INBOUND 2016: Michael King "The Pragmatic Future of Search"

Okay, so despite the fact that that gentleman has the same jacket on as me right now, I am not Michael Strahan. Mike King. So very excited to be here. And may I make some noise? That's gonna be a crowd that isn't necessarily just waiting to get Dr. So I've got a lot of slides. As I tend to do, I would encourage you to just grab this URL rather than trying to feverishly take notes. But if you do want to take notes feverishly, that's also cool too. I've tweeted this as well to encourage you to follow me, because I'm going to leave this slide. So around this time every year, people always ask me, mike, what's the future of search? I guess I'm a thought leader or whatever they call that. And I always get like this again. Like, do I really have to answer this question? Okay, so the business answer is like this. The future of search is you are such a busy person that has 50 tabs open at any point, and it's likely that you put something in your shopping cart, and then you get up from your desk and you forget about it. And then later on after you've left the office, you're walking down the street and you get some sort of indicator on your phone that says, hey, that those shoes that you were thinking about buying are in that store 100ft to your right. To me, that's the future. Now, that's a pragmatic answer for the business concept. My real answer looks a lot more like this. So how can I help you? Oh, it's just more that everything just feels disorganized, that's all. You mind if I look through your hard drive? Oh, okay.

Okay, let's start with your emails.

You have several thousand emails regarding LA Weekly, but it looks like you haven't.

Worked there in many years.

Oh, yeah, I think I was just saving those. Cause I thought maybe I wrote something funny in some of them. Yeah, there are some funny ones. I'd say There are about 86 that we should save. We can delete the rest. Okay. Okay. Can we move forward? Yeah, let's do that. Okay, so before we address your organizational method. So that's from the movie Hearts. That's a science fiction film from 2013 where, you know, this guy has his personal assistant, which is basically built off artificial intelligence, and eventually falls in love with her, and all these crazy things happen. But the bottom line is the future of search is a predictive assistant. So when you see something like the HubSpot inbound Agenda that's completely overwhelming. It can figure out for you. Truthfully, I don't know how any of y' all figured out where to go today, but nonetheless, I appreciate that you're here. You could have been at this young session. Now, does that sound farfetched to you? Maybe it does. You're like, oh, you know, no way our technology is that far at this point. Let's talk about the things that you would need to have in order to accomplish this. So you need to have an understanding of a user's location. You need to understand their context persistently, you know, no matter where they go. You need to have a big index of the world's information. You need an always on device. You need an understanding of natural language processing, and you need machine learning, plus a lot of computing. Huh. Another thing you need is speed, because nobody's going to talk to something and then wait for five minutes for an answer. And you also need a good user experience, one that is no different than interacting with the world as you see it right now. So let's talk about Google. Google's been tracking your location minute to minute for the last five to 10 years. You can actually log in and see all the places you've been on the map that Google's tracking. Google leverages what they call database of affinity to model people. Back in 2011, they consolidated their privacy policy so all the tools in the Google ecosystem could speak to each other. And now you have these things you can target in paid search called infinity seconds. So they're already modeling people, they're developing these persistent user content. Then you have the idea that their whole fundamental thing that they do is store and make usable the world's information. They have 15 exabytes. I wouldn't even know how to read that number if I didn't look up what an exabyte was. But a ton of data on everything. And perhaps you may have heard of something called Android. Well, Android is on 87.5% of phones worldwide. Natural language processing is something that Google has mastered. You know, it's fundamentally what powers search and understanding of what is being said in content and also what's being said in the queries to surface that content. They've been nice enough to open source their approach to NLP so you can be good at it too. And then Google's been working on quantum computing, so they're trying to beat Moore's law, so to speak, kind of accelerate it and make things even faster. So they have a computer that they've developed that's hundred million times faster than the computer that you have at home. They reposition themselves as a machine learning company. So I was@GoogleIO last summer and that was the main party line. They're like, we're going to be all about machine learning. It's going to be in all of our products. You're seeing new and new messaging products that are built on machine learning as well. And again, they open source that so you guys can get better at it as well. Now, speed is one of their other obsessions. From Google Fiber to building Chrome, which is a faster browser than anything else. One of the fundamental things they've done is make things faster. As far as user experience. Well, Google used to suck at design. We all know that. Like, we used to look at Google and be like, ew, this thing gets me where I want to go. But they gotten a lot more focused on that and they developed what they call a visual design language called Material, which is essentially a UX framework for building out websites. And that's why you see that everything across the Google ecosystem kind of looks the same. And a lot more people are starting to use this as well in their own hands. So basically, Google is already ready to be the Star Trek computer. They've said for a long time that was their goal, to be that ambient search interface where you can just talk to it and see if any answers. The first version of that that we saw was Google Now. Google now is a predictive service appliance, so to speak. And so what it would do is, would give you, you know, based on the data that they're collecting on you, like, oh, you go from this place to this place every day. That must be home, that must be work. So they will say, oh, 8:35, you need to leave by 9:06, otherwise you're not going to make it there on time. Oh, you're always looking up the Boston Red Sox scores, so maybe you're a fan of them and they surface those for you. So they use a lot of implicit queries, or, excuse me, implicit context based on who you are and what you do to surface some of this information. And then they had Google Assistant. So this is the real V1 of the star Trek computer and that you just talk to it, it's in your phone, you ask it questions and it surfaces those answers for you. So what Google Home is, it's Google Assistant in a box. You don't need your phone or you need your phone to set it up, but beyond that, you don't need anything else. But that device. So I got it and I was pretty distraught going back to the hotel the other day. So I asked her the question, okay, who won?

Donald Trump won the presidency. Right now Donald Trump has 279 electoral votes, while Hillary Clinton has 228.

Okay, so what do we do now?

Sorry, I can't help with that yet.

They can't answer everything quite yet, but it's still a pretty advanced problem. And so it kind of, kind of begs the question how far we've come with search. Let's talk about it. So Google started out as 10 blue links. And you know, it wasn't so sophisticated looking, but it was better than Lycos and AltaVista, web crawler and all those other things I used to use in my teens. And so they got a little more advanced recently, started to add more media to the mix. They called it universal search because they realized this answers people's queries better by surfacing these other types of media. And then they started to get faster because before this caffeine update, Google had to index things and then, or some crawl things and then index things and then start surfacing them. There could be a huge lag time between when something was posted on the web and when it was surfaced in Google. That doesn't work for us. You know, if we see something as a good article on cnn, we expect it to be in the index right away. So the caffeine update is what sped that up. Then they started rolling out personalized. So based on the things that you're doing across the Google ecosystem, the things you click on, the things you search any other time, they then surface things specific to your user contact. And then they rolled out Google Instant because they figured out that people were taking a long time to put their queries in. So they would suggest them based on essentially collaborative coding that saves people two to five seconds per search, which is pretty awesome when you think about how often you search for things. Then they started figuring out, you know, I mean, there are other updates between these, but these are the ones I think are important to the future. So there's a Panda algorithm because they realized that people were just making really bad content. Sorry, guys. Check, check. One, two. Wow. Okay. Sorry, folks. In an overflow room. There's actually some seats up here if y' all want to come up here. But anyway, so as far as Panda, Panda is Google going after the content that doesn't have utility for the user. For a long time people thought like, oh, we just make content for search engines. Like you have to write for search engines. Well, what Panda did was basically kill that. So what we call thin content is what that would effectively remove from the index. And then after they solved that, after several iterations of that and, and then they eventually added it to the core algorithm, they kind of moved towards solving link spam, which is what the Penguin update was for. And the same cycle happened where they had to update it and refresh it irrespective of the main algorithm. And then after that they added it to the core algorithm where it updates in real time. Structured markup is another fundamentally important thing that has been added to what Google will then surface. So the three search engines got together and they decided, you know, we're going to come up with this vocabulary called schema.org so that you can mark up your pages and they can extract more specific information on the, you know, the entities, we call them, the person places or things that are being talked about on the page. And then they had hummingbird, which to boil it down into like a one liner, it's basically them having a better understanding of scenarios. So if I say Barack Obama, president of the United States, you know, Michelle Obama's husband, they understand that's the same person. So you don't have to necessarily put the word Barack obama on there 59 times in order to rank for that keyword. And then rankbrain. Rankbrain is kind of the other side of that in that Hummingbird is having a better understanding of the synonymy of content. Rankbrain is a machine learning algorithm that allows them to get better understanding of your queries. So the thing about Google is that people are typing in new queries every day and it's very difficult to understand queries that are like 14 words long. It also has a lot of implications with regard to this vocal search. I call it vocal search because I like that it runs with mobile, social, local. I don't like saying voice search. But anyway, this has implications on that because of the fact that they have to understand these queries and it's very difficult to understand those longer queries. And just last week Google announced that they moved to a mobile first index. If you don't take anything away from me today, or if you only take one thing away from me today, this is incredibly important. This means that they're crawling mobile first now. So if you don't have your content available in the mobile context, you may not be updated anymore. And then because of that you may not rank. So types of search that you should keep top of mind because they are more emerging things. Spotlight search and search in Safari. As you notice here, if you type in a query like this, they're not going to Google, they're not going to a search engine. They're trying to bypass search engines altogether. And because this is like the natural place that people are searching, Apple has now inserted themselves so that they can essentially cut out Google. Same thing is happening here with other types of suggested website searches that they give you. So notice that they have a suggested website before they even get to the Google search terms. And more often than not, I mean, I don't know about you guys, but I accidentally click on that because I think it's a Google result. So again, Apple is trying to cut Google out of the picture. There's things like smart mirrors where there's all these essentially the same way that the dashboard that you would have on your desktop computer, you can have that type of stuff for your mirror. And it's interesting there because I suspect that there's going to be a lot of advertising in these types of things. So in New York, I'm not sure about Boston because I don't come here that often. We have in office buildings there's always like the TV and the elevator when it has ads and things like that. I think of smart mirrors very similar to that. So I suspect that we're going to see more pay to play placements in these than just regular natural search. And then I suspect there's going to be in car search. So think about back in the day when you would do directions on MapQuest and they would always take you slightly out of the way because they wanted you to go buy as many McDonald's as possible. I suspect that Tesla's going to realize there's opportunities around search there as well. So then you want to talk about App Store search because App Store optimization is something that's really important if you have an app, of course, and really that's just old school SEO because it's largely about on page content. Do you have your keywords there? And then the engagement metrics, are you getting downloads? Are people looking at your page? Are they leaving reviews? So it's not difficult, it's just, you know, make sure your on page is right and then use something like Sensor Tower to do research around the keywords that you should be using and also the rankings of other apps and such. And then just buy some ads so you can get more people to download your app and also in your app ask for reviews. So that one's simple, it's not like doing modern SEO on Google. It's really simple. And then brings me to vocal search. So this is the primary thing that I think moving forward or the primary mode of search moving forward. I think Google is trying to be more a part of your everyday life, not just trying to have you on your phone looking for things or on your desktop. They want you to basically just talk to them at all times. Okay, Google, how do I get girls?

Here's a summary from Instructables. One, step one know yourself. This is actually a very important step two. Step two, express yourself. Three. Step three get over yourself. Four. Step four finally. And five step five, asking her out.

Okay, that's actually kind of reasonable. So the thing about voice search or vocal search is that there's only one answer. They're not going to sit there and have the Google device. I don't know if it has a name. It's not going to sit there and read 10 results to you. It's only going to read one answer to you. And Dr. Pete Famas did a great post on this today. Like how to rank in Google home. Really what it comes down to is getting these featured snippets. So that answer that was just read to me is directly right there in the serp. And there's a lot of, there's a lot of data out there about how to do this. I'm going to talk about it a little more later, but the bottom line is like, you need to be going after this if you want to be the number one person there. So voice search is something that users are using simply because it's convenient. All this data suggests that people are like, oh, it takes too long to take my phone out or it just isn't as easy as saying it out loud. So I suspect that over time, younger generations are going to use this far more than perhaps the older ones. These other assistants are interesting as well because again, they're also trying to cut Google out of the cycle. Think about Alexa and the smaller version that HubSpot just gave me. Thank you very much. They are cutting Google out of the E commerce loop entirely. What they're saying is like, oh, you want to order something directly from Amazon? Cool. Because we don't need you to search for it anymore, just ask me for it. So I suspect that as technology becomes better and bigger and more complex, search is going to be how we interface with it. It's not going to be so much about these crazy futuristic looking interfaces. It's going to be more like this. You standing in Your house, just talking to your computer or talking to Google Home. And it's going to react to that. It's not going to be like Minority Report where it's all up in your face. It's just going to be a part of what you do normally. So let's talk about some science behind how this works. The technical stuff that I enjoy the most. So part of this, or what this is called, is multimodal search, where Google, or all search engines really, but who really cares about those other ones, are leveraging both the explicit query that a user types in or gives to the. The search engine and the context on the user. So their location, things that they've looked at before, other things that they've engaged with, they put those two aspects together to determine what it is that you need. So Google is far better at understanding the context of the user, which is why I'm far more interested in Google Home than I am in Alexa, because Google's been doing this for so long. They have that natural language processing that works with this. I have a employee who, you know, English is a second language. He speaks well enough, but I didn't think that Google Home was going to be able to understand him. And it understood him perfectly. Okay, Google, how tall is the Empire State Building?

The Empire state building is 1,250ft tall.

Okay, Google, where is it?

The address for Empire State Building is 350 Fifth Avenue, New York.

Okay, Google, how far is it from here?

You're 211 miles away from Empire State Building by car.

Okay, Google, how long if I walk?

Sorry, I'm not sure how to help with that yet. I'm still learning.

Okay, Google, how long would it take for me to walk to the Empire State Building.

On foot? The trip to Empire state building is 2 days and 23 hours.

So the interesting thing about that is it understood the context. The first time I said, how tall is the Empire State Building? After that I said, how far is it from here? So I don't have to be explicit anymore. They understand through my series of searches what I'm talking about. And I've seen some great work that Will Critchlow presented, I think, even here last year, where he talked about, you know, users walking up to a landmark and saying, you know, what is this church? And Google can answer that question because of those implicit contexts. And so the great patent, or maybe just a white paper, I'm not sure if it's a patent, but they talk about how they model users for. For these personal assistants. And it's largely about understanding the queries and the things that the user had clicked before to understand who they are, what they want, what is it that they're likely to be interested in. And they develop these contexts over a series of months, not just hours, not just a couple searches. It's over a series of months. So the way this works is they break the query that you ask into a series of entities so they can then better understand what it is that you're asking for. So in this query it says, who was the US President when the Angels won the World Series? And you see in the diagram that they then break it down into those different entities, the World Series, the Angels, the US President. And then they have a better understanding of that and they can answer that question. So Google has come out and said that entities are the first thing that they look at when they consider a query. And by that token, which is kind of fitting the word token, you need to be thinking about that when you optimize your content. So how do you capitalize on these search opportunities? Because a lot of this stuff sounds pie in the sky to you. It's like, you know, okay, we're living in the future, but Mike, what do we do with this? I got you, don't worry about it. First thing you need to do is make hyper targeted content. And what I mean by that is content for very specific Personas and also for very specific contexts. So what I mean by that is there's a series of need states. Did we break it? It's too next level for this. Can't handle it anyway. So what's on that screen when it stops blinking is it's a series of stages like your marketing funnel. So you have, you know, awareness and research and consistent consideration and all that. And you need to align the right content for the right stages and then the right KPIs. But as long as you're thinking about the right content for those different stages, you're not going to have a big problem when things get hyper focused and essentially create filter bubbles for these very specific user types and very specific contexts. So what that may mean is you may need to make a piece of content more than once to speak to a very specific audience. Type my next slide. If I remember from my recollection and crazy memory, boom. This requires content strategy. And I don't mean content marketing, which is what a lot of you may do. I mean content strategy, thinking through how your content can be useful and usable for very specific audiences, not just making infographics and putting them out, not just throwing out ebooks, not just doing blog posts, being very specific about who you're talking to and why. And so where you should probably start is from what I call Persona driven keyword research, and I have a whole deck on that. But essentially what you need to do is on the keyword level, tie it to who is the most likely user type to come and then what is it that they're trying to accomplish with that keyword. So ultimately, I want you to build robust utility driven content for specific audiences. I also want you to use structured data. Now, there's a lot of opportunities around that, but this is an example from a venue here in New England, Connecticut, to be specifically called Toad's Place. And the things that you see up here, these are all listings for events going on at this venue. But if you then click into one of these Atmosphere, one of my favorite underground rap groups, they the source is not toadsplace.com, it's a ticket site because toadsplace has not marked up their content. And Google also has these rich cards. Like the card format is essentially everything that you're going to see across the Google ecosystem. We're already seeing it, it's been around for a couple years, but you're seeing it more and more. And you can get directly into here if you mark up your content properly. So I want you to keep an eye on this. This section of Google will tell you the things that are available as far as the different types of structured formats that they'll show. Because if you go to schema.org, there's like every single possible one, but they're not all in use by Google right now. And there's another page which shows other content types that they're allowing to be marked up as well. So keep an eye on that to see what works for your business. The feature snippet is also really important. Like we said, there's only one answer per query in vocal search. And so this is from research from a company called Stat. They're a rankings platform for unlimited rankings. My favorite rankings platform. But what they found is that a lot of these queries that return feature snippets are very question based. So they have different types of feature snippets. There's the paragraph type, the list type, the table type, and based on what question word is in these queries, we're seeing a lot more of these different types of feature snippets. So for instance, if the question is around, does like, does something happen or what have you we're seeing far more paragraph feature snippets than we are the list and the tables. Also what we're seeing, you know, in aggregate, not just specific to the question type, but we're seeing, you know, a huge uptick in the list and table types and then a drop in the paragraph types. So based on that trend, you might want to think about, you know, how can I mark up my page so that those ordered lists are there so Google can easily extract that content to make a feature snippet. Larry Kim from Wordstream also did some research around this and he found that you don't necessarily have to be in the number one position to get that feature snippet. Naturally, the number one position is the one that usually does it, but you don't necessarily have to be there. So this isn't just an authority thing. It's do you have the best answer? And then the other thing is that we're seeing the sweet spot for these snippable that's not a word snippets for feature snippets is around 45 words. So not to say that I want you to write specifically 45 words because that's what people do with marketing. They take something like that and run it into the the ground. But ideally it would be 45 words. So another thing, a more specific how to on this. Check out this post from Built Visible. They walk you through how they made these optimizations and then got those featured snippets. Be mobile friendly. Again, most important thing I can say to you is be mobile friendly. So this chart that I'm showing you right here is what I suspect was Google's test for the Mobile first index. This is a profile from a client of mine, a big media site. These are just the amp. I don't want to say amp pages because pages is in amp. Anyway, these are the accelerated Mobile Pages. And for three days the traffic spiked to these pages and then it just disappeared. And so I looked it up, a whole bunch of other media sites and the same thing happened. So I suspect that was Google testing out that Mobile first index. So the implications here are if you aren't mobile friendly, you're not going to see this traffic. The other sites will. So accelerated Mobile Pages is the initiative that Google's been pushing for to make sites load instantaneously. And the interesting thing about it is that right now it requires you to have a second URL in order or it doesn't require it, but most people have implemented it as a second URL that's paired with their desktop URL. Now, this screenshot is from a post from a Google engineer that's working on this project and he says that we don't want you to do that. What Google actually wants you to do is rewrite your sites in Canonical amp. So if you know anything about amp, you're like, well, you know, the experiences aren't so robust and what have you. Well, this site right here, the AMP project site, is written in amp. It looks like any other website that you have. So you can do it, you can do it. And one of the common complaints I have about amp, because I hate it, is that you can make your site just as fast yourself if you just wrote good code and you put it on a cdn, even if it's Google cdn, it can be just as fast. But what this engineer says is that they can't verify that it's that fast. So initially when I read this, I was like, what are you talking about? You have Google Analytics, which measures how fast things load. You have Google Page speed, Insights, Chrome itself can measure it. What are you talking about? So I think what they're saying with this is they can't necessarily absolutely know that everybody going to this page will have the same experience that Google does, because Google stuff is way faster than most people's. So with amp, they can absolutely verify that. If you follow the specification, it will be that fast. However you do it, just make sure that it's mobile friendly because otherwise you're going to miss out on a lot of opportunities. And just continuing with the idea of speed. Well, there's a lot of things that impact speed. You know, the code order is really important. If you have a developer that doesn't think through this, just the way that they lay out the code is gonna make your site load tremendously slower. So there's something called the critical rendering path. And that's what AMP basically does, is reinforces that. So think about working with developers that actually know what they're doing. There's also HTTP 2, which is a new specification that's gonna speed up everything. I'm not gonna go too in depth on it, but basically the thing that about the HTTP spec is that every image is a request and a response and that takes a long time. HTTP 2 multiplexes the stream, which means that you don't have to open a request and response every time. Once you've made that initial connection, it pushes the rest of those things over to you. So I would say read up on that. And if you really want to Dig into the technical stuff because I know some of your eyes are glazing over just off those couple things. There's a blog post I wrote. It's only 9,000 words, so make your site faster. Integrate your search channels. I still see too many people that are like, well, the paid search people, we never speak to them. And you know, they have the $3 million budget, you have 50,000, but there's more traffic from organic. It just doesn't make sense. So the bottom line is that the more you integrate these channels, the more you're going to capitalize on it. Because one, your audience, your users are not thinking, oh, am I on paid search or organic search? Like, no, it's one page to them. So there's opportunity to improve on this. And there's so many studies that just show if you're in both places and you're working well together, spending your budgets the right way, there's just an uplift. And so the bottom line for that is things like feature snippets and then also rankings in general leverage ctr. So if you're getting a higher ctr, well. And all you gotta do is pay for it, that's gonna help you. So put them together. I said that already. All right, so what tools should you use? Cause I don't want you guys to just like. I don't wanna be like Gary Vee, where I just bullshit for an hour. Y' all can tweet that. I know that guy. So what tools should you use? Well, first you want to have a device library. Despite the fact that there are emulators you know, that you have within your browser and such, it's better to have the actual device to play with so you can see what are the nuances of that device, whether it's from vocal search or whether it's from organic search, whatever it is. You want to understand better than just the emulator in Chrome how this device operates. You want to use Google Search Console. While it's not the greatest thing in the world, it's the only source of data that we get from Google as far as things like click through rate, that's the only place you can truly get that for your site. You can't really rely on the click through rate models that we have from, you know, SEO, blogs and stuff. The structured data testing tool. Because I again want you to use structured data. Test it out before you push it live. Make sure that it's validating correctly. The AMP validator. Before you push your amp page. Your AMP live. Yeah, test it and then there's a variety of tools that they have for testing. Multiple friendliness and page speed. This is like the most visually compelling one. But what's up Inbound? Why is it so slow? Chrome Dev tools this is actually built into the Chrome web browser itself. There's a variety of tools that help you understand your site speed. What are the things that are effectively slowing it down? What are some of the issues that you may be having as far as like how things are being pulled across the web? DevTools, your friend, Google Analytics, demographics and Insights. Because again, we want to understand the context of the users that are landing on your site. You want to get the data on the affinity segments that Google provides so you can see which types of users, which age ranges are landing on your site. Screaming frog SEO Spider. There's not a single day that goes by where I don't use this tool when I'm at my desk, I guess. And then they also have a log file analysis tool as well where you can get an understanding of how often Google is visiting your site. If you have. Well, search metrics is a great tool because it helps you understand a lot of content opportunity. They do a lot of really in depth analysis. Leveraging machine learning to help you get ahead of some of your competitors as far as optimizing for towards these ideas of hummingbird and such. Onpage.org if you need cloud based crawling, excuse me, you know, it's one of the better tools out there. And then if you need log file analysis in the cloud as well, Botify is a good option as well. For keyword research I would say check out Moz's Keyword Explorer. It's the most advanced tool out there for this. There's other great tools out there as well and I'm sure they're all going to try and catch up pretty quickly like Semrush and all the other things that are available. But right now Moz's Keyword Explorer is the best thing on the market. And as far as ranking stat, like I said before, you can track unlimited keywords. So I didn't introduce myself. This is me. Okay, Google, can you beatbox?

Okay, put your hands together for MC Sound File.

Yeah, yeah, yeah. Yo. Yeah, okay. It's Mike King and I got a better flow. What up with that election though? It feels just like the Brexit vote. It's whatever though. I drop it in the middle of inbound like the letter O. Use my skills as a metaphor for the Google algorithm. Cause I guess you'll never know so first and foremost, I'm not little girl's dad. She's three months old. Her name is Zora. And secondly, my name is Mike King, Full stack developer, dude marketer, Moz associate and I run an agency called iPoolRank. We're a better marketing agency. These are the things we do. SEO, content strategy, paid media, machine learning, marketing automation and measurement and optimization. And I'd love to help you out. Any questions? No questions. Oh, right here. I think that's hard to say. It really depends on what you're targeting, you know, what your goals are, what is the. It's hard to say. Yeah, but you know, our clients have anything from like 500 keywords to 5,000. So it all depends. You can step up to the mic to ask your questions as well. I was pointing at you. What's up Mike? How you doing? Chill. For, for those out there with agencies or companies and they're using WordPress as many of us are and who aren't using custom websites, are there any plug ins or tools within WordPress you'd suggest? Yeah, absolutely. So for a lot of things that we just talked about, like for instance amp, there's a variety of plug ins. There's one that is specifically made by Automattic. The one that I use is actually called Page Frog because it does not just the post posts but also the pages as well. And it also handles Facebook instant articles as well. Yoast SEO definitely use that heavily. There are a number of structured markup tools. I know that Raven Tools has a plugin for it. I don't remember what it's called, but yeah, definitely that. And then pretty much the rest of that, it would just be the other tools that I talked about. So Screaming Frog, they have a free tier where you basically download it and it can crawl up to 500 pages. So I would say start with that for your analysis. Same thing for the log file analyzer. They have a free tier for that as well. So in all the tools I showed you, they're all freemium, so you could play with them before you buy speed for WordPress. So there are a number of things out there, one of which that we just started playing with. It basically will go through all your plugins and see which ones are slowing it down the most. But aside from that, you know, you want to have a good provider, a good hosting provider for it. You want to make sure that like the code itself is written the right way. So again, Chrome Dev Tools can tell you a lot of that and also put your on the CDN anything else? Y' all just want to drink, huh? No more questions. Well, cool. Thanks, Joe.