 I'm going to introduce Matt Papola. He is going to talk about stop guessing, diagnosing and fixing WordPress performance problems. Matt is the founder of SiteDiscret, a powerful and flexible WordPress hosting platform. He has a degree in computer engineering and started off his professional career as a member of the design team for AMD's first 64-bit processor. Matt is passionate about understanding and helping people solve problems. He feels that WordPress hosts should do as much as possible to provide good performance, security, and automation so you can focus on your business while your host takes care of the rest. He enjoys travel and spent four years backpacking around the world. Matt. Thank you, let's see. You guys hear me on this? I'm sure you got it. You are up and turned on already. All right, thanks everyone for coming to my talk. Obviously, we've been talking about performance today. We've got quite a bit of content to cover so I'm going to go pretty fast. If I'm lucky, I'll have some time to be in it for questions but not confine me out there and I'll post with all the nice little padded seats in the back there. All right, let's keep going. All right, so first, a little bit about me. This is probably mostly repeat from the intro. I grew up in Arizona and I went to school down at the University of Arizona in Tucson. So I'm another four years down there. So I think we can be engineering and work on. And I moved to Austin, Texas. I lived there for a couple of years and worked on my 21st for the first 64-bit processor and that's 64-bit and I was one of the least up there at the top. And after that, I left and I took a backpacking trip and went around the world and was supposed to be for a year and a half and ended up being three and a half years and I only made it a half way. I got involved with WordPress in 2011 when I started doing some plugin development and back end server and service development for a client and that project ended up being successful and went well. A few years later, they had hundreds of websites on WordPress, a bunch of clients, and they were having server performance and security issues. So by then, I ended up starting what has become the site network and we're now hosting hundreds of clients, thousands of websites on our platforms. All right, now we're going to have today's talk. I probably should stand back so I'm not turning my head around and get back to work. We'll talk about first life performance matters and how do we actually measure it. Talk about hosting and what effective has on it. Then we'll talk about your server response time. We'll talk about the full page load when someone gives a page. Just a few words on caching and then because I'm crazy and we have a little bit of time to do a quick demo at the end. So first, why does performance matter? So, I'm not used to these guys but I think they have the best marketing message at least that kind of sums it up. So how long does it take to lose a customer? 2.7 seconds, I say. That's the amount of time before someone visiting your website will stop and then start to look for your competitors. So that's one way and just summing it up. Convergence and page abandonment. So if you've got many customers, then you could be losing business if your site's not fast enough. It waits your time while you're developing and working on your site, especially at the back end and the word for this ad is slow. Perception on your visitors, your customers, when they come to your site, they have to wait for the page to load for too long and then that's going to leave a negative impression on them. Your Google ranking, if your site is extra slow that has some effect on that, no one probably decides if people at Google know exactly how much that is, but it's been published and said that's a vacuum. Your client's satisfaction, if you're an agency or you're building a lot of WordPress sites for different people, then they're more likely to come back for a new site or redesign or refer you to other clients and build them a fast website. And then I'd like to make the whole claim that speed equals happiness, and that's one of the biggest reasons so. So kind of drive that point home, think about this for a second. All right, so let's do a quick survey here of a show of raising hands. How many of you have worked on or developed on a cycle slow or became slow? All right, everyone look around and see how many people are raising their hands. Well, we've got some more questions. How many have spent time trying to speed up a slow site? Have given up on trying to speed up a slow site? Have switched hosting providers because their slides have switched off plug-ins one at a time trying to speed up a site. A lot of hands on all these. How many have a slow site now? All right, so this talk hopefully will be good for a follow-up. So I have a hypothesis, and that is why do developers and designers and site owners speed up or fix their slow sites? Well, that's because slow is comfortable and optimization is often risky and uncertain. You don't know how much. You don't know how to make the site faster, and you don't know if or how much faster your site can be. And so it's not that you don't want your site to be faster, it's that you just don't make it a priority. So I want to speak a little bit about priority. And I'm going to define priority as value divided by effort. And I started off by talking about the value of performance and speed. So what about effort? If effort goes down, then priority goes up. So my goal today is to show you how to minimize that effort. So first, before we try to fix performance, we need to figure out how to measure it and understand it. So some basics about paint speed loading. So I've created this diagram here. It's kind of a simplified waterfall view of what happens when your browser loads a page and a browser loads a page up and displays it to you. The first thing is you'll put your domain in the address bar type it up, and then your browser needs to do a DNS query. Look out, find out where that site lives. And then it'll make a connection to the server. And then it'll set up a secure connection with your site's running in HTTPS. And these days, all your sites should be. It'll then send that request and wait for the server to respond back and send back the initial page content. It's just one text file with HTML in it. And once the browser receives that, it can start doing a bunch of things in parallel. It can start parsing that HTML. It'll find out what else it needs to load, such as JavaScript, CSS files, fonts, images. It'll start making requests for those. And then it'll actually do the hard work of rendering and painting everything on the screen so you'll see the web page show on your browser. So this first part here, actually just that part right there is often called the time to first play. And that is from WordPress. That's WordPress running PHP and talking to the MySQL database at the end before it sends something back. How fast should that time to first play be? So for work, obviously, the faster the better. And I didn't talk about page caching yet, but we want to start by not covering up any performance issues. So we want to see how fast that time to first play is without any page caching turned on. So I consider very fast the 50 to 150 mils seconds, fast, 150 to 300, good, under half a second, average, slow, very slow. And some sites that I've seen have taken 15 to 30 seconds, actually even 45 seconds to return back that page back to the browser. So if you're sitting there staring at a blank screen for 30, 15, 30, 45 seconds sometimes. So for WordPress, I think under half a second has a pretty good goal, at least for uncached speed. So now how do we test that and see what your time to first play is? So one of my favorite tools is this Chrome extension. It's called page performance. And there's a link to it there, too. And what this does is it just sits in your Chrome toolbar there. And every time you go to page, no matter what page it is, it's got this little icon available here at the top. You can click on it and drop down. And we'll show you what the time to first play was for that page. Another way is you can actually use Chrome itself and open up the developer tools. And you can jump down to a little three-world button top there, down here, and develop the tools. And then you want to select the network tab. And you will need to reload the page if you do this. The network tab doesn't populate unless the developer tools are open. So you have the page, open this up, reload, and then you can click, kind of, hover on a little bar here, and it'll tell you down here what your time to first play is. And then one of my favorite tools is, and this is something you can share with other people, too, after you run the test, is a web page test. And so that's a web page test ad where if you go there, it'll see a screen like this. I recommend actually logging in with your Google account or something, because then you don't seem to need to do the capture thing, and identify street lights. And if you, when you do this, though, first, probably you want to set some settings here. So we're going to try and test the best case speed for your website. So first of all, pick a test location that is near your server, so wherever your server is actually located. Then set the connection speed to native, which means it's not going to try to slow it, but loading down, pretending you're on a slower network. Set the number of tests to one, but then click the repeat view, and the first to repeat view. And what that will do you is it'll run the test twice, but once showing the effects of browser caching that someone were a visitor site, and then they continue browsing it, or they come back later. Usually that second goal will be faster. And then click the capture video button, and this allows to see a film strip where you will actually be able to see what people are seeing in with a timeline as your site is loading. So once you run the test, you'll get a page back that looks like this. And we're generally trying to optimize the time of the first byte and speed index, and that's something you'll just find on the web page test. So we're going to ignore the load time, and I'm going to tell you why in a second. So the speed index, they define that as the time at which the visible parts of the page are displayed, basically an average of that. So web page test has a film strip view, and it'll show you over time with a little label here as far as how much time is passed, how much of the page is shown up at that time. And so the lower the better on this, and why does this matter? Because this is what people are actually seeing on your site. There's also a waterfall view. I won't go into this, but it's a diagram that's similar to the one I showed earlier in the presentation. But it's specific to the page you actually just tested, and it has a lot more detail for what's going on. This is the film strip view I talked about. We can see now here in this test here on the WordCamp website that the user was looking at a rent link screen for 0.7 seconds or so. Not very long at all. At 0.8 seconds, started to fill in. The image kind of showed up just a teeny bit after. And then the menus and the cookie bar finally filled in at 1.2 seconds. But right here, at less than a second, we can see that most of the web page is already there. So that's what people are seeing, that's what they're waiting for. They can start interacting with it. People are going to see things before they even touch it anyway. So you want to just get enough up there so they can see it. And even though the load time was a little bit longer, that the actual time that the page rendered was less. Other speed testing tools, how many people know we're having to use GT metrics or Pingdom? All right, so I don't recommend GT metrics. And so the reason for that is they list a bunch of things on the site that to me will lead you down a radical that you shouldn't be focusing on. They also require you to log in to see a lot of this other stuff that's actually interesting. So I find web page tests much more useful. So I recommend checking that out. And just to sum it up, you want to remember, is whatever tool you're using, is it showing you what your users are actually experiencing? So we don't care about numbers. We don't care about both points. We don't care about some page score, like grade or speed. We care about what people's tears like are actually seeing. All right, so let's talk about what makes up that time to first bite a limited point. And so there's two main factors in that server response time, which will often model that for your page load. First is your hosting provider. I'll talk more about that later. And next is your website implementation. And word press, that means your theme and your plugins. All right, so I've got an interactive part here. And so I forgot to announce this at the game. You've got your computer open, or you've got your mobile device here with this page right here. And you should see something like this. And hopefully, this is connected right if you start. You can put in your own domain in here. You can put in some client sites you've developed, you can put in your friend site, word press sites. And if you fill this in, it should start populating up here in real time on both your screens and this main screen up here. So what this does is we've built a tool that will go off and it'll test the time to first bite for your site and it will show up in here if you're embarrassed or you have a really slow site. Don't put it in here because everyone's going to see it. But so we can see it's starting to fill up now a little bit. And remember I said the time, ideally your time to first bite should be under half a second. Well, that's right over here. Most of the sites we've seen so far are showing up over here. We can probably be definitely a lot faster. Thank you for running that if you'd like, testing word sites. On the right here, I've kind of tried to look at the hosting providers and classify them here in terms of the average speed from slowest to fastest and if they're faster than the bottom here, that'll fall itself. So next, I want to talk about jump back and talk about WordPress hosting. So, does your hosting provider, yes it does matter. It matters a lot. So, is better hosting a silver bullet for your speed? Much of the time actually is, yes. Other times, well, you have a slow site with a lot of plug-ins, a lot going on there. It may be faster on better hosting but it might not get any faster. How do you determine if your hosting is a problem? So, well, certain hosting providers are known to be well-optimized consistently fast and not have a lot of performance issues. Other providers, it depends on the plan, box, time of day, server configuration. I don't know, stuff like that. So, although some hosting providers are just known to be consistently slower too, so. Another indicator is what web servers are host running. This is slightly more technical but I tried to keep it simplified. So, most custom managed WordPress platforms are running like IngenX and kind of a custom stack and it's a little bit more optimized. The number of those hosts are small. More common is you'll find a lot of hosts that are running Cpanel and Apache together and anyone can spin up their own hosting company using that. And the takeaway from this is that if your hosting provider has Cpanel available, it's actually an indication that your site could probably be faster. So, you could also go through a checklist and check, hey, does your server have a whole bunch of stuff on it? I'm not gonna go through these and that's kind of the point. So, it's a big long list and plugins aren't gonna help you with these. And so, if you're a server administrator or a self-hosted, you might take some notes, write those down. Otherwise, you might be thinking that. So, what's a better way, again, like we talked about the page loading. We don't necessarily care about a checklist of items. It's very important and if you're building a hosting then that's very important. We want to measure what your site, how your own site is actually performing. So, how do you know if your host is causing work or all of your speed problem? Or how do you know if your host is properly optimized for your server? Well, it's not easy to verify reliably. You don't want to go through that checklist and even that might not cover it. Well, at least not while it's on your current host. So, for another hosting provider, how? Put an exact copy of your site on another optimized WordPress host. So, there are several options for this. I've kind of rated them by how easy they are. The kind of average or harder is to use a backup or restore plugin and install plugin on your current site and host and provider. Download the backup and restore it on a new location. You can use a migration or a plugin that's kind of more integrated by hosting provider, or you can do sometimes a little bit more fully automated and I'll jump through those. So, here's some backup to restore plugins, some popular options that can be used and you can grab these from the slides later. Some hosting provider plugins that they built for migration. Here's a couple more that are provided by a service called LogVolte and for migration to Pantheon and WP Engine. They have a plugin that uses that service. And then we actually have an import that's I think faster and simpler than all these. There's more information on there. I'm not gonna show you the slides or anything about it now, but in the meantime at the end, I'm gonna be able to show it off. So, now let's say you've got your site copied over a new host. How do you test it and check to see if it's faster? Well, go back and reuse those tools that I just showed you a little bit before at the webpage test. Now we have some tools later in our live time because it's pretty short. I'll show you some of the tools that we have. We'll let you, you saw already the tool that will just test a bunch of sites and throw them all up on the screen. Now some others will let you crawl through your site and see how it's performing. So, let's talk about cost of hosting and a friend of mine came up with this term and I thought it was a great term and I didn't find it on the internet so I threw a little TM up there in the corner of it. So, I heard it here first and you all heard Bridget speak this morning, right? Everyone heard Bridget? All right. So, earlier this year Bridget decided to put up on Twitter and she decided she was gonna write a blog post about hosting her $25, you know, as an argument for managed hosting. And then she wrote this blog post and she said $25, that's what my manager costs with pressable. And some people think it's expensive. So, she argues that it's free to visit and it's isn't worth it. So, she got some funny replies and replies to this and she posted some more stuff on her blog. There's a link to it too. Some of the replies, $25 and one button donuts. I believe speaking this afternoon, Starbucks. And then I think one of the best replies, at least the one I like best is this one here. So, and Bridget, I don't mind no thoughts. So, your time is money. So, if your host is not that fast or you've got a very issues to deal with, you know, you're wasting your money that way too. That's the kind of summing up the arguments for hosting. Pretty much what the people reply to you from a blog. If it's for your business, it's important. You get a lot of things with more managed hosting than other places will upsell or it won't be bundled with the office upgrade. And again, fine, I guess your time worth it. I've actually gone through from my experience from onboarding and migrating and testing a bunch of the sites of the last eight years. I kind of went through and turned on my host and providers here, but this is just for performance from what I've seen. And so, I put up this slide last week at Riverside and then afterwards I got a comment, like I got this comment the way. My site's pretty fast but I'm in a slow column. I think my site's fast, you know. I thought, that's a good question. So, I tried to come up with some replies to that and things to consider are, well, yeah, ignorance is bliss and your site might be loading two seconds but you just haven't seen it loaded in one second. Page caching, is your site loading quickly because of page caching? And what if you're logged in or what if you're on WordPress admin in the back end? Is it still this fast? And then, yes, maybe it's good enough. If you and your customers are happy, then great. It's fast enough. And then mobile, always consider a mobile. Something that loads fast under desktop might be a little bit slower on a phone over a mobile network. So, test it. You can use web page test to set the connection speed a little bit lower to see how long it will take to put that page to load if you're on a 4D, 3D network, something like that. So, you might be thinking, wow, I just spent 20 slides talking about posting. So, yes, it is not important. And so, just this last week, I'll just sum up all the underlying parts here. I posted a reply to a question on a popular Facebook group with 11,000 members on it. And this has had a little bit of a dialogue with one of the guys on there. So, he tried the other suggestions and the results were not worth it. I spent five days, oh, sorry. And he said, the results were not worth it, but now I believe that the web host is the key. I spent five days trying, and my efforts did not work. Thanks, Mr. Pricell. So, he was able to speed up his site significantly just by switching hosts after spending a month of time with other options. All right, enough thought, hosting. Let's talk about that server response time. And let's say you've already moved to an optimized host and your time to first light is, because that is controlled mostly by your server and your hosting, let's say that's still slow. Well, that's to most people, that's a black box. And that's WordPress and PHP and running, PHP and MySQL remain and you can't look inside of it. Well, today, let's take a look inside. So, I'm gonna show you some tools to do that. But first, if you're diagnosing your site and trying to optimize it, then you're gonna go through several steps. First, you're gonna generate or wait for some traffic to the site when it loaded up or if it's live in production and you just let some people hit it. And then you're gonna review, you're gonna need to have profiling enabled and then you're gonna review those results and we're gonna use New Relic for that today. And then, you'll make a change and you'll repeat. And the difference between doing this and what most people are doing is instead of guessing for this step here, we're gonna make step three based on what we learned in step two. We're gonna have real good data that's gonna tell us what exactly is going on with WordPress and why it won't be slow. So, let's check out New Relic. So, New Relic is an application performance monitoring tool and there's a free version available. The pro version is the most useful. You do need root level access to install it on your server but it also is provided by several WordPress hosting providers. And I have personally diagnosed hundreds of performance issues across lots of sites in very little time. Thanks to New Relic. And so, once we jump into New Relic, we're gonna see an overview tab here. I tried to turn that off but obviously it wasn't successful. And it's interesting. We'll see how long the average server response time is and how much time to spend on PHP and MySQL but that's not really all that helpful. It's nice to know but what is really valuable is what New Relic added a couple years ago and they added a feature which is WordPress specific and we'll actually break down WordPress and show you what's going on. So, if we click on that page under the plugins and themes, it's gonna show us how much time is actually being spent for each of the themes or each of the plugins or your theme. So, for this example site that we tested out, we found out this A to B testing plugin here was taking 50% of the response time up and it turned out for this particular site they didn't actually need that plugin. They were just able to turn that plugin off and the site got twice as fast as it did. Here's another example. If you need more information, you need to dig down in New Relic as something called transactions views and that's for a single request that you can dig down in and find out where it's spent time. So, I think this screenshot here because this is a request that actually finished relatively quickly but then there was a hook that ran at the end and after the generate of the whole page it was about to finish up. It went off and ran this hook that synced some data to Jetpack and that took it an extra two seconds. So, what would have been a 250 millisecond response turned into a two and a half seconds, 10 times as slow as it should have been. Here's some examples, either rather old and just because a plugin shows up here as a bad one, could be configuration as a starting point. But for this, whatever module was enabled with Jetpack here and we just turned off Jetpack and then boom, you can see right away the performance of those data improves significantly. Sometimes the WordPress admin could be slow and again, this is old and might not be applicable but in this case, the US SEO plugin was making all of the admin pages a bit slower. So, before New Relic, this is what I felt like trying to figure out problems when I very first started hosting sites and trying to figure out the issues. So, right away I set up New Relic and once I did that, I set it up. Right, some considerations though. It does impact performance pretty significantly, I've found. So, usually I'll turn it off when it's not being used but you can't go back in time. So, if you have some kind of issue that only happens intermittently or you need to wait for it to happen, it might need to leave it on for a while until that request comes through. It's not a number of plugins on your site but it's actually what they're doing. And a plugin shown up doesn't necessarily mean it's a bad plugin, like I said, it's a starting point. So, check the configuration settings, check out the plugin developers, or at least a new version or something like that. And then it may be fixed already or you might just need to change or turn something off with it. Your theme can also be a major factor with your performance problem too. So, that's often harder to fix so, profile it or site it early. There's some guides too, just skip over this but if you pull the slides up, several good guides on how to diagnose repressed problems with Neurope. All right, a few words about plugins, optimization plugins. Well, they can't optimize your raw site performance. What they're generally doing is just covering up performance issues. I generally don't recommend them. And you don't need them, for the most case, on better hosting. And they can actually make server response times worse because they're doing a bunch of stuff to maybe they're doing things to try and help your page load but then it's actually affecting that server response time. One possible exception is page caching and hopefully I will talk about that later. So, I actually don't use Neurope anymore. So, on our own hosting until earlier this year we did some performance upgrades and we got in our site to begin in, let's see, our servers were already a lot faster than in several places but they were not quite as fast as some of the fastest providers that I had experience with. So, we did some optimization and our sites got twice as fast and this was more than the 20 to 30% gain that I expected and sites, they're so fast now at least on our platform it's often not necessary or worth the time to optimize them more. So, kind of the moral of this story is optimized hosting was even more important than I realized and I think there's even more room in there as a group with that. Let's talk about the page loads. We talked about hosting and time to first blank and that bottleneck of getting the page to the browser but then the browser needs to load and pull everything else up. So, what affects the rest of that? So, one thing is, the HTTP2, this is a server level thing too so it needs to be an enable on your host but there's a newer version of HTTP that speeds up things quite a bit. It does need, you do need to have the HTTPs enabled to have an access to your token on your site. It's been around for several years, has some great speed benefits which I'm just gonna pull a point through but some WordPress hosts are still not running it. So, how do you check? So, just a few more things. A lot of things that are often been recommended to speed up sites such as combining, concatenating files, making, trying to have multiple, use multiple domains to have type one or like parallel requests going on aren't as important with HTTP or even necessary because of the way the protocol works. Inlining assets, other stuff like that. So, make sure your host has HTTP2 and how can you check your sites of HTTP2? There's another Chrome extension that I have installed and it'll just light up in your toolbar a different color based on whether the site you're connected to has HTTP2 enabled or not. And not when you can get that Chrome extension there. And when it's enabled, another way is you can actually go into the Chrome Dev Tools. It's a little more complicated but you can write the link on the inside the network tab and show this section called a protocol and then it'll show me what the requests are being made with. All right, now, fonts, images, asset compression. So, your hosting provider should have GZip enabled on at a minimum, and that's, your own web page test will show up as compressed transfer. And then, roughly as a faster higher compression tool that's from Google and more advanced tools off of that. And you can also check in Chrome developer tools whether that's being enabled if you click inside the network tab on one of the requests you can look for a content encoding here and it does a VR next to it that it's using probably. Image compression, this is a big one. So, on web page test also has a pie chart down at the bottom that shows how much of your data or that page load is attributed to images. So scroll down to the bottom, take a look at that. If your purple here is really large and you did not get a good grade on compressing images, then you probably want to compress your images. One cool plug in that I've seen that I like also helps you with scaling down huge images. So if you're trying to upload your 24-megabit pixel like image up to your website, it'll scale it down to something reasonable and that's insanity. And it can resize huge files on upload, you can adjust the image quality and you can also, you can convert PNGs to JPEGs. If you have a PNG that doesn't have transparency and it's large or it looks like a photo it should be a JPEG. And you can also bulk resize your images. Here's a screenshot with a couple of settings that you might want to adjust. So, I'm just gonna skip these. There's some more advanced topics. If you really need to, if you've gone through all these things and you need even more speed out of your site then you might look at lazy loading of images, something called servo push or adding a CDN. Some of those are a little more complicated but the CDN is often quite easy especially with cloud require. Let's save that stuff for last after you don't optimize everything else. And caching, so there are different types of caching. This is the exact I think I was gonna get to talk this afternoon about caching so I checked that one out if you're interested in that. Page caching, object caching, I'm not gonna go into details about this except page caching is a common one. Is it the solution for a slow site? And I say no. And caching was designed and it's great for scaling websites with a lot of traffic. So, I would argue that your site shouldn't be fast without caching. And caching on a site only hides poor performance and often only a part of the time. You use it as a last resort after you've tried everything else. Where does page caching fall down? Well, the WordPress admin is not gonna help you there if you're working in the back end if that's slow. If you're logged in often times, page caching is disabled so the front end of your site while you're developing it could be slower. You've got a large site with a lot of pages that might not be cached effectively. If you have content that's customized to the user again, you have issues. Caches can also expire and flush. And that next request, if it doesn't affect cached and you haven't optimized your site, it's gonna be quite slow for that visitor. So, summary, you want your site to be as fast as possible without the page caching. I found this really good article on a quote from him too is, it should be possible to turn off page caching on a regular day with normal traffic and not get nervous about it. And even on a busy day, it shouldn't crash your website. For WordPress, most managers and providers actually can provide page caching. It's not, if you can't get your time to first bite under, let's say, 200 milliseconds or you just want your site to be faster, then look into it or try to get that working. It'll often be noticeably faster if your hosting provider is not, isn't high performing or if your site uses heavy technology or uses certain things like WooCommerce and page builders, certain things, tend to start to put all those together and then it can slow the site, a WordPress site down considerably. And yeah, so you're saving it for last, test it out, you might still want it at the end but try and speed up your site as much as you can first. All right, and do it all right. So, summary, measure your performance first. Think a good hosting provider or switch to one. If your time to first bite is still slow, do some profiling with New Relic. Oh, sorry. Optimize your images, use a plugin with page caching if your host doesn't provide it already and add a CDN if your visitor's been around globally on warrior service far away from your audience, or yourself. And then, feel free to ask for help, ask your host and provider for help, ask a developer, ask someone here at WordCamp, sorry. All right, and it looks like I have a little time. And so, let me do just a quick, I'll do a quick demo of some of the tools that we have, including for testing your site performance. This is our website, if you come here and click on the little sign up button, feel like I'm just gonna create a new one now and confirm this. And once I've landed inside my account here, I can click this add an existing site. And so, I can pick a site here, I've set up a test site, go to any, thanks go to any, they gave me a great deal, I think it was $1 with a domain included, so 12 bucks for a year. And so, when you click this, it goes off and it looks up. Well, if I switch test and site, it might work. There we go. This will go off, it looks up some information about the website, where's it hosted, where's DNS being provided, and stuff like that, and I can then click add site and I'll go and then we'll add it to my dashboard here inside site district. And once that's done, it takes a little, I'm gonna actually skip through this next step here, I'll go back to installing the plugin for migrating a couple of sites. But just from adding a site in here with the domain name, I've got a few tools available to me. Over here on the side, I can run a web page test directly from here and it'll keep the results on site, site district, so I can kick that off. And then I'll pop that in a new tab and it'll probably take me a minute to run. I can do a speed test and this is one of the cool tools that we've built in. This will go off and it'll hit up to 30 pages on, well, you can actually control the number of requests on the site and figure out what the server response time is throughout the site. So this will go off and crawl through the site. I don't actually have much on the site, but it's hitting different pages on the site. Looks like they're all coming in and right about the same speed. And that'll go off and finish in a little bit, but maybe an idea of throughout your site how fast is the server response time on your site. And I haven't had to do anything complicated, pretty much just add my URL and click a few other things in here. And it'll finish with a summary of the program. Then you go back over here. Well, I actually import my site. I can just pop open this dialog and install the plugin. So I think that's my username. So I can put in any WordPress administrator username and password and try to go off. And our server will go off and try to install the plugin on that program site. You know what I mean? Yay, it worked. And then I can just click through, click import site, and it'll go off and I'll verify it. And try to pull in a copy of that site. So while that runs, it's basically just a fresh WordPress install 2017 theme, nothing special on it, so it should be pretty quick. But we can jump over here and see the web page test results from running it there. So we can see, we click in here to the film strip view, we can see what's kind of going on. So on this test site here, user was looking at a blank screen until about 1.2 seconds, and started to get this gray background fade in and then at 1.4 seconds, but the site's actually showing up there. You can see how long it took that site to live. So that's the web page test there. We'll come back over here, that's still going, see. And then on a repeat view, this is with the page cache, the browser cache enabled, should be a bit faster. We'll see here, blank screen, blank screen, and then boom, everything appears at 1 second. So it's a little bit faster. Like I said, very light site, just a 2017 base install. But that's how web page test works and how you can see inside of there. This is still setting up, I was gonna show off. Three minutes, all right. Hopefully this will finish up and I'll be able to run just a couple more tests real quick. If you've got your questions, get your questions ready, and we'll see if I have time for those. Like, don't have time for questions after this or I'll just let that run. I mean, I'll just take a couple right now. So got one in the back right there. You haven't said anything about Google PageSpeed Insights as a diagnostic tool, so I'm just curious what you're thinking is about that. So the question was about Google PageSpeed Insights and I think it gives you some great insights. But again, I would say still focus on what your users are actually seeing. So back to like, how long does it take until that content is showing up to the user and I know in front of their face. And you can use web names to test mobile versions and different connection speeds, stuff like that. So I would personally focus more on that. You could use it as a guide for helping optimize that, but then just keep your end goal in mind. Two, go ahead, we'll do it right there. What do you suggest for like areas with slower internet speeds? So the question was for... So like, there's areas within this, within the country where like, not like Seattle, where they have learned McDonald's, where they're operating on Seattle to have four McDonald's. Yeah, if you've got visitors that have a very slow internet connection, then yeah, that's a tough one. You're gonna, again, just test it as well as you can. If they're located far away from your server, get a CVM in place there. But see how it's performing and then see what's slowing it down. Is that your, is that your time to first play or is it your images or some other things loading down and you can do as much as you can off. It's kind of a diminishing returns. You optimize what you can and then to really speed it up even more, it gets harder as it goes, but test it and then kind of try and see what's going on there. And web page test, I think we'll cope with that. So once we've imported this site here, we can click this tool and it'll actually load up the two sites side by side and then click this test again. We'll kind of test it over and over a few times. See how long that actual page load is for the main site, the main test. So we've imported a copy. This is the exact same site, nothing has changed. We're bringing it over from Godaddy in this case and over to the site district and you can see the speed difference for loading. Now if we want to test throughout the pages that speed crawl that I showed you, we can do a crawl of both sites simultaneously and this will go through the pages up to 30 pages on both copies of the site and show you how fast they are. So. Yeah, so our import includes the plugins. It grabs all the files underneath your main folder and then the database and this copy's in the word restores. I think we've excluded, like if you're using backups and stuff like that, we've excluded backup archives and stuff like that, so we don't pull a bunch of extra data. And so that's that ours. In this case, the site is five times faster, nothing changed just by switching posting colliders. All right, that's that for the quick demo. I won't show you relic too much because we're out of time. I think all I have left was the link back to the site again and more questions. Okay, go ahead. What happens after this? How does that translate to the admin users? Often times, yes. Yeah. Yes. Do you have to be one of our community partners to make sure that the visual work has a lot of traffic in it and that you want to make sure that you get that, do you have a seal or do you have to make sure? Traffic can make a difference if there's a lot of traffic to the site. For sure, it depends on the hosting and the setup. I think the biggest reason, this is a little speculative that sites are slower on certain hosts, is that the server is overloaded. There's this high load on the server. So that load affects the speed quite a bit. So on our environment, for example, we make sure we balance that load and give a lot of space there. So that it keeps, even if there's a spike in traffic, it kind of keeps the speed constant. So it doesn't have traffic, but it can, but it's for most sites, it's not really a factor or something. I have seen on our platform, these sites can get a lot of traffic and the site won't slow down, but I have seen on other places that once they, once you get quite a bit of traffic to the site, the site will slow down there. It just depends on how the server is shut up. Hope I answered that. Yes. I have a question on your demo there. Yes. If that had been the year on the site that we would want to move over, then all you have to do is go in and change the DNS and use it. Yes. Yeah, so, well, we charge for hosting, so you have to sign up and pay for that if you were to migrate to us. But then, yeah, the only thing left after that is pretty much updating the DNS. Other many other questions? All right, I think one or two more, right there. You're going to reach optimization. We want to optimize the user load problem of what they're going to work with. What's the practice in terms of that? Do you usually use the plug-in in terms of compression? Do you even bother optimizing the Photoshop for the other brand first? And then you have to optimize the plug-in or do you just replace it? The plug-in? That's a good question. I think you might save time by not worrying about optimizing the day in Photoshop unless that's a streamlined process for you. We, let's see, I don't have any plugins really to recommend it. I listed the one up there and that's mostly because it'll do the work when you upload the images, which is nice, and we'll take care of really huge images. We actually, I didn't download, we actually built an image optimization tool into our hosting, and we did that because we can do it 20 times faster than other services because we can do it all on the server. So, and we actually get quite a bit better optimization than some of these other plug-in tools. Not sure exactly why we're not doing anything. I was going to say we're not doing anything magic, but we're using image magic actually to optimize the images. And I've seen images optimized elsewhere, and we run our tool on our server and we'll shave off another 30% or so. So yeah, I'm not sure what the best thing to do for that is think about what your time is, but in the end just make sure your images are well optimized, and I'd recommend about 75 card jpegs optimization. One more? All right, I think you have your energy. So I was using APM, which is application performance monitoring. And so it's a PhD extension that is installed on the server. It's very easy to install, but you do need access to install PhD extensions. It's like one command on a server to get it up and running. Yes? I'm curious what you can operate with. So we're not, we have just, our servers are located in one location right now. We have to spread out. There's some of the other hosting providers because I know Kinsta is geographic. They have the usual cloud engine. They're all over the world. I think Pantheon just has locations in the US. So we haven't branched out globally yet. Our servers are more like Alphan than we know. On Amazon's infrastructure. All right, I think I'm done. So thanks again everyone for coming. I hope you've heard of something. If you have any questions, we'll be right back.