 I feel like there's this weird interesting spectrum on the web where on one end is content and the other is app, but it blends at a certain point. It does blend. And it's that blending part that I find is always interesting, where I'd hate first to be building something in a way that then search can't understand that piece. Right, because you want to find the content that lives inside the applications and you want to still find the applications, right? Right. Hello and welcome to SEO Mythbusting. Today I have Deon Alma with me. What do you work on? I work in the web developer ecosystem group here at Google. And so that's within the Chrome organization, but we work across Chrome and search and ads and all of the different areas of Google that care about the web. Personally, my passion is about the ecosystem, hence the name. So it's about looking at the health of the web and what are we doing to really help developers, but even beyond developers, hence the SEO world and the like, to make sure that people being successful, right? So that could be successful in building amazing user experiences, but it's also successful in actually being able to pay the bills. And so they can keep making those amazing user experiences. That is great. So when I have you here today, I would love to talk about SEO and the web platform going forward. So we want to look into what are the new use cases, what are the things that are coming up next, and what are the things that we from the search side should be preparing for, let's say? Do you have any highlights that you want to talk about? Yeah, so I gave a keynote at the Chrome Dev Summit in November 2018. And it's kind of going through sharing all of these great, amazing web platform updates and tooling updates. In the back of my mind, you know, you're always thinking about like, are we getting closer to search and helping the world or are we causing more problems going on? I'm actually really curious to get some of your thoughts on some of these items. Amazing. Do you have any specific things that you would like to address there? Like, is there any showcases or anything that we we should look into from a search perspective? Yeah, so one case study actually that we showed was Pinterest. And it really got me thinking because, you know, they came along and their growth team built this new PWA. That's fantastic. It was, it's amazing. Like the performance is top-notch. It's a great experience. So they've got this great PWA that's really exciting. Then they've got their kind of regular desktop site. They've got, they're using AMP for different things. And so if I'm a team like that, how do I like, I've got all of these kind of different sites in a way, which is almost like the opposite of the promise of the web at times, where we like build it once and it was over. So like, which of these, how do, what's the one that I want to give to search? Like, how do I think about that? That's a really good question. So I think generally speaking, if you have the same content in multiple ways that you present them to the user, find out the one that is most likely to apply to as many people as possible. So if you have one specific thing for, let's say, like sign up users to actually remix it or something, that might not be the thing that you want to surface to search because me coming into it from outside and not having ever used it, and I might not even be able to use the feature. And then it's like, why am I getting here? So try to think of what are people typing into a search bar? What is the purpose that they're coming from? I want to see this image board that Dion has created for Chrome Dev Summit. So I might type in like image board dev summit and then I want to get the PWA probably because it loads fastest and it is available even if I'm offline and stuff like that. So give me that. But also link to the M version because then we can also show the M version next to it. So it is not really a ranking signal. So we might see like different rankings there depending on the purpose and the user characteristics what they type into the search bar. But if you link them, then we can make the decision for you so you don't have to worry about that kind of stuff. If you then have a desktop site as well, that is great as well. If it's responsive, that's even better. If it's not responsive, then give us an alternate link so that we can decide, okay, so this is the Google crawler that uses mobile first. So we're probably going to highlight the PWA experience over the desktop site versus if someone's using the desktop site, we might give them the other version rather than that. But we are trying to make the right decisions for your user so that you don't have to worry about these things. But strategically speaking, give us a canonical that makes sense to as many people as possible so that we can surface the related information based on what people are looking for rather than any differences in the presentation. Got it. Cool. The other thing that comes up a lot is integrating our tools together and sort of having it in silos. So we've already been working. We've got Lighthouse over here. We have PageSpeed Insights. We've got SearchCom. We've got all of these different pieces. We announced the Chrome Dev Summit, the unification of some of these. So PageSpeed Insights now uses Lighthouse and the like. And that's super exciting. Wayfair and other company built this performance portal to kind of help the team make sure that they don't regress because we see that a large percentage of the times people will like do an investment, have a quick burst, get their performance good and then it will slowly go down. So they have this awesome performance culture that's trying to fix that. And I was just looking at that and thinking about it's yet another example. Like here's all the performance stuff. Couldn't we be surfacing the kind of SEO pieces into that too? How do we bridge these worlds together so both sides can kind of see the other things that are going on? So Lighthouse gets more SEO edits, which is great I think. That probably at some point feeds into PageSpeed Insights as well. Search Console integrates some of these metrics as well because performance is also important for search and sustainability. We want to like reward the content that gets the user quicker into the content rather than having to wait forever until something loads especially on mobile networks. But for external tools and external companies who want to dig into the data I think Search Console is a great way of already doing that in a packaged way. But we are working on ways of integrating external parties and external content providers and platforms to get the data that we are collecting already for Search Console. So for instance if you know one of the many larger content management systems or platforms that exist out there where people are creating content we don't want them to have to specifically go into Search Console and now like deal with the interface they are used to and then deal with something new. That's why we are bringing the data into these platforms and eventually hopefully once we have gathered enough information to understand how the data is used and what data is necessary and meaningful to external parties we will open these interfaces that will literally be an API to integrate the data from Search Console which gives you information as like click-through rate how many impressions you get from Search how many clicks you get from Search all this kind of thing how many of your pages are indexed actually because not all of it might be indexed there might be an issue you might have a markup problem right we want to surface this where people already are that can be an IDE integration that can be something like Wayfair does but at this point in time we are too early on to open that up to the general public but eventually that's going to happen. That's so exciting and some people are talking about how like they want to understand the the unknown things like they want to about like what's if I do this that or the other how does it affect my customer lifetime value and these things and we keep running into something on the performance side where they'll do this work and they'll make something that's a lot better but their metrics actually show in the field that it's slower and they're like what's going on like in the lab it's faster it's obviously faster I can see it so much when it has it slower and then it's because the reach that they got just totally changed so these people are like 2G networks yes in emerging markets that can actually really use this now then you're comparing Apple's oranges right so you've got all of these new customers and so you know the mean metric isn't the thing that matters here right now you've reached this new audience you get more revenue and the like and so like how do we know like with the search side and discovery like what are the things the next things I should wish like prioritized and the like right right I think a bunch of stuff that search console gives you already is kind of handy for that you see how certain content works and how certain other content might not work so great and that can have technical difficulty sorry technical reasons or it can be actually content reasons so what we are trying to do is we're trying to also broaden the the perspective a little bit I mean you've you said that like oh yeah we are we're in this metric right now we want to improve performance but actually has degraded and it's like well it has but it's not a bad thing because what actually happened is we opened up to more people right that happens with search as well if you look at it from a more holistic perspective so if you're not just looking at technology but also at your content strategy and what I would love to see happening is that if we bring this data to where people already are that people become more aware of it that maybe you should also look at your content and the way you present it and the way you make make sure that you're you're talking to the person on the other end who's trying to solve a purpose right yeah you're looking for something specific rather than necessarily a specific product it's like I I need this task to be gone and done and dusted and how do I do this so I think bringing the data to other people will help and I think what developers should look at is how their content is performing in search engines as well so how often do I show up in search results where in the search results that's do I rank how many clicks do I get from search results if the content looks like this versus like that so that's something that I hope to bring to more people as well cool yeah that makes sense so web components has been another big topic and we released an early version of a new one that I'm super excited about called Virtual Scroller and you know basically for years web developers have been saying like give me a UI table view on the web that can perform and scales with the DOM and so we feel like we're going to be able to deliver that but then what's that going to mean in the SEO world is search going to be able to understand this if it's all virtual and like as we kind of push the bar here for performance and user experience how do we make sure we don't leave behind I think what a lot of people outside of our sphere don't realize is that we are working hand in hand it's not like SEO versus or search versus the browser teams we are trying to figure out how we can work with a web platform rather than against it we are using the web platform yeah we are basically running a browser if you wish so we are actually looking into things like web components and Virtual Scroller specifically I think it's a fantastic thing because we have literally just released documentation for lazy loading or infinite scroll whatever you want to call it basically you know the drill like you have a page with a bazillion images or product listings you can't have them all on the page right away right you want to make sure that as the user goes through the content it comes back how often did it happen to you that someone send you a link it's just like have you seen this product and then you go to the link and you see a lot of things but not that product right and then they go oh yeah you have to scroll down for a few minutes right that's not good right web components are another thing that we are like really careful about and we have some guidance out there already and we expect to give more guidance as well because if you build new web components from scratch they look like they're semantic but they're not necessarily semantic in the strongest sense right if it's a button that just happens to look like a button but really is a diff underneath you lose a lot of the accessibility features and so does search also loses a bunch of the semantics behind it so right now we are recommending putting the content the meaty bits of your content like the actual well what it is about put that into the light DOM and then do use the shadow DOM and your components for presentation virtual scholar is a specific interesting case where not only is presentation but also the way of the how the content gets loaded and the content behaves so we are working on figuring out how to make that consistent with search so that you don't have to worry about using these new technologies because we're really excited about them as well yeah no that's great how do you feel about I keep longing for a world in the future where we add with web components these different semantics kind of high level semantics that search can understand but also assistance can understand so we can get further into transactions and life we've got a notion of you know I really want to be able to go onto a commerce experience and so I want to buy a t-shirt for Martin and everything is broken in a way it knows your t-shirt size even if I don't you know what I mean and there's a person tag and all these things and then you get a great UI from it too because it's standardized but then search could use that and assistant could do full transactions if we do that and it feels like that would be a way for assistance to kind of reach the long tail versus trying to go to everyone who's building a website there's got a lot of work to do as it is and we're asking them to modernize this and picture performance and do this for discovery and also build this thing for the assistant that is true do you feel like we'll ever get there? I think we get there probably quicker than we think because conveniently we have already pushed the idea of using structured data to expose like a bunch of information on your pages in a way that not only our crawlers can understand them but also anyone else who basically parses this information and we're using schema.org for that so we use like a standardized format rather than just a proprietary thing so I hope that gets more adoption and we see it both picked up by developers and by these assistance systems because then you can use the web really powerfully because you have all the semantic data that you can just pull together and then package in such an experience as you explained like the full end-to-end commerce for instance cool nice man thank you so much thank you mountain thank you see developer and SEO don't have to be contradicting each other also the ice tea is pretty not good pretty watery but it has that fantastic right color like you picked the exact correct problem