 Thank you for everything. Being part of our community, being part of our event, being here with us through some tough times in the last year or two and through the good times too. Thank you for coming out Wednesday or Sunday if you were here for the workshop. You might not know that a few months back we had this bright idea. I think it was Russ Jones, I'm gonna blame Russ for it. Whether he did it or not, I can't remember. I said, you know, why don't we take the longest week of our year, the most stressful week, and let's add a date at the beginning and talk for four hours. And I guess we were all high, I don't know. Like there was a gas leak or something. And we all thought this was a great idea until like a week ago. And then the fans kicked in, I don't know what happened. And we all said, what the hell did we decide to do? And you know what though, it was good. We, it went really well. I hope you had a good experience there. It's amazing to be up here in front of 1,000, 1,200, 1,400 people. But to be in those rooms with you with 20 and 40 people and hear the problems you're going through was a great experience. So thank you for that. I hope some of you will join us next year. I also want to thank Chris Smith. I'm not going to call you out, but Chris sent me a tweet the other day that encapsulated the problem I'm having with this deck and I had in creating it. And he said, I'm going to paraphrase. He said, you know, Dr. B, I really look forward to your talk, but dude, why is it always scary with you? And if you've seen me speak, you know, I apologize and it's kind of my job. My job is to look ahead at where Google is coming from and what they're trying to do and what's going to hit us next. And both from a marketing and a product standpoint, we have to know this and I feel it's important to know this and to inform you, but it is scary sometimes. And you know, the kind of thing I do, I get up here and I show you screens like this. I say, you know, type in Dr. Salary. And actually I sat with someone at lunch yesterday whose lunch was eaten, so to speak, by this kind of change, you know, because Google's just saying, here's the median Dr. Salary and here's some other professions and here's a knowledge panel on the right and here's people also ask related questions. And if you click on one of these questions, then organic is completely below the fold. If you've read any of Brittany's work on the infinite PAAs, you know, you see that if you click on these, it actually generates new questions and the box keeps refreshing and keeps changing. And you can have this entire pseudo-organic journey in this box and never see an organic result and that's scary to us as organic search marketers. I understand that. And you can see new things like this, the job listings where if you're in this business and you woke up the morning this lunch, that was a very bad morning for you. And this is scary, I think, because it's so specialized. And this doesn't affect most of you, but when you see this in the back of your mind, you're saying, when am I next? When is the other shoe gonna drop and what do I do about this? And I think that can be frightening for us. And sometimes on stage, I even scare myself. And this was 2015, for some reason, I became very frightened of my own finger. And now it seems fine, I don't know why, I don't know what happened. So here's what I want to do today. I want to take five topics and I guess we have a five theme, Chris and I, this afternoon. That's a good number. And I want to talk about five things that I think you're aware of right now but might be afraid of. Things you're not sure what to do about, things that are on your mind, things that maybe your boss or your client are saying, what are we gonna do about this? How do we handle this? And maybe you don't know. And I'm gonna try to give you five takeaways, one thing you can do for each of these things. One of them you're not gonna like, but I hope these are actionable things where you can go home and instead of sitting in the corner sobbing and drinking as you're reading the notes, you can say, you know what, today I'm gonna do one of these things. So the first one's the one you're not gonna like. HTTBS. This is the Mozcast data set, 10,000 keywords that we track. As of June 30th, 55% of the results on page one of those 10,000 queries were secure. As of about a year ago, it was only 30%. It was funny when I published the 30%, I got a bunch of people who said, well, look, what you're not telling us, if 30% are HTTBS and 70% aren't. And I said, true, good point. And then when I published it, it was 50%, so people said, well, what you're not telling us that there's 50% are than 50% are. Like, well, your math is still correct. I'm not sure how comforting that is. And I have a feeling in about a year, somebody's gonna say, when I say it's 95%, they're gonna say, well, 5% still aren't. I'm on the boat and the water's only up to here, but this part of my head isn't drowning yet. That's gonna become less compelling. And I wanna look, if we track the trend ahead, I think we're gonna be at around two thirds by the end of the year. So this is advancing. This is an 18 month timeline where we've gone from around a third to we're probably end up around two thirds. This is happening. We know Google is serious about this. Of the top five sites in the dataset we track, Wikipedia, Amazon, Facebook, YouTube, Yelp, you've heard of all of these, all of them are secure now. Of the top 10 sites that we track, eight are secure, of the top 20, I believe it's about 16. The big players are taking notice, they're switching to HTTBS. That will drive Google's ability to push this message. And we know they've pushed the message hard, we've had an update. This is not without its perils. The folks at Condi Nast who run Wired put out a very transparent article, this was last year, I think, about how the switch didn't go very well and they had some trouble. And so I understand this is scary, I understand this is difficult. I'm gonna be totally transparent even as I encourage you to do this. I ask Gary, as the numbers are rising, do you see another big HTTBS algo update coming? I know that you can be skeptical of what Google has to say, but Gary has been very close to this HTTBS project from the beginning. And he said, no, we aren't planning on that yet, but, and this is the important, but the Chrome team is very serious about implementing warnings and implementing changes around HTTBS. And so what you're gonna see in the next few months, most likely, is that Chrome is going to start warning not about just e-commerce sites, but things like lead gen forms. Maybe you don't sell anything and you think, I don't have to worry about this, but you're a doctor or a lawyer, you're a plumber and you have a lead gen form, and all of a sudden Chrome pops up a warning that says, this isn't skewer, and you start to lose conversion. That's gonna happen in the next year. They're gonna become more aggressive about that, they're gonna push this harder, and you don't wanna wake up and just find that's changed. So what do you do? This slide takes a little explaining. A couple years back, I read a story about a pen and teller trick, and they were kinda making fun of Siegfried and Roy, and they were saying, you know, they conjure up these tigers, and they've got like six tigers. We're gonna do one better. We're gonna get seven things. And so they do this trick and they produce seven bees, just to beat six, and they're making a big deal out of it, and they're talking smack about them. But then they keep going until there's just dozens and then hundreds of bees, and they're pouring bees on each other, and there's bees everywhere, and the room is filling up with bees, and you can watch the video on YouTube, and you're thinking, this seems a little dangerous, and then you look at the camera guy, and the camera guy is just decked in the full bee suit, like bee armor. And pen and teller are just sitting there with these bees, and somebody asked them later, they said, how did you do this trick? They said, there was no trick, we just did it. We got stung. This was a stupid thing to do, but we had to do it. We had to do the hard thing. And this is the first thing I want to tell you, sometimes we have to just do it. And I know it's tough, I know HTTPS is not, it's gonna be risky, I know it's gonna take time, but you need to look now, what's it gonna take? How long is it gonna take? What are the engineering resources gonna look like? What are the risks? What's it gonna cost? Because you don't wanna wake up tomorrow and find that one of these warnings is launched, something major has happened, and now you have to do it overnight. So I want you to start thinking about what that's gonna take. Serp features also can be scary to us. You know, here again, another example, veterans benefits. We've got a featured snippet, takes up most of above the fold. We've got tweets, we've got a knowledge graph, we've got one organic result, and it's the VA themselves. You're not gonna compete with that. That can be intimidating. Here's a news, a very news specific kind of query. We've got top stories, and we've got tweets, and current tweets, current news information is dominating the first page. You wanna get a California king bed? They don't know if you wanna buy one or watch the Rihanna video. This is a very weird SERP to try to compete on. Even Google doesn't know what you're looking for. 87% of the SERPs in our Mozcast data set right now have some kind of rich feature. This, I've told you this a hundred times, this is not the anomaly, it's not the exception, this is the rule. This doesn't even include PLAs, ads, shopping results. It's well over 90%. So we know we have to face this. But we don't know what to do because there's so many of these features and they're launching all the time and they're changing all the time. So here's the message I wanna get across today. Here's an example. You're trying to rank for Beyblade Battles, and you know what? You go and you look at the first page and you realize the first five are all videos. If you're trying to rank for this and you're not producing a video, and I know these are YouTube and I know that's a dicey topic, but just in general, if everyone on that page has a video except you, you're missing something. This is what you need to focus on. So how do we do that at scale? Rob showed you earlier. Thanks, Rob. And Stat has tools for this too. In MozPro, we have feature tracking. You can go into your campaign and you can just look. And this is for Moz itself. This is a Moz campaign. I won't, the legend is a little hard to understand. I apologize for that. But that scissor on the left, for example, is featured snippets. And the gray tells us that there's a lot of them and the keywords I track and the blue tells us what we're in. So we can see that, you know, there's videos in our SERPs, there's review stars, there's featured snippets, there's related questions. But there aren't local packs in the things Moz tries to compete for. There aren't news results, there aren't a ton of tweets. And so you need to go in and look at your own data and say, you know what, I don't have to compete on all these things. I need to find out what's going on in my space. Is my space a news heavy space? Is it a local heavy space? Is it a tweet heavy space? Are there a lot of videos? Are there a lot of questions that are being answered with featured snippets? And I want you to pick something and focus on it. I also want you to know where not to spend your time. This is the medical knowledge panel. I've talked to people in this space unless you're one of the big players, unless you're a WebMD mail clinic. This is content Google has created themselves with partners, with knowledge. They've put these graphics out. This isn't a great SERP to be on. Type two diabetes, that's a very, very difficult, expensive SERP. Unless you're one of these top players, don't bother. Margage calculator. Let's say you built a mortgage calculator. Congratulations, so did Google. And theirs is gonna win. And it's good, and it's fine. You know what, I think this makes us angry, but you know what, from a consumer standpoint, there are a lot of bad tools out there. Google standardized it. They realized this was useful. They did what people wanted, but this is not something you want to rank on right now. This is a featured snippet for when is Squirrel Appreciation Day. It's January 21st for your favorite squirrel. You want to think ahead. But what else do you need to know? And the featured snippet even has the three years ahead. If you want to know Squirrel Appreciation Day in 2019, you're done. And I'm gonna say if you kick forward to 2030, looking at this data, that is probably January 21st. So why would I ever click on timeanddate.com even though there's a featured snippet? Great, I got an organic result, but I'm done. It's January 21st. Stay away from this. Somebody said this before, I think it was old Reynolds, and he's absolutely right. Samsung TV. If you're not Samsung, you're gonna pay for this, sir. That first set with all the extended, that's all an ad. Those first five links are all ads. There's nine shopping results on the right. Don't be an organic agency that says to your client or says to your boss, because you only do organic, that they don't need to worry about PPC because you're lying. This is a paid search. And if you wanna rank for that, you better accept that reality and find out what it's gonna cost, or you're gonna need to get in the long tail and you need to strategize differently. So be aware of that, don't deny it, because some serves are paid. So if you don't know where to start, you don't have the data you can't pick, I'm gonna give you something to pick and we're gonna talk about this a little more in a minute. I think you should pick featured snippets and related questions and here's why. Because featured snippets are an extension of organic. You already know a lot of what needs to be done to rank for them. I think this is a healthy thing to go after. In our data set, they're about 15% of queries have it. Ours is a very headoring in the data set. I mean it's commercial terms, it's single word terms that are not great features snippet terms. So we think that number is really much higher. So we're seeing a lot of them, we're seeing a lot of growth. I think this is a good thing to focus on because you already have the skills for this. So I'm gonna talk about that in a minute. Rank brain, the scary one. And I talked about this last year, I know a lot of you are worried. I want you to worry a little less. I think this is part of our problem. This thing that they said, rank brain is one of the hundreds of signals that go into the algorithm. In the few months it's been deployed, rank brain has become the third most important signal. And we went holy shit out of hundreds. It's number three. And then they asked Paul Har, I can't remember who it was, I think it was Paul. What are the first two? And he said content and links. Well content and links is like 700 signals. Maybe exaggerating. 150, I don't know, it doesn't matter. So third most was kind of a conceptual idea. The other thing we know about rank brain is that rank brain is a layer that is operating on all searches, but that doesn't mean that it's affecting everything. It's happening, it's having more impact in the longer tail, it's having more impact with natural language. So I think we've overweighed a little bit. What is it? I was just like scouring the web for anything I could find the other day that Google had published out loud. And I found a video of Jeff Dean from the research team presenting in South Korea. I'm like, great, I'm sure nobody's watched this video, Jeff Dean in South Korea. And there's a slide that's really great. And I kind of stole it and refurbished it, but it's a real slide. Krista's like taking notes down here and I hope not reporting me for anything. But this isn't secret, it's on the web. And it's very useful and it's a common problem. It's a problem that Google is trying to solve. And here's what I'm gonna say about rank brain. Don't worry about exactly what rank brain is. Don't go out and take seven deep learning classes. Don't be like me and take the first three days of 100 deep learning classes thing quick because you realize you actually have a job to do. You can't spend your entire life taking deep learning classes. Understand what the intent is. And this is the simplest explanation I can give of the intent and this is a great slide by Jeff Dean. If you type in car parts for sale in Google, this is a classic search problem. And you look at document one. And document one has the phrases car parking available for a small fee and parts of our floor model inventory for sale. That matches all the keywords. And then you look at document two selling all kinds of automobile and pickup truck parts, engines and transmissions. It only matches one keyword. But document two is a much, much better match. Any human can recognize. This is the problem Google's been trying to solve for the years. This is part of what they're trying to solve with rank brain we strongly believe. So just understand that intent. This is an audacity course that Google does. They had a quiz. I talked about this last year. I'm not gonna go into depth. But they asked the question, if you were using deep learning for search, how might you use it? And I thought that's a very interesting question for Google to ask me and they gave me the answer. But look at what they're saying. They're saying that deep learning was a good match for machine learning and supervised learning. They're saying something else hidden in this slide which is that they're feeding the machine the input and the relevant outputs that's coming from humans. Google is looking what we think is relevant and teaching the algorithm based on that. That doesn't mean a whole new set of ranking signals. It means that we're starting to look at what people think is relevant. And that's a good thing. That's a good thing because it gives us the freedom to think more like people. This is word to beck from Google's own site, TensorFlow, I'm not gonna go into great depth on this. They were able to teach the machine just by feeding it, it's called unsupervised learning, by feeding it a massive amount of words and context. And it was able to learn things like the relationship between king and queen is the same as the relationship between man and woman. The relationship between walking and walk is the same as the relationship between swimming and swam. So we know that Google is able to do this. We don't know if this exactly is ranked brain but we know this is happening, we know they can do it. And we see it in play. We see this advanced synonym understanding. It used to be that if you typed in discount airfare, you had to find that text on the page. But now if you look at what they highlighted, if you look at what they bold, you see that they're ranking on discount and cheap and cheapest and deals and airfare and flights and tickets. And any of those things are being deemed relevant. I did a post that's still up there called Tactical Keyword Research in Rank Brain World and I did a weird thing and I went to Google and I said, tell me, what about this post is wrong? And they did and I can't tell you all of that which is extremely frustrating but I'm gonna give you the core message of what I'm allowed to tell you which is this. Google said we support, well it was Gary and John basically, Gary and John said, we support the message of this post. These are good things to do. But the truth is we were able to do a lot of this before Rank Brain. We've been, and there's actually a page on how search works that essentially says Google has been developing the synonym understanding system for more than five years. This is something that's been going on before Rank Brain. We think Rank Brain is part of this and an extension of this but this is the reality we live in whatever we call it. So how do we try and cope with this? You can go to the post, I talked about this a bit last year, I'm gonna rush through it. I'm gonna pitch Keyword Explorer just for a second. I know I don't usually push product. You're trying to rank for wedding rings. You type it into Keyword Explorer and you get back a thousand keywords. And it's all these little variations. Wedding rings and wedding rings for men and wedding rings for women and wedding rings cheap and wedding ring sets. And already you're thinking, how am I gonna create this content for these thousand keywords? And if I'm doing my job, it's gonna be hundreds of thousands, tens of thousands. What do I do? And the one thing I think you can do that's important is start to think of these as conceptual groups. So we have a tool, group keywords. I apologize for the cryptic, yes with little lexical similarity messaging that was built by our data science team. But essentially that just means, you know what, group these pretty liberally. Just, they can be pretty similar, but don't worry about it. And so we start to see that they put these keywords together and we get things like wedding bands, wedding rings, Walmart. All right, interesting. Wedding sets for women, unique. And if you look at some of these, these are representing dozens, over a hundred keywords in some cases. And I think the flexibility of where Google has headed and the flexibility of RankBrain, the good thing about it, is that we don't have to micromanage every little variation of the keyword. Every S and E, D and I and G, we can get away from that. And especially in the long tail, when you get into question research, you can have tens of thousands of variations. You can't target all of those. You shouldn't try. So I want you to pick and think about these concepts. So I just pick three. Wedding rings, Walmart, wedding bands, women white gold, wedding sets for women unique. And then start to write naturally and think like a marketer. And that's okay. Think like a human. So wedding bands, women white gold, I'm gonna go obvious. White gold wedding bands for women. Wedding rings, Walmart. Well, I'm not Walmart, but what about great wedding rings at Walmart prices? That's what people are really looking for, right? They don't have a lot of money to spend. Wedding sets for women unique. Well, what if I turn the phrase like unique wedding sets for unique women? Well, I wanna be unique, right? That's a good thing for a marketer. And this gives you the freedom to think about dozens and hundreds of keywords conceptually. But I think the important thing is it gives you the freedom to write for people. And I believe that's what RingBrain has done. And so it's still tricky and it's still a machine and it's still learning and evolving and that's gonna keep happening whether we like it or not. But as the machine is thinking a little bit more like us, we get to be more like people and we don't have to laser focus on keyword stuffing and all these outdated tactics. So I want you to play around with that. Oh, there's also a tool, Keyword Explorer, if you wanna play. There's one called Exclude Your Query Terms that basically says, give me everything that's relevant to wedding rings but doesn't have wedding or rings in it. And I think that's a good exercise. Just get that out of my head. Get those words out of my head. I wanna rank for this human concept that people understand. All right, we've talked a lot about feature snippets. I'm gonna try and give you one way to target them. You've seen these. They're in competitive commercial verticals. Here's one for credit score. My wife used to work for one of the big three. You have very expensive ads up at the top and then before they're any organic, you've got a feature snippet. You've got that number zero position, we call it. You're really trying to, this can be really competitive real estate. And even if you go to a four res high screen, you've got related questions below that and credit karma number one is way down there. These take up a ton of real estate. This is what burst stone is July and not only do you have the answer and an image but the text, you're taking up a huge amount of the screen on mobile. This is my pixel phone. It's a new phone, high resolution. It's the entire first page, a little more even. Featured snippets have tripled in the last two years in our data sets. They rolled out at about 2% so we think they've increased by 6,700% since they launched. You'll also see that it's kind of a plateau. So every once in a while, Google is going, you know what? We're gonna turn up the volume on featured snippets. It's gonna probably be low quality for a little bit and we're gonna tweak it. And once we get it right, we're gonna turn it up again. I think you're gonna see this continue. They're trying to answer people's questions. Couple of things to do. First of all, be aware that you do have to rank on page one currently to be in a featured snippet. So you have to do that organic work. It can be anywhere on page one but the relationship kind of dives. So we see about a third of the featured snippets are the person ranking in the first position. Another third in about two, three and then it starts to fall off. So we think that in that top five, that top four or five, that's a good opportunity. If you're ranking in the first four or five and actually in the first three to five, I think there's a special opportunity which is that let's say you're number three, I think it's gonna be easier to potentially get that snippet and write better content than to get to organic number one. So this can be a good investment if you're in that kind of first half of the page. Rob showed some similar data earlier. We pulled a big list and about two thirds in our list were text snippets, were paragraphs essentially. And then a little less than half of that were lists and tables and videos. Videos are kind of a special format. We're much smaller. I think they did an update to this. We got announced today, stat has a good white paper. I'll tell you the takeaway for me is that if you see Google is doing say a list snippet or a table snippet and it's a competitor and you wanna compete, consider that they think that format is important. So we'll see it a lot with how-to's. We'll see step-by-step instructions. You might want to format your content more around that. Doesn't mean it has to be a UL-ordered list. It just means you need a step-by-step kind of content. The one thing I want you to try, experiment with something called inverted pyramids. Inverted pyramids is a writing style. And the idea is that we're gonna give away the answer first and then expand. So here's our page for what is the title tag and we put it right out there. Title tags and HTML element, et cetera, et cetera. And then we expand into more and more details. We expand into things like, what's your optimal format? What's your title like? Why are they important? Why do you write one? I know this can be a little scary because you can feel like you're giving things away, but it gives Google a good thing to target for the snippet and it also naturally causes you to write in a way that's good for web users. These are people with a short attention span. They need to see they're on the right path. So they see that snippet, they go to your page that says the same thing and if they're interested, if they're qualified visitors, they can drill down, see the details and continue to read and get more information. If they're not interested enough to keep going, they weren't good prospects for you anyway. So I'd encourage you to play around with this. Think in terms of summarizing first. Put the answer at the top and then expand into more detail. This is a journalistic style. Here's a page on Northwestern's journalism site. This is not some kind of SEO trick. So I think this is something that can be good for search, good for users, good for Google, good for us, hopefully. And finally, voice can be a little intimidating. I go, okay, Google and I say, how much blood do I have because my seven-year-old asked me on the way to school. She's a little bit of a creepy question. Daddy, how much blood do you have? Do you need some of it or what are we doing here? According to Camp Wonderopolis, and there could not be a greater site to answer a creepy question than Camp Wonderopolis. That sounds like a great place to go. I'm gonna send my kid there so she stops asking me about how much blood I have. Scientists estimate the volume of blood in the human body. You don't worry about that. You can learn that later. Learn that on your own time. So if we go to the SERP and we ask the question, how much blood do I have? Well, scientists estimate the volume and that looks very familiar. That feature snippet is driving that voice search. We know that. We did a big study this year. Highly, highly correlated. This is the same engine. That's good news for us. That means that the actions we take for featured snippets to answer questions, they're impacting voice. We don't have to learn something entirely new. I'm gonna give you a small warning. Just be careful. Voice is getting into some new capabilities. So if I ask a question like, how do I make tacos? On desktop search, I get a step-by-step guide. If I ask Google Home, how do I make tacos? It says, I can help you make that. Would you like me to find your recipe? And if you don't answer, it actually keeps badgering you. I made the mistake of putting home in my office and sometimes I forget to answer it. And it's like, yes or no. Did I ask you? I went to the bathroom and I had a sandwich. I'm like, what's going on? Why are you yelling at me? But Google Home will actually go find your recipe. So Google has started to create some targeted content. We're just seeing the beginning. So you do need to go try it out. You can go to your Android. If you have Google Assistant, it's the same engine. But if you have to have one thing that you're gonna do about this, here it is. We looked at a ton of featured snippets. I think we pulled 10,000 questions and 1,000 of them had featured snippets. And we, I manually had the brilliant idea that I was gonna do a featured snippet study where I just sat there and asked Google Home 1,000 questions. And yeah, this took about two weeks and was a little bit of a dumb thing to do. But one of the things we learned was that by type, almost 90% of paragraph featured snippets generated voice results. Around about a half of the list snippets and maybe a third of the table snippets generated voice results. Because it's pretty hard to translate a list in some cases and especially a table into a voice answer. So if you're going to do something, target those text featured snippets. If you see that there's a paragraph snippet, the odds that that's gonna generate a voice result are very high. And this is kill three birds in one zone. You know, you're doing well in organic, you're getting that snippet, you're getting voice. You don't have to have a voice expert. You don't have to freak out about voice. This is part of your organic journey that you're already on. It's part of creating better content. It's a win-win kind of thing. All right, let's summarize. HTTPS is coming. Just eat the bees. So I have a confession to make. I misremembered this Penn and Teller story. I read it years ago and in my mind, at some point during the trick, they started eating the bees. Then I found the video and I found out that that never happened and I am crazy. But I had already made the slide and I really liked it. So I just left it up there. Because look, sometimes things are hard. We are such a low-hanging fruit industry, everyone's still on it. And I know many of you work hard and do the right thing, but I hate the low-hanging fruit questions. We're not willing to do the work sometimes. This kind of thing is tough and it takes work and it takes risk and we need to do it. And sometimes as professionals, we need to do the hard thing. And sometimes as grown-ups, we need to do the hard thing. So my advice to you about HTTPS is eat the damn bees. Sometimes it's just gotta be done. And I warned Stacey our captioning that I was gonna say that, just in case it sounded weird. Thank you, Stacey. Big hand for Stacey, who is amazing, by the way, doing our captioning. Because we talk fast and we say weird things. So eat the bees. Don't actually, a beekeeper actually responded to me and tweets like, please don't eat the bees. That's very dangerous and bad for these. I'm like, I love the bees. I'm sorry. Please know that's metaphorical. Don't eat any bees. Serp features is going to continue to evolve. I know this is scary. I cannot tell you what's coming. I cannot tell you what thing like jobs is gonna launch tomorrow. I don't think Google even knows what's gonna happen a year from now. This is an ongoing experiment. Don't worry about all the things that could happen. Pick one or two and get moving. And I think featured snippets are a great start. But if you have the data, if you see that you're in a news heavy industry, you're in a video heavy industry, you're seeing a ton of related questions, pick something, learn about it, try that. Don't try to do everything. Don't fix state on rank ring. Write for people. If you had to do one thing, write for people again. Stop keyword stuffing. Stop messing around with this. You still wanna do great keyword research. You still wanna know what you're ranking for. But don't feel like every single variation has to get into that content. Still write the content for people. We know user signals are becoming more important. We know Google's looking at what we interact with. We want good content that's good for people. As the machine learns more, that's what the machine's looking for. And the machine is going to recreate the ranking factors over time based on that. And that's a good thing. People ask questions. Google understands this. They have to answer them with limited screen real estate. They have to answer them on mobile phones. They have to answer them on voice. So answer those questions. Don't be afraid to do that. I think we're afraid to give things away. And yes, maybe you wanna stay away from what day of scroll appreciation day because the answer's so simple. But a lot of answers. We saw a ton of how questions. A ton of how to questions in our dataset. These are questions that take time to answer. And if you see three bullet points, you're still gonna click. Cause I can't tell you how to build a deck in three bullet points. I can't tell you how to make paella in three bullet points. People are gonna click through. They're gonna be interested. They're gonna see a cent trail. They're gonna try and understand more. And those are going to be qualified good visitors. So don't be afraid to answer questions. Finally, voice search is here. It's not incredibly prevalent. It's still new. It's still evolving. But the good news is you already know this. You know how to rank an organic. And if you can rank an organic, you can rank in a featured snippet. If you can rank in a featured snippet, you can appear in Google voice search. So you already know how to do this. Don't be afraid of it. Play around with it. Have fun with it. Imagine the first one you can get if you already have it and you show your boss on Google Home. Look, we rank for this. I'm a voice expert. Go do that. It's true, you know? So try it out. It's fun. That's all I got. Thank you. Thank you.