more important to Google.
  • 00:36:19We like more.
  • 00:36:21We moved to below the fold, and we saw like a fifteen percent increase on organic traffic.
  • 00:36:27I'm like, yeah, content is important, but again, you're optimizing for a human.
  • 00:36:35So if you and Nicholas are looking for a specific product, and you go to a page, category page of brand that you like, And you know there's a category page.
  • 00:36:44And the first thing that you see is a block of text.
  • 00:36:47You're not going to read it because you know already that you want that product.
  • 00:36:51So what's the point of the content there and only for SEO purpose?
  • 00:36:55It
  • 00:36:56doesn't need to be there.
  • 00:36:57It can move below the fold because if the user has some questions, you have the content there and help Google to understand more context about it.
  • 00:37:05Now, you don't need to.
  • 00:37:11directly inject contents in a page to render the page.
  • 00:37:16This is not like at SEO guys, but you can leverage the data that you have from the PDPs to load on the category pages.
  • 00:37:27So think about the headers, think about the description of the product, pricing, and so forth that can be loaded.
  • 00:37:36in the category page in the backend rendered to bots or in your schema markup and how you range your schema markup.
  • 00:37:43So this is also content which is not on page, but it's content that help bots and search engines to understand the context of the page.
  • 00:37:55Of course, the traditional create a content in the category page.
  • 00:37:59It works.
  • 00:38:01I did some tests in different companies that did not work.
  • 00:38:04I tried some tests on AI content generated to see if it would bring more results, more index queries.
  • 00:38:16It does bring, but then it lowers too much the impression counter, the CTR.
  • 00:38:21So it's tricky depending on the project, but people need to have in mind the user experience, why you're putting the content there is actually helpful, or is only for SEO, it can be both.
  • 00:38:37Makes sense.
  • 00:38:38We already introduced this term of enterprise SEO and just juggle it around.
  • 00:38:44And I'd like to quickly get your definition, so to say, on where do you draw the line between just SEO or regular SEO and then enterprise SEO.
  • 00:38:59So where does enterprise SEO start and like?
  • 00:39:06What are the different things you would look at and how would that then change?
  • 00:39:12Like how you, for example, prioritize different actions and how you shift gears from maybe optimizing single pages or like badges of tenth of pages to things you already described in great detail with taxonomy thinking, more category-driven thinking, etc.
  • 00:39:28I
  • 00:39:32think it's taking more than a week to crawl a website, you are enterprise.
  • 00:39:37I guess that's the answer, but you will know when you are an enterprise website because you be scared.
  • 00:39:46I don't think there isn't proper number of pages that you can classify.
  • 00:39:52enterprise or not is how the company portrayed themselves as well.
  • 00:39:56In a pure SEO number, I would say at least a couple million of pages is to the point that you cannot yourself basically investigate all pages one by one.
  • 00:40:13But I don't think there is a exact number.
  • 00:40:16or is that exact threshold for enterprise.
  • 00:40:21You can have ten pages and be an enterprise website because I don't know, maybe you have billions of views in the same pages.
  • 00:40:30or you can have billions of pages and be tiny because you don't have visibility on anything.
  • 00:40:36So it's something that if you are scared to join, you know that it's an enterprise.
  • 00:40:44You're scared to look like, okay, how many pages?
  • 00:40:47Ten million pages.
  • 00:40:47Okay.
  • 00:40:48New website and someone replaced you watching the crawler and they're like, oh, yeah, okay.
  • 00:40:55Got it.
  • 00:40:56I am on an enterprise mindset.
  • 00:40:59But yeah, sorry, it's not a direct answer, but is the favorite answer from everyone like it depends.
  • 00:41:09Yeah, so in a way, for example, chatgbt.com is already probably enterprise SEO because they're getting, I don't know if they're already getting billions of clicks, but they are probably getting.
  • 00:41:23two digit to three digit millions of clicks in a month.
  • 00:41:27And then they have like their GPT store and the custom GPTs that are like a couple of probably thousands of pages.
  • 00:41:38But this is probably an example where you have to start thinking enterprise-y, although you don't have like a million of pages yet.
  • 00:41:48Yes.
  • 00:41:49that absolutely correct.
  • 00:41:50And recently OpenAI was hiring for an SEO content lead.
  • 00:41:54I don't know if you saw that, but again, OpenAI is going to kill the SEO job because the chat deputies hire for content SEO.
  • 00:42:04So there is content poetic behind this, right?
  • 00:42:08Yes, this was a perfect transition that was not scripted, but you perfectly placed the ball on to do a penalty here.
  • 00:42:18Because I was interested in how AI, LLM driven search and like everything around that has changed your strategy at Shutterstock, for example, like.
  • 00:42:33especially after the inception of ChatGPT late in the year, just curious, how has this process looked like?
  • 00:42:45Very sensitive topic.
  • 00:42:47I don't want to say anything that is very polemic, I guess.
  • 00:42:54I'm going to give my own personal view, yeah.
  • 00:42:59So I hope that I don't offend anyone.
  • 00:43:01If someone from profound is seeing this, I apologize.
  • 00:43:07Nothing changes, really.
  • 00:43:12LLMs, they are using SEO metrics to classify their content.
  • 00:43:21They are fetching information from Google, from Bing, from any other search engine, to train their data.
  • 00:43:30There are several correlations between how chat-DPT rank companies in a chat compared to assert results.
  • 00:43:43Be successful as of today, sometime may change tomorrow, but as of today, be successful with LLMs also means have a strong SEO program, have a strong SEO foundation.
  • 00:43:57Because SEOs have been doing this for years.
  • 00:44:01How do we optimize for users?
  • 00:44:04How do we optimize for callers?
  • 00:44:06What is more useful?
  • 00:44:10And LLMs are just capitalizing all the years and years of work that we as an SEOs did.
  • 00:44:17And they are just fetching our jobs or our results and showing in a different wrapper, in a different environment, as cloud, chat, GPT, et cetera.
  • 00:44:27So it's really, really difficult for me to pinpoint a unique view or strategy for LLMs as of today, tomorrow.
  • 00:44:36This can change, of course.
  • 00:44:39And I think Lily Ray shared a very good example a couple of weeks ago of the first on her mind, and I agree, first AO-driven strategy.
  • 00:44:55that is actually something which was the integration between OpenAI and Stripe or Shopify.
  • 00:45:02So this is really a place that is unknown in terms of optimizations and is really focused in LLM optimizations, the EO-GEO.
  • 00:45:13So it's a very good example of what this can look like moving forward.
  • 00:45:17If you want to work in this space, you can believe this space is going to grow the way that things is going to grow.
  • 00:45:24This is a very good starting point to understand the nuances of.
  • 00:45:28how will you optimize your Shopify store to show on chat dpt because they have this connection, not this data agreement.
  • 00:45:39Everything else is based on what they scrape from SERPs and how you present your information.
  • 00:45:47So that's why I made a joke about the listicles icon.
  • 00:45:50read them more because ChedDPT loves bullet points.
  • 00:45:55So SEO has started to create all their content and bullet points.
  • 00:46:00As the PT goes there, I go, I have the bullet points already.
  • 00:46:03I don't need to think myself.
  • 00:46:04I'm just going to copy and paste on my answer for you.
  • 00:46:08But yeah, that's my view for now.
  • 00:46:11It might change tomorrow.
  • 00:46:13It might not change.
  • 00:46:16Annoyingly, SEO is baked on data, and we don't have data for those tools.
  • 00:46:24We don't have Google Search Console, Google Analytics for Shared GPT.
  • 00:46:29We don't have a dashboard that show us how many times our brands are showing and the results, how many impressions, how many clicks.
  • 00:46:38They do a very poor job in adding the referrer and the URLs on the citations.
  • 00:46:45But not even that, we have this ability.
  • 00:46:48So it's a trick moments to everyone in the tech world, mostly SEOs, because this L in our labs, like, oh yeah, LLMs, they're talking about search is SEO issue, which it really isn't.
  • 00:47:04It's everyone's issue, but SEOs just took the responsibility by default, which I don't know if I agree with.
  • 00:47:14Okay, very nuanced view here.
  • 00:47:16I'd like to tap into one thing that I think is not enough talked about, because a lot of posts on LinkedIn and publications, blog posts, is about how to be more visible and how to do this and that to become a citation or become mentioned in the AI search.
  • 00:47:39What I'd like to understand more is Do you think that people will search differently?
  • 00:47:46So with the rise of chat GPT and with the rise of AI overviews and then probably the continuous rollout of AI mode, do you think the search behavior itself by people?
  • 00:48:01We can also differentiate, obviously between consumers and like business users or professional users, but do you think people will search differently?
  • 00:48:11I can use myself as an example, and also there are, I think, Rand Fishkin with him, or some of these big dudes shared a very interesting report saying that after ChatGPT was launched or started to be used across the masses, the number of search in Google actually increased and not decreased.
  • 00:48:36And this is true because I myself use ChatGPT for research.
  • 00:48:40So I go there and say, oh, I want to buy a new backpack to commute to London.
  • 00:48:45And then it gives me some examples of some brands.
  • 00:48:48I'm like, yeah, but I don't like this style, et cetera.
  • 00:48:51What do you think?
  • 00:48:52And then again, give me some examples.
  • 00:48:54I get this example, and I go, should I go?
  • 00:48:57Type what I want based on the chat DPP recommendation.
  • 00:49:00And from there, I do my purchase.
  • 00:49:03Chat DPP is an excellent tool for researching and for helping you to understand what you really want.
  • 00:49:11because sometimes you don't know the word of a product, or you don't know exactly what you want, but you know you want something.
  • 00:49:18Whereas if you're trying to do this with Google, nowadays it's very difficult, because everything is very polluted with SEOs and the past few years doing over-optimizations, and it becomes really annoying cobersome.
  • 00:49:34So you either end on Reddit, or you get your idea from ChessDpT.
  • 00:49:41Now, the biggest difference today is that not a lot of people will trust OpenAI or ChatGPT to make a purchase within the platform.
  • 00:49:53So that's why I believe people are researching with ChatGPT, going to Google with a very exact type of word or long tail search and converting from there.
  • 00:50:08And yes, you will change the user behavior.
  • 00:50:12queries will be less generic moving forward.
  • 00:50:17I think having a depth of information in your e-commerce or your marketplace or your website will make a lot of difference because people not knowing are becoming smarter when they are searching for something and they don't realize that.
  • 00:50:36They just think, oh, check the PTT.
  • 00:50:37I'm going to do this and blah, blah, blah.
  • 00:50:39And then when they see they are experts in search and Google, without even realizing it.
  • 00:50:45So I think this behavior will continue to change and I am an example myself on how do I use on my day to day.
  • 00:50:54And do you think that people will like the majority of people will actually complete purchases in chat GPT because I can imagine someone completing a purchase for like Let's say twenty pounds twenty euros twenty dollars whatever, but what about something that is two hundred pounds euros dollars?
  • 00:51:19and what about something that is maybe a vacation?
  • 00:51:22so you know chat to be integrating with booking dot com.
  • 00:51:26you searching for the perfect.
  • 00:51:30Like vacation home in Italy next summer.
  • 00:51:33but then the vacation home is like three hundred euros pounds dollars a day or a night.
  • 00:51:39uh one week vacation we're talking two thousand euros dollars.
  • 00:51:44will people be comfortable doing the decision with in the chat without.
  • 00:51:51Exiting the chat to the page looking at the product themselves clicking through reading checking reviews and checking it like in full detail and not in the like synthesized summarized version that you get in chat gbt which is.
  • 00:52:08which can probably be really good, but I'm just, here I'm playing devil's advocate a little bit towards the OpenAI Stripe, PayPal, Agentech, Commerce Protocol integration, because it's technically really cool, but will the users actually adopt?
  • 00:52:25Yes, and I have a very easy explanation to you why.
  • 00:52:33Do you remember the first time you bought something online?
  • 00:52:38Yes, actually it was on Amazon.
  • 00:52:42It was a DVD series of Two and a Half Man, which back in the days I was like a big fan.
  • 00:52:50I was like fifteen years old and it was crazy for me.
  • 00:52:53Yeah, this was the first time.
  • 00:52:54Yeah.
  • 00:52:55So at the time when we bought our first things online, everyone was saying no one's gonna trust this.
  • 00:53:03They're not gonna deliver.
  • 00:53:04It's a scam.
  • 00:53:05We need to go to the store.
  • 00:53:07We need to try the clothes.
  • 00:53:09We need to touch the product that we are buying this twenty, twenty five years ago.
  • 00:53:16Now we see our parents spending more time in eBay, Amazon or any other marketplace buying shit online.
  • 00:53:24I'm sorry, buying things online.
  • 00:53:28When they say that is actually is a scam is never going to work.
  • 00:53:32So yes.
  • 00:53:34people will buy from these different experiences.
  • 00:53:38It won't be easy and it won't be fast.
  • 00:53:41Like it wasn't with Amazon, like it wasn't with the first e-commerce, or when you bought your first two and a half million different Amazon, you didn't know that all would just actually arrive from whom I'm buying this.
  • 00:53:55You didn't know that.
  • 00:53:56But when you got it, and I remember my experience as well because my first online buy was at for a CD of a metalcore band that was very obscure.
  • 00:54:07I was like, oh, yeah, this is never going to come.
  • 00:54:09But anyway, he came.
  • 00:54:11I was like, oh, wait a second.
  • 00:54:14Can I actually buy things and stay at home and just wait for it to be delivered?
  • 00:54:19So the whole experience shopping experience changes in the past twenty years.
  • 00:54:23So in the next twenty years, it will change even more.
  • 00:54:27So yes, people will buy Unchested DPT on Open High Platformer.
  • 00:54:31cloud or perplexity or whatever.
  • 00:54:33They're going to have their own custom concierge on their phone.
  • 00:54:38They're going to help them to book vacations.
  • 00:54:42This will happen.
  • 00:54:43Now is a matter of time and is a matter of adapting as CEOs or marketeers in general.
  • 00:54:52And how do we approach this?
  • 00:54:55Okay, let's play this a little bit more.
  • 00:54:58I will stay devil's advocate and you will stay like the open AI fanboy.
  • 00:55:04No, just kidding.
  • 00:55:05Just just just just for the sake of this conversation.
  • 00:55:11Why will it work?
  • 00:55:13Because if I get your argument, it's basically every innovation that people are skeptical about it first will eventually be mass adopted.
  • 00:55:23just i'm very oversimplifying of course but how is the let's say instant checkout experience in chat gpt different to the instant checkout experience that meta uh launched in facebook and instagram which they spoiler discontinued this july?
  • 00:55:44so they now i don't know if it's already like taken out of the apps but they basically said We are not going to support this anymore.
  • 00:55:51And then at some point in time, we will just deactivate it.
  • 00:55:55So there is a precedent that this instant checkout idea is not working.
  • 00:55:59Plus, I'm liking this devil's advocate thing too much.
  • 00:56:04Plus, why is there no instant checkout at Google?
  • 00:56:08Because I mean Google obviously owns all the SERP.
  • 00:56:12Like they have already introduced Google Shopping at the top.
  • 00:56:15It's a very prominent thing.
  • 00:56:19would have been pretty easy for them.
  • 00:56:20right just integrating instant checkout people using Google pay whatever.
  • 00:56:24so how is it different.
  • 00:56:27Excellent points for the open AI fanboy work.
  • 00:56:35But you have you probably you know this when chest deputy move from four to five.
  • 00:56:43what happened in the world.
  • 00:56:44Everyone was complaining.
  • 00:56:46Oh, no, you took the personality of my AI and blah, blah, blah.
  • 00:56:52Open AI is playing or something really smart, which is building a bespoke entity that knows about you and communicates with you in a way that you think is your friend.
  • 00:57:07You think is as your friend.
  • 00:57:10So the trust that you build with that interface, It's something really hard to build with Google SERP, or even meta.
  • 00:57:20Because Google SERP, if you do an instant checkout, you actually don't know what you're buying or who you're buying from.
  • 00:57:29But OpenAI play with your ecology in a way that you believe that you're talking to someone.
  • 00:57:36not a bunch of zero and ones, but if someone that actually understands you, that actually know what you want, your preferences, et cetera, but the chances for them to be successful are really high.
  • 00:57:50They can screw it up.
  • 00:57:53Of course, everyone can screw it up, but they build a system or a platform that appeal to the masses as something very unique.
  • 00:58:05Meta and Google, They already burned some bridges with people.
  • 00:58:12People, they naturally don't trust them anymore because of several lawsuits and several issues and the way that Mark Zuckerberg presented himself in the media as well is not very friendly.
  • 00:58:26Open AI, you haven't burned this bridge too much.
  • 00:58:33I can say a few things about Saul Altman, but We leave it behind, but the tool itself still is catered for you.
  • 00:58:46So I think it would be easier for them to provide the service that the user will trust because they believe that there's a person on the other side that knows them as they know themselves.
  • 00:58:57Of course, it's a very psychological game.
  • 00:59:01Us as marketeers, I think we are more skeptical on those things.
  • 00:59:07but we are one percent of the mass.
  • 00:59:09We don't represent the population.
  • 00:59:12We don't know the person that is using chat to the PT in the middle of nowhere looking for a solution for something and how they're going to help them.
  • 00:59:21For us, it's more like, yeah, I'm not doing this, but for the normal masses that are not ingrained in this world is a big opportunity.
  • 00:59:31I don't even know if OpenAI will be the company to do that.
  • 00:59:35it might be someone else.
  • 00:59:37But it will happen.
  • 00:59:38I think we're going to shift again the way that we interact with products and we search and know everyday life.
  • 00:59:47I mean, come on, you can have a subscription for bread today.
  • 00:59:51You know, you can get bread in your home every day.
  • 00:59:55So you don't even need to go to the bakery anymore.
  • 00:59:57So what's stopping us to move more forward?
  • 01:00:03Okay, a great analogy.
  • 01:00:06I will get us out of the AI rabbit hole.
  • 01:00:09now with a final question for this section.
  • 01:00:13If you think about, like, let's say, I know it's hard to predict, but let's say two, three years in the future, will it be basically the same market shares as of today?
  • 01:00:23So Google's still very dominant, and then chat GPTS runner up and then probably perplexity and somewhere there being and the rest is rather for professional use cases.
  • 01:00:34So a Claude, for example, for coders or people that like the creativity, so to say, of Claude and then GROC and the others for like one percent each.
  • 01:00:45Or will it be a Chatchabit Google head to head race in terms of especially consumer adoption?
  • 01:00:52What would be your best guess?
  • 01:00:55I believe Google will remain as a market leader for some years.
  • 01:01:01I think they build a monopoly on the search.
  • 01:01:05Like in or not, it is a monopoly.
  • 01:01:08They have more than ninety percent of the market share.
  • 01:01:12And I don't think OpenAI, chat, GPT, cloud, perplexity will, at least in the next three or four years, change this drastically.
  • 01:01:27I don't even know if this is the goal for these tools.
  • 01:01:29We don't know that.
  • 01:01:30I mean, perplexity, yes, because it's kind of a search engine.
  • 01:01:33They pretend it's a search engine, or a research engine, or whatever you want to call.
  • 01:01:39But Cloud is very focused on coding, assistance, and OpenAI, GenDPT is a more generalist.
  • 01:01:47So I don't even know if they really are trying to take the market share from Google.
  • 01:01:54But I don't think they will in the next three to five years.
  • 01:01:59It might happen.
  • 01:02:01It might, of course, happen because it is how the word moves.
  • 01:02:07But also, we don't know if Google decides to buy them out.
  • 01:02:11Google is such a big corporation.
  • 01:02:13They say, OK.
  • 01:02:15I'm going to buy you out, and I'm going to incorporate you into my tools.
  • 01:02:20The same of perplexity or fraud.
  • 01:02:22We don't know.
  • 01:02:23Even meta, they have enough capital to buy all these companies.
  • 01:02:27So we don't know how it's going to be the next move.
  • 01:02:30I guess when they start to hurt the revenue line of those companies, then we're going to see what's going to happen in reality.
  • 01:02:39But right now, Google, we remain the... market leader, I guess.
  • 01:02:48Interesting point of view.
  • 01:02:50Now, coming out of the AI rabbit hole, as promised, I always like to, so because we are already talking for quite some time, a couple of minutes, but time has flown.
  • 01:03:03I always like to, let's say, wrap up such interviews with very practical pieces of advice.
  • 01:03:14Where people can ideally end the episode and then immediately go implement.
  • 01:03:20and the cool thing is now you.
  • 01:03:22it's a little bit of like a a meter thing here because I Imagine also asking you for advice from like your enterprise SEO history for smaller marketplace sites.
  • 01:03:35and now you are obviously In exactly the setting yourself because you can draw from your experience and implemented at Lottie.
  • 01:03:44So what would be like if someone is in a similar situation to Lottie?
  • 01:03:50So smaller marketplace size, but like building the foundation, etc.
  • 01:03:54What would you say?
  • 01:03:55What would be like your two to three top pieces of advice drawing from your enterprise and tech SEO experience?
  • 01:04:07I think the most important one and people take it for granted.
  • 01:04:11is really understanding the SERPs and what is being shown in the SERPs to your target keywords or target products or whatever.
  • 01:04:23People don't spend enough time looking into the search results.
  • 01:04:29They make assumptions based on crawl data, based on technical details or content depth or whatever.
  • 01:04:40but they don't look at the search results.
  • 01:04:44So go to the search result page for the query that you're trying to rank for and see what Google is showing in that results and try to understand why those pages are being shown.
  • 01:04:58is because of the content depth.
  • 01:05:00Maybe it's not.
  • 01:05:01Maybe the page has no content.
  • 01:05:03It's just a very good fit for that query.
  • 01:05:07Maybe it is user-generated content.
  • 01:05:10Maybe our reviews are a big part of that vertical.
  • 01:05:15I don't think people spend enough time looking into the search results.
  • 01:05:19This is the first thing that I would do and that I'm actually doing is trying to understand what Google is actually showing.
  • 01:05:29Second, don't underestimate technical SEO foundations.
  • 01:05:38I guess this can become a second thought when you're trying to grow company, when you try to scale too fast, and you just bid out pages and verticals and category pages without even thinking on the impacts on the corollability, indexation, or your own website, hour or strength to sustain.
  • 01:06:05the amount of pages.
  • 01:06:07It's like when you go to the gym, if you put ten kilos more on each side, you can't hold it anymore.
  • 01:06:12Same thing with our website.
  • 01:06:14If you had too many pages and you're not strong enough, you're not going to hold it anymore, and you're going to crumble.
  • 01:06:20So don't take technical issue for granted.
  • 01:06:24Don't believe when someone says, oh yeah, yeah, yeah, no, yeah, Google can crawl, yeah, Google can see everything.
  • 01:06:30Really tested.
  • 01:06:33And three, cloud.
  • 01:06:35and LLMs are your friends.
  • 01:06:39I think Claude gave us something that we really wanted for a very long time.
  • 01:06:50that is a way of prototyping faster and is a way of showing to engineers and product managers our idea in an actual prototype because we rely a lot on words and theory which sometimes are hard to translate to certain people that really don't understand SEO.
  • 01:07:14So, prototyping fast with Cloud, lovable or an ever tool you want to use, really important and really accelerates your progress to achieve your goals.
  • 01:07:27So, yeah, there are my three main suggestions here.
  • 01:07:32Awesome.
  • 01:07:34Roberto, this has been a very, very insightful talk.
  • 01:07:39I think we already crossed the hour mark, which is always like a sign for me that we had enough to talk about.
  • 01:07:48And I think we could probably go on for another hour.
  • 01:07:52But yeah, first of all, I want to thank you very much for the very insightful thoughts and the very nuanced thoughts.
  • 01:08:02I'm not quite sure yet if i will continue to call you the open a i fanboy.
  • 01:08:08i probably won't but maybe we should do like an update episode in a year or so to see how things are going then at lotty how things are going in terms of consumer adoption of.
  • 01:08:27I started to end these episodes with a surprise question as i. announced at the beginning, so I didn't spoiler the question, but it has proven to be very interesting.
  • 01:08:41So my final question to you is, what is something that we didn't talk about or that I didn't ask that we should have talked about?
  • 01:08:51Oh, wow.
  • 01:08:55Oh, well, okay.
  • 01:08:58I think one thing that I would like the community to do more and we could talk about more is how we move away from vanity metrics to commercial metrics as an SEOs.
  • 01:09:13I think that's something that, yeah, something that the community should talk more because SEOs, they need to be hold accountable to the commercial results of their jobs as well because it gives them power of negotiation moving forward.
  • 01:09:30I think SEOs are very underpaid.
  • 01:09:33as professionals because SEO is taken from Granted and also because we did a really poor job in sales ourselves as the revenue driven channel.
  • 01:09:48So yes, that's something that I think we could have talked more or touched more and maybe we can talk more in the future.
  • 01:09:56Okay, so Let's let's do one follow up here.
  • 01:10:00if someone says okay Roberto sold like I will go out now stay up or like stand up to my worth and go away vanity metrics.
  • 01:10:12here you go commercial impact.
  • 01:10:15How like?
  • 01:10:17how can I just start getting a grasp on moving from when it tea to commercial.
  • 01:10:26Learn your data.
  • 01:10:29What people are clicking on the SERP?
  • 01:10:31How are you acquiring the users?
  • 01:10:33Are they really converting?
  • 01:10:34Do you know how the attribution model is in your company?
  • 01:10:38Is first click, last click, click?
  • 01:10:41Are you actually losing revenue on your channel?
  • 01:10:44Because you don't know the answer for this.
  • 01:10:47So learn your data and connect the dots.
  • 01:10:49I think we have the Torah of tools.
  • 01:10:54and data available to all of us to actually track revenue impact for any channel.
  • 01:11:02So yeah, don't go to your boss and say, you know what?
  • 01:11:07I want to track revenue because I want to make more money.
  • 01:11:10And then he's going to say, OK, how much money did you make in the past quarter?
  • 01:11:14And I say, I don't know.
  • 01:11:17So learn your data before you have this conversation.
  • 01:11:23Awesome.
  • 01:11:24This was pretty good.
  • 01:11:26You see why I started asking this question.
  • 01:11:30It's always great to hear something unexpected coming up.
  • 01:11:35I think it gives a sense of, I should probably start integrating that people should stay until the end because then we will have a surprise question.
  • 01:11:45So maybe it will bring even more viewers until the end.
  • 01:11:50But, Roberto.
  • 01:11:53Thanks so much.
  • 01:11:54I really value everything you shared with me and obviously also with viewers and listeners.
  • 01:12:03If people liked what they heard from you, what's the best place to follow along?
  • 01:12:09LinkedIn is probably the only social media that I actually used.
  • 01:12:15Everything else is... I don't do it, so LinkedIn is the way to talk to me and to reach me out.
  • 01:12:22Very cool and probably also if there will be any job openings from Lottie like on the marketing or organic growth as the outside people will.
  • 01:12:32Maybe I don't know UK base.
  • 01:12:35Maybe also remotely.
  • 01:12:36They will probably hear first on your profile or on LinkedIn, right?
  • 01:12:41Exactly and we do have some roles open for softening engineers hybrid and remote.
  • 01:12:47Oh Yeah, people should take a look at us And if they have any questions about Lottie, I'm happy to jump in the call and talk about it.
  • 01:12:56Very cool.
  • 01:12:58So then, thanks so much again.
  • 01:13:00I wish you a great success in the new tenure at Lottie and also a lot of success, obviously, for Lottie as a company.
  • 01:13:12And yeah, I can't wait for the feedback to this episode.
  • 01:13:17And thanks so much again.
  • 01:13:20Thank you for having me.
  • 01:13:21This was great.
  • 01:13:23Bye bye.
  • 01:13:24Bye bye.