 My name is Rob Bucci. As you know, or you've heard, I'm the founder and CEO of StatSearch Analytics. I really should have gone pee before I got on stage today. All right, if I'm dancing around the stage, now you know why I'll cross my legs. OK, yeah, Rob Bucci, founder and CEO of StatSearch Analytics. Perhaps you've heard of us. Thank you very much to Will and to Rand. I really appreciate all the kind words. What we do, in case you haven't heard of us, is we help some of the world's coolest brands and agencies measure and report on all those opportunities that are existing in the SERPs. If there's an opportunity there, we're there, we're measuring it, and we're helping our clients take advantage of it. Today, I'm here to share with you all a little bit of a methodology. Specifically, I want to teach you all how to reverse engineer Google's research. I want to give you guys a method that will help you target the right content to the right customers at the right time, hopefully every time. Some housekeeping stuff. You can get me on Twitter at StatRob. Of course, the hashtag is MozCon. You don't really need that, but it's standard for me. OK, so do you guys remember the name of that actor from that sci-fi movie with the short old guy and the weird ears? How often do you get random questions like that from your friends? And what do you do? You turn to Google, right? Because that's where you can get instant answers. But do you remember the dark ages before Google? How did you get the answers you needed? Does anyone in this room remember movie phone? It was like a phone service that you called with your landline, remember landlines? And you twirled your finger through the cord as he pushed buttons on the phone. And eventually, movie phone would spout some information about movies. And you were super stoked on that experience because it was the best experience you had available to you. But nowadays, we can turn to Google and we can get instant answers to all the sorts of weird, super ambiguous questions I might want to ask. And they've changed the world. And more importantly, they've changed human behavior. Google isn't any good at this query. Could you imagine in the 90s someone getting all huffy because I couldn't instantly tell them the name of the bad guy from Bruce Willis' seminal classic Hudson Hawk? It is a seminal classic. I'll fight anyone who says otherwise. Just kidding. No, of course not. You couldn't imagine anyone actually getting upset about that because that's an unreasonable thing to expect. But nowadays, can you imagine someone typing this into Google and getting all huffy because they couldn't get an instant answer? Yeah. Because we look at this and we go, that's not hard. Why can't Google answer that? All the information is right there. You see, Google's a victim of their own success. They're responsible for changing the way people search. And at the same time, they have to respond to those changes in human behavior that they've wrought. And they're spending millions of dollars on both sides of this equation, perhaps billions. On one side, there's this deep technological investment to get people to be able to rely on Google search for everyday use. And that involves them spending money on understanding how to interpret search queries, how to match them to search results, how to serve those search results up instantly. On the other side of the equation, they're spending millions of dollars measuring people's satisfaction with their searching behavior. What did you mean when you searched for that term? Did you find the results you wanted? Was there a better result we could have given you? Did we provide the right blend of various types of rich media on that SERP to guide you to where you needed to be? They're changing the rules. And at the same time, they have to play by the new rules. I like to say it's a little bit like a giant that's drawing a map while it's walking and its footsteps are altering the landscape at the same time. I'm here today to tell you that the SERPs are the front end to Google's massive, multi-billion dollar consumer research machine. They are conducting the world's largest and longest consumer research experiment every single day with all of you. And importantly, if we learn to pay attention to that, we can learn a lot about what Google knows. We can reverse engineer all of their hard earned research. With that information, we can stand on the leading edge with Google and we can learn to, well, we can learn to understand what customers are looking for when they search for something. By paying attention to those SERPs, we can snoop on Google's research. So with that in mind, let's talk about intent. For the purposes of this talk, I focused on consumer search intent. I like to think of there being three key stages that are roughly correlated to stages in an SEO funnel. Those are informational, commercial, and transactional. At the informational stage, well, conversion rates are typically low. We usually see this as the start of the path to purchasing a product or a service. Research has shown us that if you're there and educating people at this point, you can imprint yourself on them and conversion rates or potential loyalty from them goes up as they go further down the funnel. To make this a bit more real, I'm going to give an example. Let's say I decided to get healthy and I decided to get into smoothies because smoothies equals health, right? Sure. These are the types of queries I might use if I decided to get into making smoothies at the informational stage. Next, we have the commercial stage. Now, not everyone chooses to break out a commercial stage. The way we look at it is that the way people search is far too nuanced and complex to be modeled by just two stages. So we have this commercial stage and I think Rand said it best quite a long time ago that they're not directly transactional and they may never result in an exchange of good services or money, but they're not purely informational either. The queries I'd use at this stage are about attributes of those products, looking for characteristics, trying to refine down to exactly what I want to buy. Finally, we have the transactional stage. Many SEOs spend their entire careers just focusing on this point and I would argue out there a detriment. This is the glory stage. These are the terms that in like a last touch attribution world, you can directly tie to conversion behavior and so all that glory causes a lot of people to focus here far too much. The type of queries I would use here would be things that have the word buy, specific brand terms, things like free shipping coupons and so on. The point is at each stage of the funnel, the searcher has a different intent. They have different desires that motivate their searching behavior. They might as well be a different person. Google teaches us that demographics don't really help us understand what we really need to know, what consumers are looking for in an exact moment or where they're looking to find it. They go on and teach us that searcher intent is the new demographic and that intent beats identity and immediacy beats loyalty and what this means is that if you're there where the customer is looking, when they're looking for something, you're likely to win their business even if they have a greater affinity for a competitive brand. This is the world that Google has created and the one which us and them have to live in. A world in which instant gratification of need outweighs almost every other consideration. You can call it the age of human sloth. So, in typical stat fashion, we got interested in this area, we wrote some blog posts around intent and we decided to do some research. Now, what's different is I normally like to rely on all those hundreds of millions of data points that stat has and do internet scale research. This time I'm focusing more on methodology. I want you to pay attention not so much to the data as to the methods behind the data. I'll explain a bit more in a moment. For now it's important that you understand that we pick 14 product categories, all kind of e-commerce terms. The most basic core queries are words like toaster, electric guitar, blender, microwave, laptop, et cetera. We then modified those queries through each stage of the funnel. So, in appliances, the term blender and information will became blender reviews and then in transactional it became by blender. Now, I'm gonna riff a little bit on what Will said because he's 100% right. Keyword research is way more art than science. You actually have to speak to people. You cannot shortcut this work. For the purposes of our research, we kept this very simple. We wanted to make this easily understood by a broad audience. We wanted to scale it up. So, we use these relatively formulaic transmutations or transformations of keywords as they go down the funnel. The truth is people don't really search like that. They search a bit more like my blender example, my smoothing example. So, you have to go out there. You have to talk to people. You have to research. You have to watch how they search for your product or service. Use that information to inform the moments when you actually start to turn to keyword research tools. Key thing I wanna emphasize here is that this whole methodology is dependent on getting this right. It's garbage in, garbage out. And if you shortcut or you don't put the effort in at this stage, all of your findings will be called into question. Now, if you follow Google's research, the think with Google research, you've undoubtedly seen this metric. They call it share of intent. And in typical Google fashion, they're very opaque about it. Basically, it's a way to measure the visibility of a brand at a particular moment in a customer's journey to where it's purchasing a product or a service. We look at this and they go, well, that's just share of voice. So, we create a very simple share of voice score. Again, in reality, you might choose to use a click-through rate model. And you might choose to use another sort of weighting system that is more relevant to your industry. We kept this very, very simple because this is a broad piece of research that focuses more on methodology than on actual data results. So we have a weight. The weight is correlated to a ranking position that's multiplied by search volume. And thus we get a share of voice score. That share of voice score is basically a way of measuring how visible a site is, how visible a ranking is, how much potential traffic it might send, perhaps. Usually share of voice is represented as a percentage. We kept it as a whole number and it's very simple. The higher the number, the better it is. I just don't want you to forget. The data from our research is specific to our keyword set. Your results will differ. So if you start tweeting photos of the pie charts and all that, you're actually misleading people because these are not universal facts. This is all about the methodology. So, focus on those methods rather than the data. Okay, so for each stage of intent, we map the SERP. And this is something that we do day in, day out. We parse out result types. We look for featured snippets. We look for images. We look for news results. We look for places packs. We look for people also ask. We look for video. And we look for shopping results. And as we went through them all, we asked ourselves three key questions. One, how can we unlock Google's insights into what searchers are looking for at each stage? Two, what points of leverage do we have in each stage to make sure that we're visible and importantly, that we're useful? And three, how can we reverse engineer the winning strategies of those key competitors at each stage? Here's a 10,000-foot view of what we found. Boom, 90% of the results on the SERP, regardless of the stage we're at, are organic results. Google tells us that you have to be there, you have to be useful, and you have to be quick. And what they're saying is that the organic algorithm goes 90% of the way to providing results that are there, that are useful, and that are quick. That's a pretty amazing thing. That algorithm goes 90% of the way of lifting all the weight and all the utility that people find from Google. But today I wanna focus on that 10%, not that 90%. So we're gonna zoom into that 10%, and here's what we see. Whoa, 60% of the remaining non-organic results on the SERPs are shopping boxes. I think this is insane. The European Union agrees with me. Whatever, I digress. The point is, these are e-commerce keywords. You have to expect that you're gonna see shopping boxes there, ignore this data at your own peril. If this is your space and you wanna play with it and you wanna win, you definitely need a paid strategy. Now, I'm gonna come back to shopping. That's an important feature of all these SERPs. But for now, let's zoom into the other non-shopping. I call these our points of leverage. I call them our points of leverage because these are the universal result types that are going to allow us to go that extra mile towards meeting search or intent. On the basis of a strong organic ranking strategy, these are the result types that are going to allow us to go that extra step further to helping our customers, giving them truly useful experiences, cementing the utility of our brand in their mind, ushering them towards their destination and generally providing all sorts of awesome signals of engagement for Google. To start with, we're gonna look at the informational stage. We're gonna work our way through three stages and then we're gonna wrap it up. I call this the topology of the informational SERP. This is basically a count of the non-organic result types we saw at all those keywords in the informational stage. This allows us to understand what kind of moments our searchers are having. When they search for terms in the informational stage, what are they experiencing? What are they seeing beyond the organic results? A lot of shopping, images, news, video, people also ask, answers boxes, also called featured snippets, places. So 42% of our leverage in the informational stage are the shopping results. Here's what Google knows. When people use these super short tail core product terms, there's a high likelihood that they just wanna buy it, like right there, right now, they don't give a darn about the model or the size, they just want it fast and they want it cheap. Google knows, immediacy beats loyalty and what they're doing with their shopping boxes, really, when it all comes down to the end of the day, is they're auctioning off immediacy to the highest bidder. It's at the very top of the SERP. 21% of our leverage are image boxes. Now, it's hard to say who wins when it comes to image boxes on the SERP. We have this paradigm of ranking things from top to bottom, images are in this strip. Now, I would actually say this is one of those rare moments where it doesn't matter if you actually rank on the main SERP itself, what matters is that you rank in the image search results because someone's gonna click on one of those that they wanna see images and they're gonna scroll through those image results. Those image results are actually a really wonderful user experience for broad and unrefined queries. They allow us to get to what we're looking for without expending the necessary energy of using human language. We can just look with our eyes. As a sidebar, you guys, there are an infinite number of cat litter scoopers available on the internet. They come in all variety of shapes and sizes. I don't have a cat because my husband is mean, but if I did have a cat, his name would be Samages and I would love him forever. You can tweet my husband, it's at Mr. Underscore DAC and you can tell him I deserve a cat. I would really appreciate that, but digress. The point is, if I did have a cat, Samages, my lovely Tabby, I would need a cat litter scooper and I would turn to Google image search and I would type in cat litter scoopers. I would scroll down the results until I found the one I was looking for. It's the one that's like shaped like a rainbow, doubles as a slingshot and I'm like good to go, right? So that's a really good experience for me when I'm looking for something that's super broad and there's all sorts of various like iterations and versions of it. And if someone does click on one of your images because it's enticing to them, you actually have a pretty good opportunity to convert them. You have a lot to work with here. You have the image, of course, that's what drew them in. You have a great title that you can make actionable, you have a description there and you have a link for them to visit your website. Okay, news boxes. 18% of our potential leverage in the informational stage. This chart shows the average search volume per keyword per result type in this informational stage. Basically this says when a result type is triggered, what is the average search volume that that keyword has for that result type? And news boxes like massive, right? So I look at that and that's very, very enticing. I want to evaluate the potential going after them. The first thing I want to do is I want to say, who's winning? And here's what I see. 102 media sites across 128 separate news results. The dispersion is super broad, like everyone and their dog has a news result. These five domains are the only domains that have more than one news result. And so like really, you look at this and there's like not really a true incumbent here. Everyone, the vast majority only has one result. The next thing I want to know is on average, who's sending the most traffic per news result, i.e. who's winning the most valuable news results? And here's what I see. Interestingly, these five domains are not the same domains that were on the previous slide, meaning that all of these sites that are sending all of the search volume are doing so on the basis of just one news box ranking. How can you build a strategy around this? You can't. Here's the thing. News results suffer from an irrelevancy problem. I'll give you an example. I love chocolate pudding, you guys. I just like, I can't get enough. And let's say I wanted to go and I wanted to buy some chocolate pudding and I typed chocolate pudding into Google and what I would see was a news box for Meryl Streep, spotted eating chocolate pudding at a fashion show. The fashion world is aghast, like something like that, which is completely irrelevant to my actual query. That's because Google does this on purpose. You see, what Google knows is that when someone searches for a core product, a place, a person, there's a possibility that that product place or person is the subject of a news event. And what we all know is that Google's relevancy algorithm largely is based around links and citations and you can't accrue links and citations at the same pace as the news cycle. How do you deal with that problem? Very simple. You exact match the query with the headline of the news article, thus covering off any possibility that the person is looking for a news event. Here's an example of that irrelevancy problem in action. The term is jacket. It might seem like a really good core product term maybe, but it's just a broad news box. I can't use that at all. Here's another one for desk. I snuck a little sexy Canadian Prime Minister in for you. You're welcome. But here's a great one, best laptop. Look at that. Look at the visual real estate that has. It takes up most of the screen. It's a great consumer term. It's probably directly related to what someone who's like a world traveler is looking for. It's also completely ephemeral. That result will disappear as soon as it ages out of the news cycle. As a side, this is at least partly how Google is responsible for the erosion of quality journalism over the years. News orgs are so desperate for that traffic because it drives the advertising revenue that keeps them alive that they're willing to constantly regurgitate the same article over and over and over just to chase that algorithm quirk. Let's add another example of how Google's changing the world. My tip to you is this. Don't go running after news boxes just because they rank on high traffic serves. If you're looking at the data the way we are, you will consistently see news boxes appear for massive search volume queries. And you're like, wow, that's super enticing. Unless you're Forbes or Gizmodo or some media company, there's no point. Just ignore them to death. Okay, 8% of our leverage is video results. I love this. Video results are a great way for you to get like a little bit of extra like real estate and visual oomph on the SERP. But also, look at this. Bed, Bath and Beyond. It's a branded term. That goes directly to a product landing page. I bet you that performs very, very well for them. And then you have sometimes these smaller video thumbnails when the video is a primary piece of media on the site. You can get these video thumbnails that will appear. So there's some good opportunities to get some great extra real estate on the SERP. The first thing I wanna know is, is there like a big winner? You guys, I have a secret for you. Let's all keep it in this room so we don't ruin it. Google has a bias towards YouTube. Now you know. So we look at this and we're like, okay, well these are e-commerce terms. I don't really wanna send my traffic to YouTube, right? Right, here's a mantra for you to walk away with. Be the Bed, Bath and Beyond. Because that's what you wanna be. You wanna be that site that ranks that large video thumbnail directly to your product page. And if you can do that, that's awesome. And if you look at the data and Google saying, hey, people when they search for these queries, they want to interact with video. And you look at your site and you go, I have video assets, I can take advantage of this. That's a great opportunity and you should run with it. Featured snippets. Everyone loves featured snippets. I love featured snippets. I'm tired of saying the word featured snippets. They're probably one of the most blunt examples of Google conditioning and modifying the SERP to like directly meet consumer or search or need. And they're a great opportunity for SEOs. Many people have been driving fantastic results with them. So I definitely wanna take a look at the potential of going after them. The first thing I wanna look at is what is the most common featured snippet format? I look at this and I go, okay, great. I don't have to make major site changes. I don't have to put tables or worry about making, or changing my content around because it's mostly paragraphs. And all I have to do is make some small optimizations to take advantage of that. The next thing I wanna know is how valuable are these keywords that the featured snippets are appearing against? 23,300 is the average search volume. So I know that these featured snippets are appearing against very high value terms. The next thing I wanna know is who's the incumbent? Oh, it's Wikipedia. Now a casual inspection informs that the Wikipedia featured snippets are mostly around like broad product knowledge, like who made the product, what's the history of it, why does it exist, how do I use it, et cetera. So you have to ask yourself, I'm really enticed by this opportunity. 23,300 average search volume, featured snippets at the top of the SERP, do I wanna fight with Wikipedia? Is it a good use of my time? Probably not, right? Right, there are greener pastures. Let's sum up the stage quickly. What have we learned? First of all, we need to recognize when we need a paid strategy to win. You guys can't be polemicists about SEO, you cannot sit in an ivory tower and ignore the fact that the landscape here dictates that if you want a piece of this action, you're gonna have to pay to play. That's just the fact. So you need to break down the walls that you might have between you and your paid team and work with them on this one. You can't ignore those shopping boxes. Lesson two, we need to look towards strong bias for certain sites. These incumbents like Wikipedia, they're currents that we have to swim against and you really wanna not be swimming against the current if possible. And when you see an incumbent like Wikipedia, it's pretty hard to compete with them when it comes to featured snippets. So if they're dominant in your snippet space, maybe you wanna look for greener pastures. Lesson three, looking at those SERP topologies tells us what Google thinks these searchers want. This is not a mistake, it's not random. The fact that there's a certain percentage of video results to featured snippets, to images, to answers boxes is not at all an accident. It is the result of years of Google's fine tuning the SERPs and their research into engagement patterns of people. So if you look at that data, you can understand what kind of moments those searchers are having, that Google thinks are best for those searchers and you can use that information to inform what you're trying to do. Lesson four, news boxes suck. More importantly though, you can't just blindly chase search volume. People get like really excited about big numbers and they chase after it without really thinking twice about it. It's a waste of time. Always critically interrogate the data, go a level deeper to see if it's something that you can actually take advantage of. All right, commercial intent stage. Here's a topology. Whoa, shopping boxes, right? We also have images. We have videos. We have answers boxes. Those other wedges are too small. The news box is about 2% of the leverage here. I didn't put the number on, but now you know. Last time we went through every single result type and that was quite, it took a long time. I spent quite a few minutes doing that and I wanna do it a little bit quicker. I wanna show you a method that's gonna help you get to the finish line faster. We're gonna do this with share voice. Share voice is a great way for us to measure the visibility of a result. So here is the average share voice contribution per result type. Remember, Nord News, it's like the siren song. I look at it and I'm like, oh yeah, look at that. No, you can't have that. But maybe answers boxes. So that gives us some ideas and maybe where we wanna start. But before I go through every single result type, why don't we start at the end and work our way back? And what is the end? Well, the end is who's winning. So the way we're gonna do that is we're gonna look at the domains that have the highest aggregate share of voice in this segment. Consumer Reports and TechRator are almost neck and neck. Then you have Amazon third, CNET and PCMag fourth and fifth. The next thing I wanna know is these are the winners but who has the most real estate? I.e. who has the most rankings? That's weird, right? Amazon's in third place but they have 3,117 rankings on these SERPs compared to the first place site which only has 299. Instantly I'm like, what on earth is going on? The technique we use to understand and figure out what's happening here is to attribute the share of voice back to the result types that are driving that share of voice for each one of those winners. And that's this chart. So first of all, I can point out Consumer Reports. Super badass. Driving a massive amount of share of voice in first place primarily on the basis of 299 mostly organic rankings. That means they're ranking high for super high value keywords and that's a very strong place for them to be. Then you have TechRadar in second place. TechRadar is also badass for a different reason. They're driving a ton of share of voice, more than 50% just on the basis of featured snippets. Both of those have like sub 300 rankings. Then you have Amazon 3,117 rankings in third place like the giant in the room just swinging its club winning on the basis of volume alone. But you have to know that a lot of those rankings are not very high value. A lot of them are quite poor in fact and that's why they're in third place. You have CNET who's bringing in some games with video. You have PC Night that has a blend and you know what you see almost none of? It's news boxes. So I should just call this talk news boxes suck, but anyway. Okay, so moving on. I wanna show you all how you can segment my query modifier to see which result types Google likes. The queries that people use are such a strong indicator of the mental state of the desire of the mode that the searcher is in, right? So if we're able to pull out certain key terms like buy, review, compare, sell, et cetera, we can really begin to get this fine grained knowledge of how Google is conditioning the SERPs to meet very minute moments of intent. Here's the topology when the word best is used in a query at this commercial stage. If your percent is shopping, but answers boxes. Remember what Will said? When people use the word best and they see featured snippets, what they want is some guidance from like a trust of third party to help set them off on their journey, to help them find what is actually best. This is a good way for them to jumpstart what they're looking for. So Google has given people featured snippets to get them to that destination really quick and by simply segmenting on the word best, we're able to tell really quickly, hey, you know what, if we're optimizing for queries that have the word best in it, we better consider some featured snippets. I do not advocate running headlong towards optimizing for featured snippets. I advocate taking a look at the data, digging deep into your businesses, strengths, your weaknesses, and your opportunities and deciding if this is something you can take advantage of. Look at this, compare. When the word compares is used, 42% of the non-organic leverages videos. Google knows that when people use the word compare, they wanna see videos of people comparing products. New. When people use the word new, they wanna see images. Remember, I'm gonna point it out again, this is not random, this is on the basis of very careful meticulous research and feedback loops that Google's built into the SERP. Reviews, we see a really great blend of both videos and featured snippets. Just don't forget, you have to go deeper than the surface data, you have to dig into that share of voice. So what have we learned about the SERPs for our commercial keywords? One, we've learned that by using share of voice, we can measure which domains are winning. That's how we get to that very, the finish line. Who's winning in the space? We aggregate up the share of voice. And we can see that even though, TechRadar has like 3,000 fewer rankings, they're still beating Amazon on the basis of share of voice, which is the leading indicator of their visibility in this space. Lesson two, we can perform share of voice analysis on result types to assess their potential. Thus, we can reverse engineer exactly how TechRadar was winning. It's on the basis to feature snippets. Lesson three, we can segment even deeper on those query modifiers, words like best, compare, reviews, buy, et cetera, to see how the SERPs adapt. Thus, we know that when people use the word best, the searchers want snippets. When we use the word compare, the searchers want video. Final stage, transactional intense stage. Here's a typology. It's like the birth of Pac-Man. It's getting bigger and bigger and bigger and bigger. The SERPs are getting more and more distilled. Now, I want to use the exact same method as I did in the last stage because it's very efficient. We're going to start at the finish line and we're going to work our way back. So here's who's winning. Based on aggregate share of voice, Amazon, Walmart, Overstock, Best Buy, Sears Outlet. Let's attribute those back to the result types. Whoa, it's all organic results guys. At this point, everyone is winning on the basis of organic results. No one is pulling in their wins because it featured snippets when it comes to the transactional terms. Here's the thing. The further you go down the funnel towards those glory keywords, those high converting keywords, those moments before purchase keywords, the more distilled the SERP becomes. The more Google wants to funnel people towards their paid product and the more Google wants people to rely on the organic rankings. The things that get people to their destination really, really quickly so that they can buy something immediately because we need to be there, we need to be useful, we need to be quick. And the organic algorithm goes 90% of the way of doing that. This is basically good old fashioned SEO and it's probably pretty gratifying for all everyone who's like, you know, get off my lawn when it comes to featured snippets. It's like it all comes back to good old fashioned SEO. And of course, PPC because you got to play that shopping game. So at the risk of sounding reductive, the advice here is really simple. Are you ranking high? Are you ranking the right content? How do you tell if you're ranking the right content? Well, one of the things we do is we scaffold out all the URLs that we see for a particular site ranking. We then look for the patterns in the URLs. So if we see a URL that's like sat slash C slash and we go to that page and we see it's a category page, we're gonna wild card tag every URL as category. And then if we see slash P slash and we go to that page and we see those are all product pages, we're gonna wild card tag all those as product pages. That feeds into graphs like this. Which UX is ranking the most? Which UX is the most visible? Category pages. This is borne out across all of the winners. Category pages. Category pages. You see, what Google is doing is the rewarding the best user experience that displays the highest engagement. So people like really labor these decisions around, oh my God, like what do we choose to rank here? Like what sort of user experience do we wanna give people? You don't have to reinvent the wheel. You don't have to really have a ton of conversation about it. You have to look at what's already winning. Because Google has decided to reward those things because they're helping people. And you can use that information to decide how you're going to help people. So what have we learned about the SERPs for our transactional keywords? One, we've learned that the deeper we go in the funnel the more we have to rely on organic rankings. Those organic results are the best way to push traffic in this stage. Lesson two. We've learned that analyzing ranking URLs to find out what UX, sorry, I'm gonna repeat that. We've learned that we can analyze ranking URLs to find out what UX satisfies the searcher's intent most. The majority of those pages that we saw ranking were category pages. We can just move in that direction. We don't have to swim against the current and try to like shake things up. We know that's working. We know people like it. We know Google's rewarding it. Let's just do that. Okay, so at the end of the day, what conclusions have we reached? Remember Google's advice. Be there, be useful, be quick. How do we be there? First of all, it starts with quality keyword research and thorough segmentation. You cannot shortcut this. Take a page out of Will's book. Talk to people, interview them. Ask them what they're looking for. Ask them why they search for certain things. Ask them why they clicked on that result. Get in the mind of the searcher. You have to get out of your office and onto the front lines in order to do this properly and you cannot shortcut this. This is crucial and underpins everything you're gonna do in this methodology. How do you know if you're there? Well, you know you're there by measuring share of voice. It's a great proxy for general visibility. How do you know if you're useful? Well, you can look at those topologies and you can find out what Google thinks is useful to searchers on the basis of their research. Take a look at this and tell me that if this was your business and this was your data, you wouldn't look at this and go, whoa, people want video. Google has learned that. I'm gonna piggyback on Google's learning and I'm gonna see if that's an opportunity I can use for my business. How else can you be useful? You can create the right types of UXs by looking at what's already winning. How do you be quick? Well, that's a little bit outside the preview of this talk, but there's lots of great content out there on site speed optimization. Basically brevity, omit needless words, get right to the point. Now, many of you might go, hey, Rob, you know, really like that's great, but 90% of the results are organic and at the end of the day, it's organic that's really winning when it comes to those converting keywords. So I'm just gonna focus on that. The truth is, is across each stage, those non-organic results make up 15% 11% and 10%. And then you look at this, that birth of Pac-Man. You might go, it's just not worth it, right? I mean, Google's squeezing us out. And if I just focus on the organic, that's my comfort zone and I can stay there and I can probably win. Yes, maybe, but I wanna point out this stat. This chart shows which result types, on average, deliver the highest visibility, the highest share of voice. Shopping's at the top, because Google's perverse. Shopping's at the top because Google put shopping box at the top. Organic is under 2,000 average share of voice contribution. Everything else, those answers boxes, those images, those videos, on average deliver 3,000 share of voice contribution. So you definitely need a strong foundational, organic SEO strategy. You can't get away with that. That's the basic. That's like the minimum thing you need to establish in order to win in any space. But by optimizing for those various result types, the ones that correlate to a strength within your business where there's an opportunity for you to take advantage of it, you're going to be able to pull away from the competition. Go further towards meeting the needs of the searchers. So here's some key takeaways for you. Optimizing for universal results is the best way to ensure that you're satisfying searcher intent. We can snoop on Google's findings on search intent by looking at those SERP topologies. As we move down the intent funnel, different result types rise and fall in value, and that's really important. People have these cookie cutter one size fits all approaches to this. But the truth is, answers boxes aren't that useful in the informational stage. They're not there at all in the transactional stage, but in the commercial stage, they work quite well. With that knowledge of intent, we can start crafting the right content for the right moment of the searcher's journey. This analysis is actually very easy to do. Here's what you need. You need a SERP tracking tool that can break out result types. You need some search volume data. You need a share voice model if you're choosing, and you can do it in Excel or SQL. I did all of this in Excel or SQL just to prove to myself that it is possible, and in fact, it's quite easy to do. There are tools that you can use that will help get you a little bit further along. Moz has this wonderful SERP features feature. SERP features functionality. Serp feature thing. Anyway, the point is, they do this great job of aggregating up the different appearances of result types across the SERPs, and it really helps you get to the finish line sooner. Stat does this as well. Both of us have information out there that you can look at in the partner hub where you can come talk to people, and they'll show you how you can do this sort of research with their own tool sets. If you're interested to learn more, there's a whole white paper on this topic. Getstat.com slash MozCon. There's a lot of information around like people also ask boxes related searches, things that I just couldn't fit into the time today. So go ahead and download that if you're interested to learn more, and thank you very much for your time. Hope you have a great day. Thank you.