 All right. So today we're going to talk about big data. I call it bigger data because it's not really big data. I'm not like a data scientist or anything. But what I've realized is the tools that I was using to try to help my clients were failing. And when you go to big data or bigger data, you need new tools. And the question is, do you have the right tool? Let me give you an example. Where can I put this water? There. Not good. Not flat. Guess I'm not drinking. All right. So let me give you an example. Moz, clap it up for Moz. They let me use their data for this presentation. So that's some transparency right there. So Moz gave me 55,000 of their keywords, or not 55,000. I went into their paid search account. And I extracted 55,000 keywords. I'm going to make up the time frame and just say it's a year so that you don't have all their data. So then what I did is I wanted to pull the top 20 for each one of those keywords. So as you can imagine, I'm already at 1.1 million rows. And then the data that I'm bringing in has all these columns. Whoa. Hot clicker. Let's go back one more. So let me show you what it's like when you're trying to work on a 1.1 million row file in Excel on a computer with an i7 processor, 16 gigs of RAM and an SSD drive, which many of you might not have. All I wanted to do was take the URLs out of stat and turn them into domains. Look at how long it's taken. Still going. All I wanted to do is clean a URL list. That was it. Still going. Look at the number at the top left. See how it's just scrolling through like, I'm dying. Six minutes, 34 seconds to go from a full URL to a clean domain. So now let me show you what happens when you use Power BI for that, starting at the same spot. I go to clean it. I make a new column. I start typing in what I want. You can use machine learning to figure out that I want domains. So I don't have to put in any formulas. Then I'm done. I'm saving it. It's updating the file right now. I'm going to get to one minute and three seconds, eight seconds, and it's going to be done. Now I'm already producing visuals for my client to show them how they stack up against their competitor. In this instance, my client is Moz. And one minute and 34 seconds, I'm now looking at 70,000 unique competitors. 70,000 unique competitors in a minute and 32 seconds because I changed the tool. If you go into Excel and try to pull the files, or you go into AdWords and try to pull the files so they're smaller, I'm going to crush you. If you send me a file that doesn't have three digits in it and the file size, I'm going to crush you because you're using the wrong tool. So instead of shifting to a new tool and doing better work and leveling yourself up, you try to cut the data down like, oh, I only want to see things that have at least 10 impressions because you don't know how to function with the data. Let me remind you of this. Excel won't even look at more than 1.02 million rows. So the minute you go to get 1.1 million, they just stop. And then you scroll alphabetically and realize that all of a sudden, a bunch of your words are gone. Now, that's just to do that. All I wanted to do was change your URL into a domain. God forbid if I wanted to marry that data to my Google Search console data, my ranking data, my hot star data, my CRM, or the freaking Census for client out loud, you want to do a VLOOK up on the Census in Excel. Good luck with that. So I have a goal today. It's one goal. I want to show you what's possible. I think this timer is a little short, but whatever. I'm going to try to see what I can do. I'm going to show you what's possible. Don't take any fucking notes today. Keep your eyes open so you can see what's possible. And I'm not going to leave you guys hanging. Usually I come up and I try to show people stuff they can actually do. That's what YouTube is for. So this is the only thing you should take a picture of or write down and then pencils down, hands up, or whatever they say. This URL will take you to our site where we have the homework you were supposed to do. If you didn't get it, we ain't helping you today. You do the homework, it'll show you all the videos of how to do the basics to get started. The first thing you've got to be willing to do is to admit your old way with shit. If you're not willing to admit that, you're going to fight the data. Like, oh, it's directional. How can you be sure? I'm like, okay, well you keep looking at 100 rows when I'm looking at 1.2 million and see who figures it out. It's actually embarrassing. When I look at old deliverables just from six months ago before Sear made this massive change to using Power BI, it was embarrassing to look at the way that I used to service our clients in certain ways. I just can't even, it's like a disservice to them to do it. And I got a shout out, Katie P. Katie Pulaski had been at Sear for six months. She had just graduated from St. Joe's. And she was seeing that I was struggling trying to do some categorizations inside of Excel because that's the only tool that I knew. And she said to me, she said, hey, Will, I use Tableau. Here's the tool that I use, yada, yada, yada, yada, yada. And she started teaching me some of the stuff in Tableau. Now I found Tableau to be kind of hard, so I switched to Power BI. And it costs money. Power BI is free. And if you're on a Mac, just use Turbo.net and use Power BI that way. The other person that I work with that I have to shout out because these are the people that are the reason why I'm on stage. This shit doesn't just come to me. It's being surrounded by some smart people who push me to be better. So Ethan's like, dude, do you wanna do this keyword research project? I'm like, sure, I got no problem doing it. And I'm doing it. And he's like, I'm gonna show you how bad you are at keyword research. And he's like, you know what? And most people are probably afraid to say that to their boss. Thank God, hopefully most of my team isn't because Ethan is the reason why I'm here today because he ran an analysis with a bunch of big data on keyword research and destroyed me. Destroyed me. You know, I'm going in, I'm not hitting enter. I'm thinking I'm all smart because you guys clapped me up and shit and you guys submit these great reviews. So I think I'm the shit. Ethan destroyed me with big data. And what that showed me is, is if you can get bigger sets of data, you can beat people out at stuff that they just don't see. I don't want you to put the pressure on yourselves to think that you're gonna do this stuff tomorrow because it's not gonna happen. It's been a year in the making for me. Look at my first tweet that mentions Power BI. It's a year in the making, this presentation. And I've been practicing it in some other little small venues to try to get it right for you guys. So let's break through the limitations that you believe have been in front of you all the time. First one is keyword research. Like now it doesn't matter if I'm evaluating 5,000 words or 15 million, which is the most I've done so far. I use the same process. I get my clients focused on revenue and my team focused on revenue instead of search volume. I hate search volume. You're about to see why. And it's gonna make sense. The other thing it allows me to do is get my client focused. So the old way, I would ask one of you, if you were one of my clients prior to six months ago, what keywords do you wanna rank for? You give them to me and I drop them into some tools and shit. The new way is probably gonna take you about a month. So if you can get here in a month, you're in good shape. I don't ask them that anymore. I pull their paid data and I say, let me show you what's already making you money, but it's not ranking in the ideal spot. Let's show you what that looks like for us. So you look at your axis, on my left axis, my Y, I got ranked. So the lower it is, the better. On my X, it's conversions. The further out it is, the better. You can see that I'm looking at 55, 54,000 keywords, 9,000 conversions, $632,000 in paid search spend. You see that I can slide rank as I want. So if the client says, hey, we wanna look at what's striking distance. I just slide it over and we have a new conversation. Stop doing PowerPoints. PowerPoints require you to go, oh, here's my slide. I took my picture of it. And then when the client asks you a question, you gotta write that shit down. Look, I'll get back to you. My client asked me, well, how would the same thing look if it was 11 through? Because you can't do this in Excel because if you go to filter all these rows in Excel, it's gonna choke and you're gonna be sitting there waiting and making jokes with your client about how bad Excel is and how bad your laptop is. When I'm getting insights, I can then sort by conversions. These are all options. I can say do I want brand or unbrand and I created some categories that I can just click on. Those categories are so easy to create. You don't even have to know formulas. You just tell it what you want and it starts categorizing all your content. So you're probably wondering, what are we gonna see, well? Well, let's dig in. I'll see Rio in a couple hours. All right, so watch what I'm doing. I'm clicking on unbranded, notice my numbers change and I put my mouse over all these bubbles. Each one represents a word. But what I'm looking for is that one up there. What's that one? Oh, SEO checker. Hey, Sarah Bird, you spent that much money, six grand. Got 41 conversions on that word, but we ranked 41. See, I think how am I gonna talk to an executive? All right, so if all the, I just, think about how this sounds. I looked at 55,000 words for you and I found the one word that you get the most conversions on where you don't rank in the top 10. Might you want me to go after that word organically? There's money in that green little bubble there. Okay. So the word was SEO checker. So notice that in my filter here, I'm looking at the word check. I'm looking at the word check because I wanna look at how many other words there are. I'm not gonna be like, oh, let's go up to the word SEO checker and build it. No, no, no. Just type in the word check. That's five keystrokes and enter. And then you see there's 4,900 different words that just include the word check. I can also see how many conversions they are. So when I go to talk to Sarah, I can say, hey Sarah, does that seem like a worthwhile number of conversions to go after for you? Hopefully she says yes. And then I can tell her how much she's spending on these words and paid. So now I'm grounding her in, well, you spent this much on paid. You're willing to spend this much money to get in front of these people for this word. She's like, yeah, you're right, Will, because all these are facts. No opinion yet. But 5,000 keywords is a lot to process. That's a lot to process, right? So I decided to just put on a filter for at least one conversion. Once I get to all the keywords that have one conversion, it's 214 words. I can manage that shit. I can look at the cost and hear all my bubbles. Everything above that line is my content strategy. Let me go get it. Why? Because I know it converts. See, I got my SEO and PPC data sitting in the same place. You don't. That's why I can do this in two clicks. And then I look at all the words and then I'm sorted by conversions. So every word that I show Sarah, I know I'm going in an order that I can say, this word does drive you real business. And I start looking, all I'm looking for is the teens. 18, that one, 15, that one, 51, that one, 15, 15, 18, 34. Let's go get it, because I know it converts for you. It's a good place to be when you can say to somebody, this word converts for you and you're willing to spend money on it, you're still spending money on it, let's go get it. And the great thing that I love about it is it focuses my client in theory. Sarah Byrd could print this thing out and say, all right, I could say print this out, Sarah, put this up in your cube. If I come back to you in three months and then bubbles don't freaking drop, somewhere we didn't execute or somewhere a serious strategy failed, screw your long report and the same words you use every freaking week that, hey, how you doing? I'm your client. I don't want you to ask me about my freaking weekend. Show me that you're gonna make sure that I make money first and then ask me about my freaking kids. Print this out and if I come back to you in three months and those bubbles haven't moved and I failed or we couldn't execute, it's powerful. Are you guys ready for the next one? That's what I hope so. Or I hope you don't want the next one because I'm gonna crush all of you. Competitor analysis, the next one. The old way I would do it is I'd say to my client, give me five or 10 of your competitors and then I would drop them into a tool. It was horrible, new way to do it. I analyzed 70,493 different competitors from us. They have 700,493 unique domains ranking in the top 20 for keywords that they spent money on last year and this is what that thing looks like. It's a thing of beauty. Notice at the bottom, 54,000 words, you saw that before. I'm looking at 233,000 URLs, 70,000 domains like I mentioned and there's their cost. So who are Maz's competitors? You tell me, name one. Ahrefs, name another one. Google, who said that? God damn it, you're good. I thought that was somebody on my team who saw my fucking slide deck. Let me show you, you cheated. Neil Patel, Neil Patel is the number one competitor to Maz when it comes to unbranded words, they got at least one conversion. And then it's Word Stream, then it's Google, then it's HubSpot, then it's SEMrush. How many of you thought SEMrush or Ahrefs were gonna be their competitor? Now this is their content competitors, okay? I'm not saying that they gotta go and steal people's content and put it on their blogs or anything, you know, to win. So you think like okay, let's go after those guys, yeah, let's like put Ahrefs into the tool and figure out their content strategy, you would've missed Neil Patel. You also would've missed thousands of other sites that Maz could just take their freaking lunch money. I've never been to SEO site checkup in my life, but it ranks for 160 words in the top 20 that Maz gets conversions on and all these other freaking sites, Optimost and Link Assist and take their lunch money. So keywordtool.io, I put my mouse over the box and I go okay, they have 209 different words. What's really cool is when I click on entrepreneur.com and at the bottom it shows me this. Hopefully you can see it in the back. Let me look at it my damn self, whoa, I'm getting close to that thing, don't wanna do that. How to build backlinks to your website, they rank number one in three. It's gotten us four conversions, not eight, it's duplicated because they rank one and three. So I can say man, I think Maz is probably a better piece of content than Entrepreneur Magazine. Now there is one problem when you use PPC data. Is last year Maz was improving one of their tools, so they didn't invest as much in paid until that tool was improved. So that's something to consider when you're using paid data. See the value in paid data to me isn't trying to return an ROI. It's in knowing how to deploy my resources in something that's gonna make my client money. How many of us have burned up our client's internal resources thinking that we need to improve page speed and then the page loads faster and the rankings don't freaking change. Now you've used your internal political capital so next time you're like, no, we really have to do this. They're like, sure we do, like that page speed thing you said. Can I do another one? I hope so. Feel free to say no, I got a flight to catch tonight. Long tail, how long is the long tail? You know clients ask you questions like that, well how should we be concerned, the old way, I'll tell you what I used to do. I do, find some article, find it on Mars or search engine land and send it over with an analysis. But it was always somebody else's data and their best practices. And God, if we've seen anything in this conference, we've seen that SERPs are different in so many different industries. So let me show you the way I would do it now. All right, this one's a little bit hard, so pay attention. I like to group my data into these groupings. So what you'll see here is the circle represents all of Maz's conversions according to paid. So you can see that about 28% of all of Maz's conversions come from keywords that have gotten over 1,000 conversions. But I've made a tooltip that shows there's only one word there, that's their brand. So when I look at the second highest group of converting non-branded, I believe, keywords, they're the keywords that converted once all year. All year. Those words rarely have any search volume or very low, which means most of you never see them, but it's the biggest group of converting keywords for Maz and for every other client I've ever looked at this for. So when we look at unbranded, look at this, I can take my mouse and put it over this and go, okay, well, it's actually 40-something percent there. It's one-time conversions. You can see all the keywords they have, 25% only got one click all year. And then you can see that over 50% spent less than $100 all year. How many of you are sorting by CPCs and volumes and missing all that? That's where most of the conversions are happening. On the words that spent less than $100 all year. But you know what? Those words don't have a lot of volume. So guess what? You go to your tool and you sort by search volume and I let you all battle over the same 10 freaking words. We slide in on the bottom, find the other 10,000 words that nobody looks at and optimize for those. And because nobody's looking at it, it's a quicker optimization and I get quicker results while your client's still pissed off at you. So now you look at the costs and all these metrics down here. Now I'm gonna control and I'm gonna click on the less 100 and click on the 500, so I'm gonna sum those up. That number goes from 375,000 to 247,000. Two-thirds of Moz's paid search spend is on key, no, two-thirds of Moz's conversions from paid are coming from keywords that spent less than $500 all year. Those never show up in the tools you use typically. Can I do another one? Not, that wasn't good enough. Thank you. This is my baby. I've been working on this shit for a year for you fucking people. So a client asked me this, how much content do I need? The old way I used to do it? Oh, you know, you got 30 keywords and about five keywords per page and six pages. Client's like, yeah, sure, okay. I mean, I guess, best practice, sure. The new ways, more intermediate, it's gonna take you some time. Pay attention to this, cause this is fucking good. So now I've got my box of all my domains that I'm competing with. I got all their 200 and something thousand URLs. That list is just URLs, URLs, URLs. I actually have a filter on this, so it's only top five. So you can see it's 37,000 URLs. Sorry, small data. It's cause I filtered them down. Cause now I know I'm only talking about keywords that rank in the top five. And then I've got my search terms that are applied to that keyword. So let me show you how this works. First of all, out of all the URLs I looked at, 17 ranked for over a thousand unique words in the top five alone. So what do you think I look like when somebody says, yeah, I think we should do about 10 keywords per page. What the fuck, well I read it on search engine land. Pull your own data, come up with your own best practices. So let's pick one of the domains so we can make mods some money so they can pay for all this awesome shit. Let's pick on Neil Patel. I'm not picking on Neil Patel for any other reason then. He's got a big box and I want mods to own that box. So what I do is I click on Neil Patel. What you'll notice is in the bottom left, I can see the page that he's got that ranks for 2,400 and 24 words that I got conversions on last year. No, sorry, that's all words that I bid on last year. That one URL ranks for 2,400 and 24 unique words. And then I can see all the search terms for that one. So when I click on SEO Analyzer, keep in mind you always sort by conversions. So if you remember anything, be like always sort my boxes by conversions so we can sit with an executive, you're always talking about money first. I don't know about you. Would you rather be paid in search volumes or money? How many of you are sorting by search volume right now? I know I was and you guys thought I was smart. So now I sort by conversions and I can see all those teen rankings. That's Maz's rank. But I know that Neil Patel, look at that filter below my categories. Below my categories, see how my filter is a ranking of one to five? So whatever these words are, Neil Patel's ranking between one and five for all of them and I'm not even close. And it drives a total of 302 conversions. Let's go take Neil Patel's lunch money. You click on it and you go into focus mode and Power BI and you start seeing that there's a theme here. Damn, we just gotta start using the word Analyze a little bit more and we could probably take Neil Patel's lunch money. And because I've got conversions there, I can, you know what, having conversions there keeps you having empathy for your client that they gotta somewhere pay the piper on revenue. It keeps you constantly focused on, oh wait, let me make sure this is converting for you. Oh, the snippet showed up. Well, does it affect my conversions? I don't know. Will it get the fuck? I need to make money for my freaking customers. I can't, thank you, I appreciate it, but if you can't tie that to revenue, I don't know how to prioritize it. All right, so we found that one. Let's do the next one. Sorry, Neil. So now I just go down to the next one. Scroll down, click on the next URL. 1061 unique keywords that Maz has bid on last year. I click on it. I'm sorted by conversions. Free keyword tool, best keyword tool, blah, blah, blah, blah, blah. This is one of the ways that I'll say to a client, like they might go, well, we don't really think we should write that content. I'm like, but you're fine bidding on it and getting a lot of conversions for it. How is the customer different just because they scrolled down 30 more pixels? Explain that to me. Why would you not want to go after that? Yesterday, Dr. Pete stood on stage and I disagreed with him hardcore about something. He said, well, maybe people want review sites and we shouldn't really write that kind of content. And I go, you can because people trust your brand. Like if Maz, I don't know how many of you guys have seen Rand stand up on this stage or talk to him or see his blog post where he's like, hey, you know what, if you need that, this other tool's actually better for you right now than the tool that we have right now. That builds trust out the freaking ass. So I would say Maz could write that data all day. If anybody hears from HubSpot, just Google best CRM software. HubSpot ranks number one for it and the content is good. It's possible to rank for best and review keywords if your ethos as a company actually doesn't mean you have to talk shit on every one of your competitors like they all suck. But if you're in the kind of company where you don't have the balls to do that, then you can't go get these conversions for free. I just gobble up Neil Patel's rankings all day. Seein' a little bit, Rio. And I do it for keywords that I know convert and make this man some money. Can I do another one? I hope so. Let's do it. People used to talk about optimizing for brand. The old way for me was I used to talk shit on those people because that was a tactic that a lot of crappy SEO firms would do to try to get their clients to be like, oh, I rank for 1,000 keywords. You're like, but yeah, they're all your brand or you're getting ripped off. So for the longest time I was like, you are so dumb. You are really dumb for real. But actually I was the dumb one. I was the dumb one. Because I missed an opportunity to help those clients save money. This is something you can do probably in a month or so. When you bring in all your paid data, one of the things you can do is you can do a report like this. Show me the boxes. Each box is the size of the number of keywords or the bigger the box, the more keywords. See how my slider is set to one in one? So it's only for keywords that, only for domains where they rank number one and they aren't Moz for branded keywords. So I'm like, Mozilla has a bunch of words. So all I gotta do is click on the box. Boom! Oh shit, people are searching for MAC addresses and some weird Moz extension. And they've cost me 1700 bucks. I'm gonna start doing some negatives on that shit. That's why I should have been tracking rankings for my branded keywords as well. Because by doing that I went, whoa, whoa, whoa, wait, wait a second. How is this domain ranking for words that are my brand? And watch what happens is I click on box after box after box. Tools Moz or test Moz, that's a bunch. Get rid of that test Moz stuff. That must be somebody else. And there's Moz Angelus, which is some T-shirt thing. I clicked on Facebook. John Lewis has this thing called Moz the Monster. We got some clicks there. Save that money, get all that money back so you can go and say, hey, now that I've saved you this money, here's what I think it's gonna cost to write that content. And remember the whole time I'm calculating conversions to make sure that I don't click on something that could convert. Because if I click on 99 signals, I go, whoa, well, they rank number one for a lot of these words. And to be honest, I think that's another conversation that we could have with Moz. Where I can say, hey guys, I think people trust you enough to write Moz versus SEMrush or whatever, right? And if they already have it, they're not ranked number one for it. So let's help them, because I know it converts and can drive them money. All right. When you look at things like Cora, you're like, damn, Cora's answering the question, what is PA, what is Page Authority? In the number one position, you're like, well, maybe we should go get that back. All right. So one of the sidebars I have for you is when you put this much data together in one place and all you gotta do is check boxes and slide little things, you start just being like, why not check fucking boxes? So the speed by which I hypothesize is very fast because my processor isn't trying to sort a million rows. So I'm like, check, check, nothing, check, check, nothing, check, check. Oh snap! You spin all this money on this and then I can take the template once we build it and then apply it to every client in our data warehouse and then they all get that template. One of our clients said, hey, Will, I'm wondering, let me go back. One of our clients, I love this one, one of our clients said to us, hey Will, I have this hypothesis but I don't know how to test it. He's like, you already have all my data sitting in your little data warehouse. I go, yeah. And he says, I wanna know if I have to use more words for canned words than what words, than whatever, whatever, for when I'm trying to get, when I'm trying to write certain types of content in the financial space. So in about an hour, I put this together for him where I have the median word count and I can say, what's the median word count from canned? How, what, where? How did I get the word count? I took all 200 and something thousand URLs, ran them through Screaming Frog and got the word count from Screaming Frog and joined it at the URL level and now I can show the client exactly if there's a big difference on people who rank high for canned words versus what words versus how words versus where words. And those words are anything. Any word he came up with, I could have made a filter for and then showed it as a checkbox to say, yeah, for these words, you need a higher, you need more content to rank in the top five for these but these, you need like a thousand words. And when somebody goes, well, why did you say a thousand words? I go, because I looked at your data and the people outranking you for the words that convert for you. That's way different than, I read this on Search Engine Land. All right, got another one for you. You ready for it? Are you guys exhausted? All right, let's keep going. Which snippets matter the most? We talked a lot about snippets today. The old way is, there is no old way because I've never seen somebody talk about snippet impact on conversions. But you gotta imagine when a snippet shows up, it probably impacts click-through rates and paid and other things, right? This one's a little bit harder, but I'm gonna show you the easy, easy one starting with people also ask. So what I do is I get all my boxes where each, so on the box, the size of the box represents the domains. Let me get this right. Represents the domains that are the answer in the people also ask expansion. Okay? All right. I've got all the keywords that Maz has been bidding on sorted by conversions, so I know I'm talking to Sarah Bird and the team about money. And I go, all right, so you rank in the top, I think I did top five, for the word keyword research. And it triggers four different people also ask questions and you're none of them. So when somebody searches for keyword research which gets you a lot of conversions, all the people also asks, you don't show up in. Maybe we should prioritize that a bit and see what happens. I don't need Dr. Pete Fixen, how much does Google make in revenue? I need to knock out these guys because they're sitting for words that convert for me. SEO audit, entrepreneur magazine, what? How do I do SEO for my website? Is one of the people also asked for the word SEO audit and they're ranking entrepreneur magazine and last interval, Maz got 33 conversions on that. Do you think that Maz could beat entrepreneur magazine out on how to do SEO for my website? Yeah, Screaming Frog, all those URLs showing up and then that's how you see if they've got H1s, H2s and how they're structuring their data. So then you do that at scale across hundreds of thousands of URLs and you go, hey, I looked at everybody that's outranking you for rich snippets that you want and I use a Screaming Frog to pull all their tags and their hierarchies. And what I found is that in your space, a lot of people have H2s, great. All right, let's do that then. All right, I'm gonna skip ahead. All right, Pantsless Rio says, new feature time. All right, so what I want you to watch is you gotta watch this very closely because I'm gonna show you something called a drill down in Power BI where I drag one of like 150 different selectors under my tree map, watch it. I drag it in under domain, nothing changes. But what it allows me to do now is right click, drill down to get the individual URL that's showing up for all the words that I'm bidding on and the people also ask questions associated with it. So word screen, when I right click and drill down, boom, I can see that word stream has a singular URL that Google trusts a ton across 3,000 different keywords that we bid on last year. So in 3,000 different keywords that we bid on last year, one of the people also asked was answered by this one piece of content and I know that it converts for me. I'm getting tired. Can I do one more? All right, all right, all right. This shit's fucking hard. All right, trust but verify. I don't believe tools anymore. I'm gonna show you why with data. The old way. Trust the data. Fuck that. The new way is gonna take you a couple months. So I'm not gonna name out names because people get all hooked up on the freaking, the name like, oh, he was shouting this person out. What I did is I took all my words or search terms from paid that converted between 10,000 and my max number of conversions times and I counted up how many words there are in my list versus this other source of a keyword corpus. They're missing one. Not bad, cause especially cause my unique words could be a misspelling and other things that stem. But it's 96,000 conversions. So you miss one of those, that could be a big deal, right? Then I went from 10,000 to 1,000 conversions. They miss five. All right, after stemming maybe it's only two or three but it's still 39,000 conversions for this client. This isn't Maz's data. Then I did 100 to 10,000. So individual words that converted between 100 times and 10,000 times. And now I'm at double. So if you're using a keyword corpus where you drop a domain in, I now know the shit you're never gonna find because it doesn't exist in their corpus and I've attached it to conversions so I can prioritize which ones I go after. Does that make sense? Isn't that disgusting? You think that's bad? Wait till we look at the freaking longest of the long tail. Half a conversion to 100 conversions. I have a list that's 60,000 words last year that drove between half a conversion and 100 conversions. And if you go to this keyword corpus that all of us use, you're getting 2,531 keywords. Damn, this is fun. The amount of times where my jaw hit the ground where I said, how long have I been doing this wrong when I found this stuff? Because seriously, if we are competing in the same space, you drop the domain into the keyword corpus and I take the paid data? You're missing 95%, let's say after stemming. Maybe you're still missing half. I'll just go after that half because you're never gonna find them unless you take the paid data and bring it in. All right. All aboard. I got five minutes left and I'm gonna go through some quick stuff. I'm gonna skip this part. The one thing I'm gonna tell you is the refresh button is your friend. If you set this thing up into Google Cloud or you set it up into folders, you just drag a new ranking report in and everything updates after you hit refresh. None of that pulling all this crap and normalizing it anymore, you just pull it in. It works really fast. I'm gonna skip over this one. The one thing I'll say is I'm trying to find a way to get more dates, published dates, because then guess what I can say to my client? If you write this content, half of the URLs that rank for this were published in the last three years. So now I take their yearly number of conversions and multiply it. You're like, most of the content's been up there for years, so if we write something fresh, we might be able to disrupt them, which is what kinda happened here. So I wanna show you some inspirational things my team members have done. Some of them couldn't make it because they're on vacation, one's on a honeymoon, you know? Maz, honeymoon, I guess so, Croatia. Look at the click-through rate curve for Sear. See, the idea of looking at click-through rate curves generically is a farce, but I'm here to give you a tool to actually understand your individual click-through rate. Notice that the difference between mobile versus desktop for Sear. That's a massive freaking drop. 25% in position one to 10. That's a massive drop. Position five gets almost five times, four, three and a half times, the amount of clicks on the mobile. How many of you are looking at your own click-through rate curve so that you know where the disruptions may or may not come from? Imagine if I had all my different categories over in the left. I could get click-through rate curves for each one of my categories. Imagine if I had snippets as a bunch of checkboxes where I could go, hey, when this snippet shows up, how does it affect my click-through rate curve? It's just a click for me. It'll be a click for all of you if you watch those videos that we put on YouTube. Kim Jones, Kim Jones, put this thing together. Y'all don't know who Mike Jones? There we go, there's some of y'all in here. Some hip hop heads, all right. So what Kim did is she looked at crawl for her client and she compared it to their highest converting pages. Do you notice that this client's second and third highest converting page don't really get crawled nearly as much as some of the other pages that don't convert as much? Now we're talking about improving crawl depth because of money. Go ahead and talk about crawl depth to a CMO. Most of them are gonna look at you like you've got three heads, they don't understand it. It's not their language. This page doesn't get crawled as often and it's your third highest converting page. Let's just say. Okay, great, then what can I do with that? Well, we probably wanna get that thing crawled more often because every time you make a change, I want you to get freshly in the index as fast as possible as you can start to recoup the value of the investment you made to that page. Oh, okay, so you want me to, and you know it's not gonna be a waste of my time because it's one of my pages that converts a lot. Yeah. I had to edit my slides and put this in because this guy Josh, who's only been with us for like eight months, right out of college built this thing. He got with a bunch of other people on the team and did some X-Paths, which I don't even know how to do, in a Screaming Frog. And he pulled a lot of different features out of YouTube. So now when you see YouTube ranking, what we can do is we can say, it's one thing to know YouTube is ranking for words that matter. It's another thing to understand exactly who those channels are that are ranking for those. And to know the keywords that are ranking for those and do those keywords potentially convert. We don't have conversion data. We don't have PPC data for this client, so that's why you don't see it there. But I can look at that by category. Click, click, click, click, click, just going through. My local SEOs, Bing's maps are awesome inside of Power BI. So I can look at different regions. I can zoom in, zoom out, drag it and whatnot for different regions and see how my different competitors stack up in different regions. So every pie chart represents each one of the competitors. You can see in different regions there's different competitors that have different levels of saturation. I can just drag across the whole United States showing that to my client. This one's really cool. That was Scott Taft that did that one. Tracy McDonald. Tracy McDonald was like, well, you know there's an R engine inside of Power BI. I'm like, uh-huh. She goes, Will, if you know how to copy and paste, I go, I know how to copy and paste. She's like, if you can copy and paste in a script, you can create your own ranking factors, guides with your own data. Oh, this is a correlation plot. So now what I can do in R, which sounds so sexy when you say, oh, I used R to get a correlation analysis on whether or not a store page is highly correlated to rankings, and the answer is it's not at all, according to only your data and your competitors' data, not some bullshit that I read on somebody's site with a bunch of other people in a different industry. Oh yeah, why would I ever look at something from somebody in another industry? It might not apply to me. Bing, frickin' O. All right. Here's another one that's kind of sneaky. Katie Pulaski is doing this one, finding keywords that have no conversions, but a lot of assisted conversions. You know why? Because everybody wants to ignore those, so therefore you're not gonna get the assists later. This is a very hard report to pull. She likes using Tableau because it does many to many joins, which I didn't even know what a many to many join was up until about eight months ago. You can learn this stuff, I'm telling you you can. All right, are you ready for the next one? There is no next one. I gave you all I could. They're gonna yank me off in a minute. I wanna say something though about our industry is this is a competitive advantage for us big time. Seriously, think about the last competitive analysis you did in the way that we do ours. And I'm not saying that to belittle anybody. I'm saying is you gotta be willing to admit that your shit's not where that is yet, and I'm showing you how to do it. I want you guys to compete with me. The better my competitors are, the smarter I have to get. The more I have to innovate. The more I gotta figure out the next new thing to do. And our whole industry was built off of us sharing shit with each other. All of us here mostly learned what we did because somebody was willing to share something that a lot of other people would be like, oh, that's proprietary, you can't share that on stage. Fuck that. That's how we all learn. And I want you guys to all try this stuff, share your templates with people and share it so we can all get better together because I'm still our newbie at this. I wanna say thank you to my entire team because I am just a mouthpiece for so many amazing people that have inspired me to build this stuff. This is not just me on my own. Get a bunch of people in your company who geek out about this. I'm thankful for this team over here who is getting up at 7 a.m. EST before work every other Friday so we can all geek out together and get better at this. We're gonna be sitting with our laptops and our computers to help all of you who did your homework until they throw us out of this fucking place so that you guys can speed up your speed to get better at this because I think we all need to. So we're gonna be sitting somewhere. I don't know where they are. We all wear blue shirts and I'm like one of the 10 black guys here. So find us and until I either have to go to meet Microsoft because they wanna talk to me about this shit or until my flight leaves, I'll be here to help all of you guys get your first steps. Thank you for having me for my eighth year. Let's go kick some data ass.