performance

View on separate page

Transcript

Intro: Hey, everybody. Welcome to Episode 101 of How I Built It. Today I’m talking to my good friend and second repeat guest Chris Lema. Chris is going to tell us all about how he and his team at LiquidWeb built the Managed WooCommerce forum over there. Their goal was to make WooCommerce and e-commerce in general more accessible to people. He’s going to talk through his whole process as well as some important information you need to keep in mind if you’re building an e-commerce website with WooCommerce, or any platform, but we’re going to focus on WooCommerce today.

That is the episode that we’re going to talk about today. Before we get started I do want to tell you about our sponsors. Pantheon, who you’ll hear about a little bit later on in the show, and a new one, Loxi. Today is brought to you by Loxi, from the team behind The Events Calendar, the most popular events plugin on WordPress. Loxi is the online events calendar that makes it easy to display and publish your events on any website. It was built with design in mind to easily fit your website no matter where it’s built. Squarespace, Shopify and beyond. You’ll hear a little bit more about Loxi later on in the show. For now, let’s get to it.

Joe Casabona: Chris, how are you today?

Chris Lema: I’m doing great. Good to see you.

Joe: Good to see you too. I’m excited for this episode. I’m also excited because we are recording this before we both leave for CaboPress, which is a fantastic event that I’m deeply looking forward to.

Chris: It is the best business conference in Cabo. It may also be the only business conference in Cabo, but it’s the best business conference in Cabo.

Joe: See I was about to agree with you, assuming you were stopping at best business conference. So, with the qualifier it’s 100% true. Today we’re going to talk about your Managed WooCommerce hosting from LiquidWeb, but first for those for those who don’t know, why don’t you let us know who you are and what you do.

Chris: Sure. Like you said, my name is Chris Lema. I am the VP of products over at LiquidWeb. The majority of what I spend my time doing is doing product development and managing product. That often ends up being focused in WordPress and WooCommerce. We spent the first year that I was here working on WooCommerce, and then from there building out on top of that starting with WordPress, and then building on top of that the Managed WooCommerce offering. The dynamic is that most people know what Managed WordPress is because it’s been around for nine years. People have experienced that difference. If you go back in time, you remember back when if you were running WordPress on a regular shared host, or someone at Rackspace, a managed host.

If you called in and you had a problem their answer was, “Do I need to give you more RAM or do I need to give you more disks?” That was it. And you’re like, “No. But I’m talking about the application.” And they said, “No. We don’t pay attention to the application.” Nine years later, everybody has had an experience with a Managed WordPress host where you call and you have a question, they know WordPress so they can help you and that’s awesome. But if you put a plugin on top of WordPress  and you’re like, “I use this classifieds plugin,” they’re like “No. I don’t get into the applications that you run on WordPress. I just go up to WordPress.” What’s amazing is, with more than 3 million WooCommerce sites on top of WordPress, that was still the state of affairs.

You would have a WooCommerce store and something would be slow, and you would call, and they’d be like “Here’s what we see at the WordPress level.” And you’re like, “What are you seeing at the WooCommerce level?” “No. We don’t get into plugins. It’s too complicated. There’s too many of them. It’s too complicated.”  So we said, “Let’s build a dedicated solution that much like Managed WordPress was differentiated from managed hosting, let’s differentiate Managed WooCommerce from Managed WordPress and bring a lot of expertise, performance optimization, thinking about it differently for WooCommerce stores into the market.

Joe: Nice. That’s fantastic. Because those who have been around a while, like  you said, will remember calling your hosting company. The answer that I got more often than not was  “WordPress is inefficient, so it’s probably something in WordPress.” And I’m like, “That’s probably not the case.” Then they would ask, “Do you need more RAM or more processing?” But the Managed WooCommerce platform is  the first of its kind. Especially the way that you guys offer it, which comes with a ton of great tools. As we get into the research question here, the first thing I want to hit is you mentioned that you had to build out Managed WordPress hosting first. When you came to LiquidWeb there was no Managed WordPress hosting, and you couldn’t  make the jump to Managed WooCommerce hosting. Why is that?

Chris: A lot of what happens when you start, if you’re building a managed application in a managed hosting company, the managed hosting company pays attention to network and operating system, and disk and RAM. The monitors are set for that. But what you do after that, like if I went into my into our ops team and said, “Tell me which of our customers on our cloud VPS are getting a ton of traffic.” Meaning, look at the ingress routes for HTTP or HTTPS. If I said, “Show me the ones that are getting high traffic that are still HTTP, because I would like to offer them a migration over to HTTPS.”

They would be like, “We don’t look at that. Why would we look at the ingress player for HTTP versus HTTPS traffic for sites that are running on their own boxes? We’re the cloud service provider.” When you shift into applications, you have to go, “I need to monitor different things. I need to interact with it differently, I need my support to be different.” Thankfully LiquidWeb is already clear they were doing this. They were like, “We’re going to do Managed WordPress,” because they had done a scan of all their accounts and all their services and then gone, “We have 30% of our customers– 30% of our servers are hosting WordPress. We should do this.” What I added to the equation was, “Yeah. You should. But after that, we should do something dedicated to WooCommerce.”

That’s when people started scratching their heads going, “That sounds interesting. What does that mean? How big could it be, and what could we do, and how would we do it?” And  to your earlier point, it is different. It’s different. There is a ton of hosts, really good hosts, hosts that I have no problem recommending who will allow you to deploy WooCommerce like any other plugin. At one point when Bluehost and our friends at Bluehost launched, they said, “One touch deployment for WooCommerce.” And you’re like, “It’s one touch for every plugin. You click a button and it downloads it. There’s a lot of hosts and good hosts that can deploy WooCommerce. That’s not different.” Our competition never in my mind, it never was Kinsta, [Doty Engine], Flywheel, SiteGround, Bluehost, DreamHost, Pagely. Those were not my competition. My competition was Shopify and big commerce.

Hosted e-commerce platforms. When you compete at that level you realize that those people are optimizing their system for high traffic load. Whereas, imagine if you go to a regular WordPress shop and you install WooCommerce. A normative WordPress environment, when we talk about how you build it and you start looking in the details. A normal WordPress environment, if you’re a host, you’re going to want it highly cachable . The default over at Kinsta is that you get two PHP workers. There’s absolutely nothing wrong with that. Two PHP workers, in order to understand what a PHP worker is, you’d think of it as a cashier at McDonald’s or Burger King. You walk in the door, there’s a line of people, and you go “How many cashiers do they have?” Two PHP workers is two cashiers. You go, “OK. Why only two?” It turns out if you’re running WordPress a lot you’re going to put a caching layer so that most of the requests never hit PHP. Whether you’re talking about full page cache, that’s normally that we’re talking about. Something like a varnish or a cache plugin, you want that hit ratio to be 90-95%.

Which means only 5% is coming through to your PHP workers. You can be highly scalable and have two PHP workers. But try and go to any of these places and run WooCommerce and you’re going to quickly go, “Wait. This platform wasn’t really designed for WooCommerce, it was designed for WordPress. It was designed for high cache utilization and low PHP workers.” In our world we go, “No. I need 100 PHP workers. I need 300 PHP workers.” If you want 100 PHP workers  and you go to one of those other places you might be paying $1,000-2,000 a month, if they even let you have it. Ours is obviously much lower. When we were building this out we had to go, “OK. First let’s get everything right for WordPress.” But now, “How do you layer on top of that and still tweak it differently so that you can get high optimization, high performance, on a platform that– Yes, you’ll use some cache. We offer varnish and we offer [Retis] for object storage, and  there’s ways to do that. But most importantly, I need everything to be super-fast.” So then I got to pay attention to my infrastructure and shape it for e-commerce logged in transactions, not just non-logged in page views.

Joe: That’s the big difference. On WordPress I publish a post, it’s highly unlikely that I’m going to change that post for a long time. You can reasonably assume that people will view the same copy. But if I’m buying something, I’m  going to be logged in, the experience is going to be different from another person who could be buying another thing, and you can’t reasonably cache that information.

Chris: You start getting nuances. Let’s say I want to cache the product page. You’re like, “It’s the product page. Everyone is going to see the same.” You go, “Yeah. But the product page has variations and variations are changing in inventory. I don’t want you to pick a blue shirt that’s size large when that large blue is no longer an inventory.” So now I’m dynamically changing that and you’re like, “This is going to be more complicated than just a regular blog post.”

Joe: Or if you’re split testing prices or something like that, there’s  a lot of things that are moving parts with an e-commerce store. You went pretty deep there talking about  PHP workers and stuff like that, but you also include a lot of extra services for WooCommerce shop owners. Can you talk about what the decision making was there? I see you include Jilt, you include Glew, and things like that.

Chris: Here’s the thing. Again, if you get to the point of saying “My customer, the competition I have for that customer is buying Shopify right now, or they’re buying big commerce right now. If you are– Let’s say you go to Google and you Google “WooCommerce versus Shopify,” or  “WooCommerce versus Shopify review.” If you pull up the top ten links, top ten articles, you’re going to read those, and you’re going to hear a story that goes something like this. “WooCommerce has these features. Shopify has these features. WooCommerce has this interaction, Shopify has this interaction. WooCommerce is free but you have to buy a theme, and you have to pay for a couple of feature plugins. Shopify is cheap, but you have to pay for features. You also got to pay transaction fees.”

You get to the almost end, the penultimate of the article. They’re about to tell you, and they go, “Through 90% of this article, it’s a tie.” Then you get the final paragraph, and it says something like, “But with WooCommerce you have to figure out your own hosting, you’ve got to become your own dev ops, or you got to hire people to manage it. You’ve got to support it, especially in high traffic, you’re going to have your own engineering team that has to do this and that. So unless you’re a developer, we recommend that you choose Shopify.” We looked at that and went, “OK. What we have to do is take that last paragraph out of the reviews.” How do we build a product, how do we design a product where people no longer say,  “But also I have this worry about managing it, hosting it, configuring it, getting it optimized for traffic and performance? What if I didn’t have to do that?” And you go, “That’s managed WooCommerce.”

But then when you do that you then end up with– Hold on a second. We know some things. We know that there is no native cart abandonment with WooCommerce, which means Shopify is going to do it and you’re not going to do it? That’s not right. Or we know that Shopify, even though you have to pay, you got to go up the chain to get their advanced reporting. You’re not going to get it for $29. But eventually when you go up, like let’s say you’re paying $299, you’re going to get their advanced reporting, but their advanced reporting is not going to impact the performance of your store. We all know if you’re running advanced reports in WooCommerce you’re going to be using the same resources and doing the same stuff that your store uses to deliver value to customers, and that’s going to kill you too. So we went out and did the deal with Glew so that we could have a best in class analytics and reporting solution that didn’t have performance impact on the store.

At Jilt, abandon cart is also a SaaS, so it doesn’t affect the performance of the store. A lot of people are like, “It is cheaper if I just go buy this plugin.” And you go, “Right. But if that plugin does all the work on the same server, and it’s pulling data and running queries, then your site slows down.”  So we started looking at some SaaS partners to integrate so that we could keep your site going fast. Speed and performance being the most important thing for an e-commerce store. So, we did that. We turned around and went, “OK. What else do we want to give these guys? What else should be in this package? We tested 50 different WooCommerce themes and found the fastest. Inched out in front of a couple others by barely a little, but Astra was the fastest performing theme for WooCommerce. You go, “OK. Astra is free which is awesome, but Astra has a pro edition which is a little plugin that goes with it and that opens up a lot of configurations for WooCommerce.”

And we went, “Let’s go do a deal and get Astra pro here for everybody.” Everybody who signs up to our business plans which is the $250 and up, they get that. They get Beaver Builder too because Astra is great with page builders, and of the page builders Beaver Builder is the one that is not only incredibly powerful but also really good with caching. So you go, “OK. Let’s get these together.” And we said, “OK. We’ll get you the page builder, especially for your landing pages or product pages where you’re doing something custom. We’ll get you Astra for your overall theme to be fast. We’ll get you reporting through Glew, again if you’re on the business plan, and we’ll get you Jilt to do this stuff. We’ll get you a high amount of PHP workers and give you a good amount of RAM.” When you go look at the chart, and you see what you get for that price, nobody who’s running a real store has ever called me up and said, “Can I get a discount? Because this is expensive.” In fact, a majority of our customers who have come from other hosts have lowered their fees. They were paying $1,200 and now $750, and they were paying $600 and now $250.

Because we increased their RAM or we increased their PHP workers, or we increased their feature set, and they didn’t have to pay for some of those other plugins. They go, “You’re building the whole solution.” And you go, “That’s exactly what I’m doing. I’m building you a solution. I’m not just saying, ‘This is about WooCommerce.’ Most companies would say, ‘Yeah you can install WooCommerce here.’ They’re like, ‘Yes you can install the plugin.’ Our goal was to build a total solution.”  So for all our business plans, which is the standard, the plus, the pro, and the enterprise. We’ve created a solution that comes with a whole bunch. Then of course as we launched in the last 30 days, we said, “Not everybody has an existing store. Not everybody is spending $1,000–” Yesterday I was on call with a guy who was spending up to $100,000 a month in hosting who had just shifted over to an AWS plus Rackspace combo which brought it down to $30,000, and my quote going out to him is going to be something like $6,000. We’ve looked at all the data, and what they’re paying for is not always just the hosting.

They’re paying for the support, or they’re paying for the knowledge that can help them with this, or they’re paying for the ability to extend their servers when they need to do stuff. Dynamic adjustments. But they’re paying a lot, and almost everyone that comes over starts paying less. That’s because they’re serious. 400-500 concurrent users on the site at any given moment in time, you’re like “Yeah. All of us would love to have stores that are that are running 400-500 people at the same time. Hitting that store and trying to buy stuff.” There’s a lot more people that are that are looking for starter plans, so over the last three days, we rolled out three of our four starter plans. The beginner plan, now we have the basic plan, and we have a marketplace plan.

The one that will come out in the next couple of weeks is our drop shipping plan, and those are all for people who are just getting started. The beginner plan starts down at $39,  and then the others are $99, $125 and $149. The dynamic there is people who want to get started, you tell me you want to get started with drop shipping, I’m going to get you started with drop shipping. You’re going to be ready to go. You want to start with a marketplace? I’m going to give you the marketplace Dokan plugin, and we’re going to help you get started with Dokan so that you can run a marketplace.

If you’re a beginner, we wrote some additional code that locks in either how many SKUs you have,  so you can say “I have only 15 SKUs but I want unlimited orders,” or you can say, “I have way more than 15 SKUs, but I’m going to tap it out at 150 orders.” That lets us constrain it which allows us to shape how many resources we put against that, but you get started for $39. At any point, if you’re like, “I want to take off the constraint.” Our basic plan is that we run one command and we’ll take the constraints off, and you shift into the basic plan at $99. If you are making enough sales at $39 to say, “I’m ready to take off the constraints,” then $99 is not a problem. So we started with the serious plans because we started with serious stores and customers that were running serious stores. Then we said, “Let’s back up and fill in the bottom part of this to make sure that people who are just  getting started had an easy way to do that.”

Break: Today’s episode is brought to you by Pantheon. WordPress 5.0 and the new editor Gutenberg are coming. Are you prepared? Do you want to learn about the changes in advance? Pantheon has gathered resources to help you prepare, including webinars and tutorials. Pantheon also has made it easy and free to try Gutenberg with your site before the official launch. Visit Pantheon.io/Gutenberg. Let them know that How I Built It sent you. Now, back to the show.

Joe: You mentioned a lot of things here that boil down to this. You are helping people make money. I’ve always said to people, “It’s a little easier to sell the price of an e-commerce store or a website that is directly making money,” because you could say,   “Right now you’re spending this to make this. I can help you make this if you increase your spending a little bit.” For the beginner, if they’re spending $39 bucks a month they are– If they make 2, 3, five sales or whatever, that’s justified. They’re able to go to the next level where they are getting more resources and more tools. When you get into Jilt, Jilt’s pricing model is incredible. Because the first 25 recovered carts are free, or something like that. By that point, you’re like, “They’ve already recovered x amount of dollars for me, so it’s worth paying for. You are helping people get started with online and with e-commerce. You’re making them customers for life because you offer the best service.

Chris: That is exactly right. Our goal is to eventually to make sure that people think of WooCommerce in the same way they think about Shopify. Today people go, “Shopify? I don’t have to think about it at all, and WooCommerce, there’s so much  to figure out.” And we go, “We can mitigate that so that there’s not a lot for you to think about.’ But once people start working with it, you start realizing, “I have some constraints in Shopify or big commerce world that I don’t have in WooCommerce. The flexibility, the open source, the speed of customizing something  is dramatically different.”

You get customers who are saying,  “By the way, I have this one-off inventory system.” And you’re like, “There’s a plugin for that.” And they’re like, “That’s great. Shopify couldn’t do it.” Or they go, “There’s a plugin that is part of that but doesn’t do all the rest of it, but it’s open source.” Your development team can take it and layer on top, and they go “This is amazing. That just took two months out of our development time.” And you go, “Right.” The flexibility of WooCommerce allows us to compete better with Shopify if we can do the rest of our part right.

If we can get rid of the headaches around the hosting and everything else, then we can do all that. Then where we spend time is helping people speed up their stores, optimizing it, making it go faster so that they make more money. We had a customer who was using WooCommerce on AWS, and they crashed. They just straight up fell over minutes after a launch. I got a call about 7:00 PM that said, “Can you get on a phone call with these guys because they want some help.” I’m like, “Yeah.” So we started the call at 8:00, just like  this on a Zoom call, and I told him “I won’t get off this call until we have a plan.”

We were on the phone with him until 2:00 in the morning. By that point I pulled all their code down, we had done a scan over a bunch of stuff so I could figure out what was going on, and I went “This is a problem. “This is a problem. This is a problem,” and the development agency was on the call too, and they were like “We’re out of our league. We didn’t know. We don’t know what to do.” And it was fine. There was nothing wrong. They just admitted they designed the entire theme using ACF as their page builder and they hardcoded things, so if you hard code something part of what that means is that the full page won’t load unless each component loads, but that component was tied to a plugin that was poor performing.

So when you turn off the poor performing plugin, then the panel doesn’t load, and if the panel doesn’t load, the rest of page doesn’t load. You’re white screen, and you’re like, “Guys. There’s a bigger problem.” We said, “You have a whole audience that is ready for this launch. Let’s go spin up a new store, a little pop-up store  for this one product, and we’ll get you making money, and then we’ll turn around, and we’ll integrate back in.” In two hours they generated– We did that. We helped them spin up the new site, get everything configured, get it running. Bring their product inventory over for that product line. We launched within 48 hours. I don’t know. There’s not a lot of agencies that would be like, “If you call me on Sunday I can have a brand new site for that product line launched Tuesday night.”

Because most of them are like, “If you call me on Sunday we can have meetings this next week, and then  we can get on a deal, and then we can put out a broad plan and maybe in three months–” But these guys had an audience that was queued up ready to buy product, so we did a pop up store, got it up and the first two hours. They did more revenue in two hours than they had done in their best month of the year last year at Christmas time. What we focused on was the engineering, the infrastructure and at the low level, the code that would make sure that their page could load fast. When we first started working with their pages, they were loading in 28 seconds. I’m like, “It doesn’t matter if it goes down because even when it’s up, it’s not up.” Then we got him down to 1.7 seconds, and it was fantastic.

Even as we were doing it, and I don’t recommend this to anyone, but while we were doing it there seemed to be some problem with the payment gateway. People were queuing up and getting stuck. You could watch it in the Google Analytics real time. We’re watching it and seeing there’s this choke point, people are sitting at the checkout, and there’s too many people at the checkout concurrently, there’s a problem there. We started looking at the API calls to PayPal and going,  “There is a problem. But it’s not on our side, and it’s on PayPal.” So we say, “Let’s switch the payment gateway to PayPal’s standard where we route them to PayPal, and they can check out there. We’re not going to use their API.” I’m not joking. We did the change in real time.

So I got a developer who’s on the phone with me, and I’m like, “OK. Install the standard. OK, get it configured. OK, leave it inactive. OK, I have it here. Now, product guy on their side, you look at the numbers. Where are we at? Let’s get to a point where this goes down for just a second. OK, we went for 800 down to 200 people concurrently on the checkout. OK. Now, swap. I’m going to turn this off, you turn it on, and everyone will move on with their cart active straight to PayPal.” Not only did they not lose a thing, all of a sudden you saw in real time, “People are no longer choking up here. The checkout page is not aggregating all these concurrent users. But also, look at PayPal now just bringing in money. One order after another, after another, after another.”

And we were like, “OK. We’re good.” And the CEO of the company is like, “Do you always do that? Do you have to do that? I’ve never seen that happen.” And I’m like, “I don’t recommend changing payment gateway in the middle of a product launch, but you had a problem, and we had to solve the problem.” Because what you don’t want is everyone to look back later and go, “That was a mistake, and we couldn’t get that done. But lesson learned. I guess we’ll learn for six months later for our next product launch.” And you’re like, “No. There’s money on the table right now.”

Joe: Right. That’s an expensive lesson to learn.

Chris: That’s right.

Joe: Especially one that you could fix in the moment, if you can. That was– This is not a video podcast, but if you could see what I looked like while he was telling that story it was the same look I had when watching The Last Jedi. During the fight scene with Ray and Kylo Ren. So, that’s an incredible story. I haven’t even gotten to the title question yet. We’re coming up on time, but for the title question I want to ask you something a little bit different. Because we got a lot of the details for how you put together this hosting, but if I want to build a fast site, how would you build that? What would you recommend I need for that?

Chris: To build a fast WooCommerce site?

Joe: Yeah.

Chris: There are several dynamics that are at play here. First of all, you need image caching in CDN, you need image compression. People will upload images that are way too big that they don’t need. They’re only going to show 600 pixels wide and they upload a 5MB image, and you go “No.” So you want to compress your images, you want to use a CDN for the non-image files, whether you’re talking about your JavaScript files or if you have PDF files or anything else. You want to be able to get those in the  broadest and most distributed dynamic so that anyone anywhere in the world who pulls this up is getting it from the edge, rather than running all the way to your server, which allows you to have more transactions permanent on a server because they’re not worried about doing other things.

That’s another piece you want to do. You want to do something we call “Ruthless prioritization,” which is look over every single plugin and figure out if you need it or don’t. There’s a lot of plugins on sites that you just go, “No. You don’t need that, you don’t need that, you don’t need that.” What you  want to do is evaluate the performance impact of any site. Which means you need to get good at testing, but you want to understand the performance impact of any plugin on a site and you also want to understand the revenue implications on a site. If you have a pop up that is not super-fast, has a couple of issues etc. but it generates 30% lift on your revenue, you keep that baby there.

But if you have some that you like because you think it’s super cool, there’s a little video of me that slides up into the transparent background and I’m talking to my customer. And you’re like, “Not only is it poor performing but no one’s ever clicked it.” You’ve been like, “Come here. Click here to go to my–” And you’re like, “No one has ever. You have zero clicks. Turn this off.” “But I like it, and it’s me talking, and it feels good.” You’re like, “Turn it off.” The ruthless prioritization gets gone a lot, and another thing we see in there is that sometimes you have plugins and you’re using one feature on a plugin that has 48 features. Yet you’ve got to  load a bunch of JavaScript every time that page loads, you’ve got to load a bunch of images and other scripts, and you’re like “No. This is bad. Let’s just take the one feature you really need and load it into something like code snippets so that it does that one feature.”

That’s again, a lot of mistakes that happen when a store is built by someone who may not know how to write that line of code, so they just keep grabbing plugins off the shelf. And you’re like, “No. That’s suboptimal. Because you’re causing PHP engines of your server to do work. To parse and process stuff that you don’t even need, because it’s not a feature you’re using.” Then I’d say there’s a whole bunch of– Either it’s in WordPress or WooCommerce, there are settings and features that  you can turn off. Perf Matters is a plugin that I use often. Perf Matters will let me quickly turn a bunch of these things off . Windows used to have a writer, Windows Writer which let you write blog posts from their desktop.

There’s codebase in WordPress to let you hook into that. If you don’t have a Windows computer at all, you’re not likely to be writing your blog from there and if that’s not part of what you do, turn that off. Part of what happens is you use something  like Perf Matters to close it off. But Perf Matters also has features like pre-calling and pre-fetching sites and resources. Pre-fetching and working with DNS pre-fetching that’s connecting to servers or other places, other domain names. There’s a lot you can optimize there so that things move a little faster, and then having a lightweight theme which we talked about already. From there  you’re in a pretty good place to do the last step, which is performance test over and over again. Because as you performance test you’re going to find little things that you go, “What’s going on here?” And that’s when you realize you need to optimize X, Y or Z.

Break: This episode is brought to you by Loxi. Your website needs a calendar, and Loxi is made up from the experts behind The Events Calendar on WordPress. Loxi is the platform agnostic events calendar you can embed into any website with a simple cut and paste. Loxi makes it fun and easy to build a calendar and add events with robust listings. It’s free to build your calendar and you only pay when you’re ready to publish. I can’t stress this enough. If you need an embeddable events calendar no matter what platform you’re using, Loxi is the way to go. You can check it out by going over to Loxi.io. Exclusively for How I Built It listeners you can use the code GETLOXINOW for your first 3 months free. That’s GETLOXINOW your first three months for free. Now, back to the show.

Joe: You mentioned Perf Matters, that’s a WordPress plugin. I just picked up WP Rocket, which maybe does something similar. Do you have a preference, one over the other?

Chris: I like WP Rocket a lot. On our particular platform, because we’re offering [Retis] and because we’re offering varnish, and because we roll in a couple of other things. Some people go, “I don’t think I need WP Rocket to do that.” But if you’re not on our platform and if you’re just trying to get a site fast, WP Rocket can be an incredibly useful product to help you navigate through a couple of these different things. Not just one thing. You’d want to look at a couple of these different tools to see what’s right, but it does some optimizations and it also provides some cache, and it lets you connect to other solutions if you need to. There’s a lot in there that’s very  powerful.

Joe: Gotcha. Cool. I was between the two, and I saw your testimonial on the WP Rocket page. Because I do have LiquidWeb Managed WordPress hosting, but I don’t have Managed WooCommerce hosting yet.

Chris: We can work on that.

Joe: Cool, so there’ll be a bonus episode after this where we talk about that. But there’s a lot of really great information in there, so if you are not on LiquidWeb hosting then these are definitely some really good takeaways. We also talked a bit about your plans for the future. You mentioned that you’re rolling out a drop shipping tier, or level. I just spoke recently to Anton Crowley from Drop Ship Lifestyle. That episode is coming out after this one launches, but drop shipping is an  interesting topic to me. Maybe in the last few minutes, your plans for the future? Why did you choose drop shipping as a specific level?

Chris: Because when we talk about people who are getting started in e-commerce, the two that are the hottest in 2018 predicted to also be hot in 2019, are drop shipping and marketplace. It doesn’t mean that people aren’t going to do courses, it doesn’t mean people aren’t going to do membership sites, it doesn’t mean they’re not going to do lots of other things. But the hot ones, the ones that are tracking and people are searching for in Google trends and the ones that people are looking at in other platforms like Shopify.

You look at these things, and you go, “People want this.” One of the things that they have in common, drop shipping and marketplace is that people don’t have products. When you don’t have products, but you want to get into e-commerce, what do you do? You let other people supply the products, or you let other people create their stores in your store and let them sell their products. Either way, a drop shipper marketplace is there. So our drop shipping integration is a SaaS product called Shop Master, and most people I talk to have never heard of Shop Master. But again, the people I’m talking to are WooCommerce people, so they’re always talking about,  “I did a Google search for WooCommerce and drop shipping,” or “WooCommerce and ali express, what are the plugins there?” And there’s “ali drop ship,” and there’s “Woo drop ship.” And there’s a couple of those, but none of them were doing what I wanted out of this.

That’s where I went, “OK. Let’s go figure out what happens if I don’t just want to be with ali express, what if I want to be with Amazon too? Or what if I want to be with other pieces? What if I have more than one website?” You start getting into these nuances, and you realize, “No. I need a better one.” Of course, as you know I’m a big SaaS fan, it’s why we do so much of what we do the way we do it. So we went looking for a drop shipping SaaS that we could integrate with, and that’s Shop Master. We’re excited roll that out in the next couple of weeks.

Joe: Cool. I’ll be sure to try to pair the drop shipping episode closely with this one, so  that you get a one-two punch.

Chris: That’s right.

Joe: For your 2019 plan you can get the LiquidWeb drop shipping tier, learn how drop shipping works and maybe have a lucrative 2019. It’s always a pleasure talking to you, but I do need to ask my favorite question. You provided a very good answer to this last time, which is do you have any trade secrets for us?

Chris: Yeah. I can give you something interesting. One of the things we know when it comes to e-commerce is that speed is everything, performance is everything. In that space one of the things we know is that when your store takes more than 2 to 2.7 seconds to load between pages, people’s session length drops off by 50%, which means people will leave. That’s our experience too.

When you go to a website and you click and you have to wait a few seconds, and you click and you have to wait a few seconds, eventually you’re like, “Forget this.” Speed is a big deal, performance is a big deal, it has direct impact on session length which has a direct impact on revenue. What you really want to do is performance test your store. People go, “Yeah. But I don’t know how to do that, and I don’t even know what test to create.” What we’ve told people forever is, “Heads up. LiquidWeb will do that for you. If you come over here and you get, especially with our business plans, we have a whole bunch of different tests we’ve created and we can run your stuff through tests and tell you where there are things that are poor performing. That’s it.”

My little insider secret is not that. It’s not a pitch for “Come to LiquidWeb.” My insider secret is that those tests are public somewhere. All those tests that you can potentially load up at load impact and run on your own, they all exist somewhere publicly. So, A. You can look around for them. Or B. You can email me at Chris@LiquidWeb.com and I can send you a link to the repo. If you’re one of those people who goes,  “I want to tell my customers that I performance test their store,” or “I want to know how to do this for my store,” that’s our secret. They’re public already. We can help you get access to them so that you can run your own.

Joe: That is a great call to action, and a great piece of advice. You already told us, but where can people find you?

Chris: I’m over at LiquidWeb. You can find me at Chris@LiquidWeb.com, you can also find me blogging over at LiquidWeb.com. I also have a blog over at ChrisLema.com. The easiest way to reach me is on Twitter, which is @chrislema.

Joe: All right. Chris, it’s always a pleasure. Thanks so much for joining me today, I really appreciate it.

Chris: All right. Take care, buddy.

Joe: Thanks out there to everybody listening. Until next time, get out there and build something.

Outro: Thanks so much to Chris for joining me today. It is an absolute pleasure talking to him whenever I have the opportunity to do so. I appreciate everything he talked about, not just from a building a platform standpoint, but good advice for any e-commerce store. Talking about the importance of performance and how you can improve performance. He let us in on a little secret that drop shipping is very popular. In 2019, I’m going to have a guest talking all about drop shipping. Definitely look for that in the future. Once again, thanks to our sponsors Pantheon and Loxi. This show would not be possible without them.

The question of the week for you is, what are some ways you can improve the performance of your e-commerce site? I just went through this myself by working my way through some performance issues over on Creator Courses. I’m curious, what are some ways that you can improve the performance of your e-commerce site? Let me know on Twitter @jcasabona or email me Joe@HowIBuilt.it. If you want to join the conversation with other listeners, you can head over to HowIBuilt.it/Facebook and join the community over there.

You can find all of the show notes and everything that we talked about at HowIBuilt.it/101, and then head over to Apple podcasts and leave us a rating and a review. It helps people discover us. That’s everything. Thank you, Chris, thanks to the sponsors, and thank you for listening. Until next time, get out there and build something.