Music
Jessica: Hey everyone, welcome to another episode of Greyd Conversations. Today I’m joined by Zach Stepek from BigScoots, and we are going to talk about Hosting for Hundreds: What Agencies Need to Know Before They Grow. So this topic will be about, hosting most likely. And, but specifically from an agency context and what you should think about when it comes to hosting not just one website, but, many websites and especially when they get bigger. What challenges are you looking at? And I think with Zach, we have a great guest here. Welcome, Zach.
Zach: Thank you. Yeah. It’s good to be here. Excited to join the ranks of all of the great guests you’ve had on this show. Thank you for having me.
Jessica: Yeah. Awesome. So you’ve been an agency owner, and you’re also director of partner programs at BigScoots. So you’ve seen basically both sides, of this topic. One is an agency owner from the user perspective, if you will. And now from a from a hosting from the perspective of a hosting company and, So before we dive in into all of this tech and strategy, what goes wrong when agency agencies start scaling, their hosting needs? What is the common thing that goes wrong?
Zach: There are a few things. One, you know, when you look at hosts as as partners and you’re you’re looking for a partnership. There are a lot of shiny things out there that look really great. And these hosting programs that promise you the world and, you know, it’s, these partnerships are all designed to be a sales tool, right? The the hosting company is trying to sell more hosting plans. That’s the underlying goal of most of these agency partnership programs. And so you have to kind of look past that and look beyond just the the shiny features, the the bullet points of what they’re trying to sell you on. And remember that, you know, the underlying goal is for them to sell more hosting, right? So once you look at it through that lens and you can kind of compartmentalize the things they’re telling you that you’ll get, it’s, it’s really easy to compare feature for feature instead of, you know, comparing promise for promise. So yeah, that’s really the biggest thing is just making sure that you’re focused on what the tech provides, what the platform provides, not necessarily just on what the partnership is promising You.
Jessica: Oh, okay. Cool. So when you start for planning, what should you look out for? So if you have like just a couple sites, very small, but there is the possibility that they may grow into something bigger over time. What is something you need to look out for there?
Zach: So one of the things is to make sure that whoever you’ve chosen as your hosting partner can grow with your customers, right? So now some hosts have a growth ceiling. It just is built into who they are as a company. And once you hit that, you reached the limit of what they can operationally support. And now so you need to know kind of where that growth ceiling is and have a scalability plan for what you’re going to do for your customers if they move beyond where that growth ceiling ends. There are other hosts that don’t really have as much of a growth ceiling. And you can, you know, grow a site to enterprise scale, but that may become price prohibitive or cost prohibitive because, they have a very set standard in which they do things where they’re renting infrastructure from a cloud provider like Google or AWS or, open metal or Azure. Right. And so all of those have fixed costs that grow as the usage of those platforms increase. So, it can grow exponentially to where you’re, you’re spending, thousands or tens of thousands of dollars a month, a hosting bill for a client that has a massively large site or an e-commerce site that’s featured on shark Tank or Dragon’s Dam, right? These are things that cause massive traffic spikes. And, if those things happen, you have to know. Yeah. What exactly is the plan here? And with a lot of hosts, if you go over your limits, the cost of going over those limits is astronomical. And now they do that because they want you to upgrade to a bigger plan, pay more per month, and that’s you know, obviously that’s the growth pattern for most hosting. But sometimes those overage, charges can get gigantic. And if you’re trying to manage everything yourself on AWS or Google Cloud, there’s there’s no way around those overages, right? If you go over what you’re supposed to have, first of all, if you go over disk space or memory on those servers, they’re just going to stop working…. Yeah. Yeah. That’s just you’ll hit a wall and you’ll get 500 errors and the site will go down. So now that’s kind of the difference between managed WordPress hosting, where you have somebody watching resources and making sure those things don’t happen. And the kind of DIY, you know, we give you a panel to manage your own hosting or, we’re now going even deeper into AWS or Google Cloud Administration. The difference is you have people watching resources, making sure your site’s going to stay alive. So that’s another consideration, to make is, you know, what is what does support look like at the company that we’re talking about hosting? Because support becomes a very important thing. Unless you want to become an infrastructure expert, which, you know, more power to you if you want to. But it’s a journey, you know, it is quite a journey to understand scaling a website at, you know, massive traffic levels. So yeah, those are the things that I would, you know, consider moving in to have a scalability plan, know where those limits are on the plans you’re on, know what plans you can move to, know what the cost increases are going to be. That way you can communicate all of that to your customers well before any of that becomes an issue.
Jessica: Yeah, and I think that’s important to know because like if you have a feeling I mean, sometimes you have just have very simple corporate websites that do not have an online shop or anything specific to it. They probably won’t need any scaling at any point of time. If, you know, like, okay, it’s just a company that does its business in a certain area or whatever. But as you said, when you have like an, a TV appearance, or whatever specialty is coming up, then it just goes and goes down, very quickly, because if people just want to know information and it’s been broadcasted on TV or somewhere else, hitting YouTube channels, whatever. That’s that side has to, withstand a huge amount of traffic that comes in in such a moment. So, what do you think are the biggest red flags? You see, when an agency plans to host, dozens, or even hundreds of sites?
Zach: Well, I think one of the things is we’ve got an a habit as a community of following the example of hosts that offer what I call, deeply discounted or bargain basement hosting. Right. And so a lot of agency owners who haven’t experienced sale or, you know, used to these prices that are, you know, 3 to $5 a month, right. And what they’re doing, receive those prices is putting hundreds of websites on a single server. And so a lot of agency owners, when they get Ahold of one of these panel companies, you know, I saw clockwise right when they get Ahold of a panel company, they think, oh, I’ve got two VPNs. I can put as many sites as I want on this thing. And then if it starts to slow down at the point that that starts to be an issue, then I’ll scale up. And, you know, they don’t realize that if they hit that point site start going down before they have an opportunity to scale. So, that’s one of the big pitfalls that I see with people who are going to be running hundreds of sites. The other thing is, you know, if you’re running hundreds of sites, how are you managing the basics of those sites? Are you handling WordPress updates? Plugin updates, theme updates? How are you making sure that core web vitals stay in shock? Right, so that, ranking stays high? How are you handling, you know, search engine optimization? How are you optimizing for a, how are you, you know, doing all of these various things? How are you handling accessibility concerns? How are you making sure that you have things in place for GDPR, GDPR or other privacy laws that, you know, we have these requirements that we have to meet now that are mandated by government entities. And so, you know, there are so many little things that we have to worry about. And anytime you can pick a partner that can take some of that concern, some of that complexity, and they can just handle that for you, and you can trust them to handle that piece for you. It is is your burden as an agency. So you know, when I was looking for partners, when I was running my agencies, I was always looking for people that made my team better. That augmented what my team could do internally…. And so that’s why I chose the partners that I did, because how they were equipped to be able to handle issues that I couldn’t, and they made my life easier. And so those two things were really the most important. They make my life easier. Do they make my customers lives better? If there were issues, how quickly did they respond? What was the, you know, SLA for support? How quickly did the issues get resolved? How quickly was I alerted to issues when they happened? And yeah, so all of those things really mattered. And then when there was a large issue, like, for example, the the whole reason I got into WooCommerce in the first place, which was, you know, a friend calling me on Thanksgiving, my friend Noah, and saying, hey, my site’s down. Can you help me? I never touched his site before. And, he had been interviewed by, Jeff Zodiac, the voice of the Chicago Bears, the Google football team. And you know. Oh, here. And they were playing against the green Bay Packers. Wisconsin’s football team were neighbors and rivals. Right. So, during that game, an interview with my friend Noah, who runs a veteran service organization called ask Mike and aired, during the pregame. And they got about 14 orders in, and the site crashed. And he didn’t know why. He just didn’t understand what was going on because he had never run into that before. Right? He had never he had never hit those limits. And turns out they were on a VBS that was donated very graciously by a company out of Chicago. And, it was underpowered for the level traffic they were getting. And it was also Thanksgiving, and it was a smaller hosting company, and nobody was in the office because it was holiday. And so there was nobody to call, couldn’t get a hold of anybody at the hosting company to reboot the server. So, about four hours later, we had it back up and running because we had to figure out how to log into it, how to get into cPanel in order to reboot the server, do all these other things that we needed to do in order to get the site up and running again. And like I said, I had never touched the site before, so, by the time we had done that, the window had unfortunately passed in this veteran service organization that’s, nonprofit, lost some of the great opportunity they had there to, you know, become a larger entity than they were and help more veterans. So I ended up working there for 18 months after that, rebuilding their site, optimizing it for performance. That’s how I ended up getting into that whole WooCommerce performance area. But the key is you don’t know what you don’t know until you realize you need to know it. And so having things like scalability plans, having all those, in in advance of having an event like that happen helps immensely.
Jessica: So you already mentioned WooCommerce, and, WooCommerce itself is a very resource hungry plugin already. And, you just need a good starting point with the hosting already to even run WooCommerce smoothly. And, so when WooCommerce is a beast at scale. What are your top strategies for keeping it fast and stable when all the volumes explode? So as you mentioned in your example, it was 14 orders, right? Just a very tiny number. And then the site went down because just there was just too much traffic on it, which is not very much 14 orders.
Zach: Well, and the funny thing is in that in the root cause analysis, once we finally got in and figured out what it actually crashed the server, it wasn’t PHP that crashed the server, it wasn’t the database that crashed the server. The hosting company, trying to be helpful. Had put, had both an email server for the entire organization on the same server, the same VPNs and WooCommerce generates, multiple emails per order. And so the emails spool of outgoing email from the same server ran the VPNs out of memory. That took the site down. So if the email hadn’t been there, they probably would have survived at least a little longer. They would have been able to handle, you know, 4 to 5 times the number of orders that they had at least. But the email ended up taking the entire server down. So, and that kind of transitions well into, you know, what are my strategies for keeping that fast and stable? So WooCommerce is, is, as you said, a very complex plugin. That’s what happens when you now put an entire e-commerce platform inside of WordPress, right? E-commerce is not a simple operation. It’s something that has a lot of different, things that all have to work in harmony from, you know, product management. You order management to payment processing, inventory management, all of these things have to work together and there’s, you know, 17 or 18 different systems at play there that all have to work in concert, like an orchestra. And so now if you try to fit an orchestra onto a tiny, tiny stage, you have people trying to play in the wings and trying to play on the ground. And, you know, there’s just so that’s part of it is, that you have to have an environment that is meant for handling something at that size with that level of orchestration. So that’s the first thing. The second thing is that when you are running a high volume website, it generally, and it has what I call a death by a thousand cuts, which, my, my brother is a fan of, of samurai films, has been here for know most of our life. And so I’ve seen a lot of samurai films. And now there’s this, this thing where the more you get caught, the worse things. Yeah. Right. So as as you’re running one of these larger sites, small little cuts, even paper cuts hurt if you. Yeah, if you have a ton of them, it’s going to hurt a lot. So these tiny little cards, these tiny little things that are affecting the performance of the site add up. And so we have, you know, 100 or 150 decisions that were made that aren’t necessarily great decisions. Those may have a very minor impact on their own, but when you add them all together, it’s a major impact. Right? So yeah, if you have 40 different things that have, hundred millisecond impact on page speed, 40 things at 100 milliseconds each, that’s a four second change, right?…
Jessica: Yep. So if people click away when it takes like four seconds to load for a website.
Zach: Yeah, you have three seconds. That’s the statistic that most, people, including Amazon, have found, is that you have three seconds to make your, your first impression, and actually you lose about 40 to 45% of your market if it takes longer than two seconds for your page load. Oh, wow. So, you know, that’s just it’s huge. And so yeah, when you’re in a business where the entire site is transactional, the goal is to sell products or sell services or sell downloads or, you know, anything that is transactional in nature. And you have all these little decisions that are having to gather a large impact. The biggest thing you can do, the best thing you can do is start to reverse some of those decisions. So, when a lot of people start out, they start out wanting the site to do everything right. I want to process my orders through the site. I want to ship my orders to my customers, print my shipping labels. I want to do all of these things from inside the site. That may not necessarily be the best thing to do at scale. So when you take things and you move them outside of WooCommerce, that don’t necessarily have to be done inside WooCommerce, you free WooCommerce up to do its core tasks, which is show products to customers, let them add it to a cart load, then go from cart and checkout, let them pay you. Those are the most important functions of WooCommerce. There’s four things right there, and anything else that you’re doing that doesn’t support those things is something that can be done outside of WooCommerce just as well as inside. So now the the optimization strategy for me is generally let’s get down to the basics. Let’s make sure the basics of the site work really well. And then once we’ve done that, let’s take a look at what we need, what we’d like to have, and what we can eliminate completely. And let’s just build it out of what those things are. You now have column four need to have it’s business critical. Nice to have. It’s not business critical but it supports the business well. And we like to have it and things we just realize we added and don’t necessarily add anything. The bottom line get rid of the entire don’t need to have column. And you might have eliminated, you know, 20 cuts of your 40. Right. And then you go through and the nice to have. So okay, how can we do this in a way that doesn’t impact the site performance. And then you take those and move them to external services, SAS products, other things. Right. And so one one of the big things we did with Oscar Mike is we moved to an external shipping solution. And that was, you know, at the time we were using Aurora, which is a company out of Boston, great company. You know, Austin, Texas, a lot of fun. Great team. And we were using them as our order processing, so we didn’t log into WooCommerce anymore to ship an order. We were doing that in an external tool. And by doing that, we eliminated a whole bunch of load on the server for doing the order processing on the back end. Because the only thing that had to happen was a simple API call from them back to WooCommerce to say this order shipped. And here are the details. And that was it. So rather than having people in the WordPress admin making changes and so these these lessons, you know, later on led to running even larger sites that Oscar like significantly larger some of the the largest WooCommerce sites that existed in the world before, before 2014 when I left my my agency and yeah, we had the largest WooCommerce subscription site as a customer for a while. That’s Universal Yums. I believe they may still be the largest, but at the time they had 150,000 people on their their subscription boxes. That’s around. Yeah. And the orders renewed on the same day every month. So it wasn’t it wasn’t is seen thousand orders that had to go through all on that day. When they first came to us that that process took seven days of cron jobs. Oh wow. And through some very great work that, that, folks, on the team did like job or divorce ski. We brought that down to under 18 hours, and, there’s this was before high performance orders, storage before any of the optimizations had really been done to action Scheduler or, any of the other things that that sit behind the scenes that make all those things work. So, getting that down to one seventh or less of the time that it used to take, had a huge business impact. Yeah. And then we had a grocery chain that we worked with that had, hundreds of thousands of products because it was like a full grocery store. And they had market based pricing and market based inventory, and we made all that work. We had, a deodorant company, that, started on, the founders dining room table and grew to have the 2 million person email marketing list. And, we were keeping their servers up and running, you know, so we were able to take those same basic lessons, that I used on a much smaller site and apply them, across numerous other sites. And, it still stood the test of time across all of those. And of course, that wasn’t all just me. I had a great team of people supporting me and doing all that. And the in the and I actually had my keyboard taken away because I hired people that were better at coding. I was, so, yeah, we we got to a point where I was just kind of running the administrative side of the business and, and the sales and marketing. But we were still, you know, fundamentally doing the same things we were optimizing sites for performance and, eliminating some of those thousand cuts that slowed the site down. One by one. So that’s, that’s the key.
Jessica: That’s very interesting. So how do you decide for a client? What is a nice to have when it was a when what is they don’t even know or don’t even really need to? I mean, some of the things are very obvious that need to happen, but like, I think oftentimes clients are just really wanting to have some things because they know it’s on another shop that is their inspiration for their online shop. It just so happens so often, how do you make those decisions? Like, is it anything you try to convince your clients to not do things, or do you also sometimes maybe had to give in and say, okay, we are doing this specific thing because you’re asking for it, but then we do not we don’t we do not do these three things, for example.
Zach: Yeah. So a lot of it is just getting to a point where we have a deep understanding of the customer’s business. Right. So, this is not a simple transactional relationship in any of these cases. Right. These are much deeper relationships than, you know, maybe a smaller agency is used to, because you have to understand when the order happens, even even beyond what the website is doing, how is an order processed, how is it picked, and how do they warehouse their own goods? What are the things that happen after the sale? Yeah. How is how does shipping tracking work? How do customers get those updates? How does inventory management happen? How or how is cycle counting happening to make sure that inventory is accurate? If you’re using third party logistics companies, how do they get the orders? Now, where does that integration lie? It does. Is that, a plugin that already exists or is it an API? We have to call and or or do we just send them an email and hope they process the order? How do you know what level of interaction we have there? So it’s really just about getting to a deep understanding of how the operations side of that business works. And, part of what we did do that was we would go on site with these e-commerce companies and, watch them do what they were doing, watch them hack orders, watch them ship things, watch them now and participate in that process so that we understood at a deeper level what exactly it was we could do to support that. And so that’s a big piece of it. The other thing is that it’s really important to understand that in these cases, the customer is, in the end, going to be the person making the decision. The store owner will tell you yes or no as to whether or not something is they need to have a nice to have or a we don’t need. Right. And they are the arbiter of whether or not that is the case. So your job is to present your case for whether or not something should be in one of those categories in a way that is delicate, with understanding, with care, but also with a little bit of a, with a little bit of force, sometimes a little bit of, you know, trust in yourself and, and whether or not you’re making the right recommendation, so you can’t be afraid to push back a little if you know that something probably could be done better. But part of knowing that is you have to know how it can be done better, too. So, and I think we, because of the formation of the team that we built with my former agency, multiple members of the team had been embedded in e-commerce companies that actually run operations in e-commerce companies. So we understood at a deep level how these various systems needed to work together. So if you’re an agent singer, just getting into that space, I think, if you have customers already that are e-commerce customers, go out and work a few days with them. Just go out and pack orders and ship things and, you know, do those things and, you’ll gain a much deeper understanding and honestly, a much deeper appreciation for what exactly it is they’re doing.
Jessica: And I think it also helps understanding their processes, like entirely and make sure that, okay, I now understand how they are working. Can this, WooCommerce plugin, for example, really, like is it a good fit for this process, or is it maybe something else that, needs to be chosen in this context?
Zach: Yeah. And you know, the other key here is to know that the WooCommerce team is actually performance focused at this point, right? They they spent a long time trying to serve the small, you know, mom and pop store owners because that was that was where the most growth was. And you know, it was on an understandable strategy at the time. And I was there, and agencies like ours were they’re running these much larger sites and the WooCommerce team just wasn’t focused on making those much larger sites work at that point. I hadn’t quite the same level. Right. So there was a realization over time that things like high performance or storage were needed, because, you know, inserting an order with the traditional way that WordPress does things, you can’t bulk and post matter in WordPress, right? So, yeah, post matter is a single call per record and an order contained a minimum of 50 posts about it calls. So inserting an order was 50 separate database cells,
Jessica: 51. Yeah. It’s for for the order itself. And then the post middle on top. Yeah.
Zach: Yeah. So you look at that it’s like okay at scale. Well let’s say we’re getting 100 orders a minute.
Jessica: Oof. Yeah. It’s a lot of database requests.
Zach: Exactly. So high performance orders, storage reduce that to three. Yeah. The order of the order itself. An initial record. And then, you know, all the post-match gets inserted into a single record now, and then plugins may add more to that. Right? But the core is these three calls instead of over 50 now. And so, you know, that performance difference means that with even you know, it was less hardware than before, we can handle more traffic because high performance or storage does the work. The the problem is and you know, the, the foundation that led to high performance or the storage being a possibility, they first introduced the CRUD classes create, retrieve, update, delete classes inside of WooCommerce in 2017, and that abstracted away the way the data was interacted with by WooCommerce core code in most cases, and then it took from them until 2024 for plugin authors to get the message that these CRUD classes, if you’re not using them, and we replace the way one of these data, types is stored, everything’s going to break. So they had to have an outreach campaign with plugin authors saying, we need you to use these Crud classes, because if you don’t, once we make high performance order storage, the default for new stores, which it is. Now, if you install the WooCommerce plugin, as a fresh install new store, high performance or storage is on by default. So you know that that’s. Yeah. Yeah. It’s not for old stores, right? You do have to migrate with it. But by all means, if you’re listening to this and you have an older WooCommerce store, you haven’t migrated to it. Yeah. Work with a team that understands high performance or storage, test it in your staging environment and make the switch because, your growth ceiling goes from here up to here. Right? So, that’s it’s an important thing. And so we we really, we had to wait seven years, though, for the industry to catch up to what had been done in 2017. And, if anybody’s been following me and my podcast with Do the Woo for the last seven years or, you know, listening to me talk in the marketplace in general, you’ve heard me say that plugin authors need to update and support all of that so that these things work. But it was even a problem internally. Like there was once all that stuff came out. WooCommerce had its own reporting inside of WooCommerce. Didn’t use the the CRUD classes for that first. So, you know, it was just a it was an uphill battle. The change was there. You know, the foundation had been laid. We still needed to build the walls. And then once the walls were built, we had to get all of the other people to come in and take care of, you know, drywall and, and mudding and, you know, laying carpet and tile and, putting in the appliances. Right. We had it took seven years to build this house. So we’re finally at a point where now we have the house. Right.
Jessica: So it’s at the end. It’s all for performance, essentially. Yeah.
Zach: Yeah. So, you know, that that was kind of where we were going with that was the WooCommerce team has a performance team inside of it now that is solely focused on performance across the board for all store owners. And so that’s a huge thing. And it’s really, really helpful, to have them, you know, standing with us now, helping make sure that these larger scale stores are successful. And I can tell you from conversations with leadership inside WooCommerce, that is absolutely, where they are now. They they want, you know, source to be able to operate at scale, and they want, you know, support that the best that they can.
Jessica: So, so with, yeah, WooCommerce itself also focusing on performance, I think it’s also worth to mention that it’s not just like WooCommerce that should, look after performance like everyone has to do who creates websites, whether you’re a freelancer or an agency. And, I remember that especially when working with WooCommerce, like, caching is a bit of a pain in the ass. A Few will say, so, and do you have any tips which database and caching strategies actually work for larger WooCommerce stores?
Zach: Yeah. So there’s there are a few things that work really well. One, if you have page caching enabled, obviously it’s going to help with the less dynamic portions of a WooCommerce site. There are points where page actually work, like the cart, the checkout, you know, anything that’s in that dynamic, process that happens, you know, at the end of the sale, when the cart itself can’t really be cached. So the session making sure the session integrity, across everything that’s gosh, is is solid is important. Yeah. So my first recommendation is always to make sure that the site is performant without caching on, because putting a bandaid on something that is bleeding, if Band-Aid isn’t big enough, it’s still going to bleed. So yeah, we we want to make sure that everything is as performance as we can make it. We’ve optimized deeply, and then we can layer things like caching on top. And so now there are multiple ways we cache within WordPress. There’s no object caching which caches, queries to the database. Right. And we do that a lot of the time with Rattus or memcache. You know, I personally I’m a fan of around this object cache. And so Cruz has done great work with Object Cache Pro and with relay and really optimizing the way the object caching works. WordPress. So those are some of my recommendations there. Then we have page caching, which caches the results of what PHP has, has executed. Right. So the PHP has done executing. We’ve generated what becomes of that more exact page. We can cache that result so we don’t have to go back to the database to make that same thing again. And then the third layer of caching is edge caching. So we can use a content delivery network and we can cache the results of these pages at the edge. So using something like a Cloudflare Barnes ETN or yeah, some of these other resources where we can cache some of the site or yeah, the static page cache, even in some cases, out at the edge close to where the request is coming from. So, the way the content delivery networks work and, you know, primarily is we have an origin server, then we have edge servers and multiple points of presence around the globe. And so when you deploy to an edge cache, those points of presence, get a copy of those resources. And when somebody requests that resource from close to one of those points of presence, rather than going all the way back to the origin server, multiple hops away from where that request is happening, we go to that edge server that’s closer, and we get it from, you know, 1 or 2 hops away from where the visitor is. And so that that can really optimize that process as well. There are other things that you can look at. But as far as caching, those three layers need to operate, you know, effectively, but they also have to operate in concert. They have to work together. So yeah, if you update a post or a page, you want to make sure then that when that page gets updated, when the hook for that happens in WordPress, whoever you’re using or whatever stack you’re using, listens for that hook and invalidates that cache across all three layers. Because if it doesn’t, you end up with stale content. Stale content leads to problems, right? So, so those are really the keys there. Those three layers do work really well together. The other thing on the database side that I will mention is that, there is a point where a single database server may not be effective anymore. Right? So when you get to a certain scale, you may need to horizontally scale your database servers. And there are a number of ways to do that. Most of those ways result in something called replication lag. And replication lag is really bad on an e-commerce site, because if an order gets inserted into one of the database servers and then the next request comes in to display the confirmation page that confirms the order by querying the order. And in that time period, the traffic is dictated that the load balancer move that request to a different database server. That database server doesn’t have the order. Well, what’s going to
Jessica: there is no confirmation
Zach: … and there’s no confirmation the page error is out because the record doesn’t exist. So there are a number of ways that we can approach this. The easiest way I, in my opinion, is to use a Galera cluster, which, allows you to have a master master set where, a write has to happen to all of the replica servers at the same time. That way everything stays in sync. But you still get the benefit of being able to distribute load across multiple servers. So, that’s kind of the best case scenario. So, for handling data at scale. But, you know, as long as you have those kinds of things in place, you’re going to be able to achieve tremendous scale, with these, with Hammerstein’s.
Jessica: And I think it’s not just like about WooCommerce sites, but also larger websites in general. So anything performance related can also be applied to a website that is not running WooCommerce. But like instead, I don’t know, has maybe is a multilingual website or is, like spread across many countries, or many markets, different markets even doesn’t have to be a 1 to 1, translated website. So, in case everyone notices that we’re just talking about WooCommerce, it’s like this is where, websites scale up massively. But they can also apply to anything, that is not a WooCommerce website, but still a large website for whatever reason. And yeah, in the in the business, the business world, there’s like lots and lots of, different websites that are not WooCommerce websites, but still, maybe have their own custom system for whatever reason.
Zach: Yeah. I mean, there’s there’s a ton of, of things that I say about WooCommerce that just apply to WordPress in general. Yeah. You know, we we generally deal with those things more in WooCommerce than anywhere else, but, large learning management systems, large content sites, you know, millions of posts, these are all things that, that have similar, load, you know, needs or, hardware needs. So, you know, the other thing that I didn’t mention before is there is another way you can cache some data that people don’t always think about when they’re they’re thinking about strategies. And I’m tied up as an open source plugin called Elastic Press that allows you to use elastic as a way to store, data. You know, mainly posts and, that are in your WordPress posts, tables and the thing that elastic does really well is it does search really, really well. So, SQL is not really great at search. It’s, it’s great for creating relational data. It’s great for storing that. But if you are searching that data deeply, it doesn’t do that incredibly well. It’s not efficient.
Jessica: That’s true.
Zach: So elastic on the other hand, is really good at searching across millions of records very quickly. So when, you know, when we’ve been working with sites that have a ton of, a ton of products or a ton of, posts, tons of content, if you put that into elastic, it’s more searchable. But the other thing that elastic perhaps does that people don’t really talk about as much is that it overrides the WP query. So any time a query would be happening to a post, it replaces that with an elastic query. And so even your category pages and other things that you can reliably query on the site gain the benefit of those much faster search results. So you can go from searches that takes seconds to searches that take. Now 100, to 150 milliseconds. And that drastically changes the performance of the site.
Jessica: Yeah, that’s a lot. That’s a huge change. That’s true. Yeah.
Zach: So it’s just it’s cool that there are so many tools that we can employ, for things like that.
Jessica: Yeah, that’s that’s great. And performance isn’t just like, one dimensional. And as you said, there are many different ways of improving the performance. And I think this is still some something that we need to still mention because I often also get to hear, yeah, but I need to improve the performance of my site, even if it’s just a freelancer and it’s like, yeah, it’s not just a plugin. You install a button, you press, then your site’s performance. It’s unfortunately not as easy as that. So, yeah, but that’s very interesting. I didn’t actually didn’t know all of this about Elasticsearch, that it replaces the queries like all the queries. So it’s good to know it’s you.
Zach: Yeah, it’s it’s a really cool side effect of, of using it is that it just speeds up all the, all of the queries that are querying the, those tables. So yeah, pretty cool.
Jessica: Okay. So let’s move on to workflows. A little bit. So what does a healthy deployment process look like? When you’re managing a lot of a lot of sites. Do you have some suggestions there.
Zach: Yeah, actually I do. So there are a number of things that we can take from the more traditional software development, workflows, you know, the, the software development lifecycle, that a traditional, developer would use. And so things like, you know, continuous integration, continuous deployment, you know, CI, CD tools, these, these help out a lot. And it is as long as you are managing, you know, as you grow your agency, you’re going to have more developers that are attaching the same files. So things, having things and version control as agencies grow, being able to track where changes are coming from, being able to handle conflicts when they and availability arise when two people work on the same file. If that’s not in a version control system and you are cowboy coding directly on the server, you end up with people overwriting each other’s work. And, I will tell you that that is a terrible waste of time and resources and very frustrating to both of the developers. You just spent time working on a feature?
Jessica: Yeah. That’s true. That is true.
Zach: Yeah. So, you know, having having some kind of version control system, whether that’s GitHub or, you know, subversion or, you know, any of the git related tools, you know, you can use GitLab, you can use, you know, any of the other hosted, solutions there. But having a version control system managing change matters a ton. And then going from version control to a tagged release and using that tagged release as a, as a deployment trigger, and that deployment trigger being the only way code gets production. So everything else can happen in staging, right? Let people deploy their new code to staging all day long, you know, or bill individual staging environments for each branch. And I have dev staging environments for each branch. But, and that’s probably another thing that should be talked about at some point. But, you know, get flow and branch, branches for each feature, having some kind of way to track feature development and, and have things, you know, in these compartmentalized commits inside of version control man matters to you. But yeah, then deploying from your version control repository using a continuous deployment tool, and then having that push to production, that’s the enterprise workflow that we all, you know, dream of, right, is the. Yeah, the ability to work with a large team and use all those tools to funnel into one release. That’s a coordinated release of all these features. And yeah, we’ve tested that all these features work together. They’re not there no conflicts. We’ve, you know, gone through and and run all of the, automated testing because we wrote test cases and all these things are working correctly. And now we can deploy and when we go to production, having that be the only way that code changes in production is really helpful. That does make it harder for, say, clients to update their own plugins. If you don’t have a way to pull those changes back and, but, you know, there are a couple of different ways to handle that. One is don’t version the plugins that you know they’re going to need update. That works to a point. The other is to just make sure that you constantly are keeping, your development environments in sync with production for plugins by watching what updates have happened. Now have something watching the the WP content plugins folder on the server and notifying you when anything changes. That way you can go and grab those changes back and then integrate them into your dev branches. So those those things make sense. I like showing those updates down completely and just committing them through version control, but that’s the better way long term. But for it’s not as it’s not as manageable on the client side. So yeah, it’s true. There just multiple ways to handle that. There are some tools out there from, from companies that allow you to look at, you know, multiple sites in one, one environment and kind of see everything that’s going on, across multiple sites and in one location. I mean, there’s this thing called Greyd.Hub, I guess that, that does that, too. You might have heard of it. That’s, you know, being able to manage plugin updates in a, in a quick and easy location like that is really helpful. Yeah. There are other tools that do the same thing. But maybe. Yeah, in a different way. And I find, I find that those tools are really helpful for managing on a large number of sites as well.
Jessica: Yeah. I think it also maybe depends a little bit on the client, themselves. Like, do they want to do updates, on their own or are there to completely fine to have, have them version controlled and be updated for a fix, let’s say every week or every two weeks or every month, depending on like how much is still going on on their website. And I think it’s also like the website itself. Some may need very regular, maintenance, and some are fine if they’re just being there for months and, not being up there. Of course, for security updates, you should always go, super quick and have them implemented. But I think it’s very much comes down to the what is the project actually like. And yeah. It can depend there as well. What is the best solution to to maintain the website.
Zach: Yeah. I mean, and on the security side, you know, having, having things like Patchstack in place helps a lot, to ensure that things like the zero day exploits that come out, are already patched, right? That’s what Patchstack does. They release tiny patches that fix problems and plugins that have not been addressed yet. And by having that kind of rapid response generally, you can beat the, the bad actors to even trying to attempt to use these on your sites. So it’s just yet another layer of security that you can add that, just provides more defense. Yeah. And then things like blocking that bots and all of the other, you know, things that can be done at the edge. So the server is never even touched by bad actors, with rules like Cloudflare Enterprise and just Cloudflare in general. Very helpful, very helpful. So yeah, I think those are, are key to, making sure that, you know, your customers are has supported and secure as possible at all times.
Jessica: All right. So now maybe let’s move on to the topic of partnerships. So partnerships with of from agencies to let’s say hosting companies. And how can agencies turn, hosting from a cost factor to a business opportunity.
Zach: Well, now you’re speaking my language now. We’ve we’ve been on the same page the whole time, but, yeah, this is so this is what I do now, right? I, I run our partnership programs at BigScoots. I’ve been involved in, agency partnerships now in the hosting space for a while. And, what I find is that the biggest thing I was looking for as an agency owner is something that’s honestly kind of difficult to find. And that is a true partner. You know, when, when I’m picking a partner that I want to work with, I want someone who is as invested in my success as they are in their own. And that’s a hard thing to find. And it know, honestly, is, a lot of these partnership programs that are out there are really just an affiliate program by a different name. And, you know, while, while the partner programs that I’m running right now do include an affiliate component. Yeah. Because that’s what creates the recurring revenue. We’re approaching it a little differently, and we’re doing that by making sure that that commission is not a single time thing. Right? We’re we’re invested in keeping that profit sharing component moving forward. So as customers are signed up through our agency partners, those agencies get a recurring commission. And that recurring commission is for initially two years, because they are agency partner. So and then as long as they remain active with us, we’re planning to extend that beyond that. So, you know, commission structures are great. That’s how you turn, you know, hosting from just a, a very simple, no cost line on your, on your piano to something that can be considered an asset and, profit center for your, your agency. But the other thing that really does matter is part of that profit analysis for hosting is how much time is it taking of your day to manage that relationship? Right? Are you still having to go in and fix problems for your customers when they happen? If they’re things that the host could very easily fix themselves? And so that’s why, you know, things like managed WordPress matters so much for agencies because, a manage WordPress provider, a host can take a lot of the load off of the agency for managing that customer relationship on the infrastructure side, and know some hosts go above and beyond that and, do other things that are ancillary to hosting that help with those agency relationships as well. For example, one of the things that we offer is a performance services package where, we’ll monitor core web vitals and make sure your customers say, in the top percentages of, the core web vitals, you know, sitting in that 95 and above percent, score. And, you know, those, those core web vitals and making sure not only do the scores look good, but in performance, you know, in practice, the performance is actually matching what the scores say because sometimes the scores are more of a vanity metric than anything. But actually making sure that they’re really doing those things, and doing that on a monthly cadence and an ongoing basis with alerts that tell us when things change. So now those are ways that we can support, our agency partners beyond just, you know, we’re a really good host with great support. So those are the things that I looked for when I was an agency owner. First, how good was the support? And when I, when I wrote in and in chat or if I called or if I opened a ticket, did I have more knowledge about what was going on than the hosting companies support? People? Did. And there were a number of instances where I did have more knowledge than the host company did, and that may not have been a good partnership for me. So, you know, how quickly does it take to get from in in an environment where there are tiered support groups, you know, to get from that entry level to somebody who can actually diagnose and fix your issue, if you have to go through three levels of support to get there. And each of those transitions, each, each time that the, you know, ticket is forwarded on to another level of support, you have to wait an hour or two hours. You know, if you’re if you’re in an a, an emergent situation where the site is down and what is it taking you six hours to get to the right person to fix that. That’s a it’s that long. Yeah. Right. That’s that’s a problem if that’s the case. So Yeah. What’s the what’s the SLA for support? And one of the things that I’ve loved as I’ve, gotten into my role here, BigScoots is that. Yeah, we, we have a 90 second SLA for support tickets. So, yeah, responding to a support request in under 90s is the expectation. And it’s actually the norm. We have how we consider support to be a very important thing. So I found in the places I was looking to work, I was looking for the same dedication, support that I had wanted when I was an agency owner. And I believe the last two places I’ve been really both embodied that, you know, Convecio support was, a very important thing. And now excuse support is the most important thing. The way we treat our customers is the guiding light for the entire company. And, it shows, you know, when you have a Net Promoter score of 95 for the year as an average, and some of your support reps have a 100 for the year, which is nearly unheard of. It really shows that culturally, support matters here. So yeah, even if you’re not with us, I want you to look at support as an agency owner is one of the key things. So support stability speed and scale supports first. Stability is second. What’s the uptime look like. Kind of keep your site up and running even through these massive traffic events. You know security everything and keep your customers secure. Are they going to keep your customers customers secure? Right.
And then speed. Yeah. Are the servers fast? Do they, you know, can their hosting handle the workloads that you’re throwing at it. And those four things are the, the assets that I think about when I’m looking at, hosting partnership. And so and you really are looking for a partner if what you’re getting is a vendor, you don’t have a partnership, right? If the people you’re talking to, if the people you’re talking to when you have an issue are sales people, if the only contact you have with the hosting company is with a department that is a sales department, and they’re just telling you to spend more money, that’s not a partnership, that’s a vendor. So we strive to be a partner, not a vendor. And that’s, if you look back, even back in 2015, when I gave my first WordCamp talk, that was one of the points on there that I, they used was find a partner, not a vendor, and now it’s still rings true today.
Jessica: Yeah, I think it’s it’s very important because like hosting is like if you have no hosting then you have no website. So essentially it is the basement you’re working on. And the ground floor, basement, ground floor, whatever, you know what I mean? It’s a base that, that, the website is running on. And if it doesn’t work, yeah, you have not you, but also your client loses money eventually if they are having a shop on it or whatever, even if it’s just a regular website, if they can, if it cannot be reached. Then there’s just nothing you can do and no business can be done.
Zach: Yeah. And and I encourage business owners to look at hosting, in the same way that they look at physical space. So if they were trying to open a retail store, for example, you don’t want your retail store to be in a back alley somewhere. You don’t want it to be in some of these closet. Right? You know, I have to go through some of these living room. You got to it. Trust me, there are actually hosting companies that exist that it’s a server in the owner’s closet. Like, that’s the reality. It does happen. But they they build themselves as a full scale hosting company. Right. And that’s not frequent. It does happen, though. I’ve seen it before. So when you’re looking for a retail space, you’re not looking for this back alley, nondescript door you’re looking for, you know, the place on Madison Avenue in New York or on Michigan Avenue in Chicago, right? You want frontage. You want a place where people are going to be able to get to your location, and they’re going to be able to come in. And it’s beautiful right now, things are efficient. They run well. And when you’re looking at hosting, really you’re looking at digital real estate. And then that digital real estate matters. So how do you want your digital real estate to be in some back alley, or do you want your digital real estate to be at a premium location and a tier three data center that has, you know, all of the, the power back up and redundancy capabilities that you need? Do you want a host that, understands the the layout of their infrastructure and at a deep level, because they run their own servers? Or do you want to, you know, rent space from somebody else who’s renting space? And this is something that I think is really important distinction to make, because a lot of hosting companies, they’ve built on top of Google Cloud or AWS, and so they’re renting space from Amazon or Google and then reselling that space to you. You are subletting your hosting environment. And so if you wouldn’t do that for your your physical location, for your business, why would you do that for your virtual location? And so, you know, it’s just you’re just looking at those things now. And so I have, you know, the privilege of working for a company that has their own data centers. I have badged biometric access to go look at the server racks that are our sites are in, we have, locations in the Chicagoland area and Chicago. We’re downtown Chicago by McCormick place. We have a location in Elk Grove Village that we just spun up, which is just outside of where O’Hare International Airport is in Chicago. We have a location in Virginia that we’re we’re setting up. We have presence in England. We have presence in Germany. The that we’re expanding. So but those are our servers. It’s our hardware. And because of that, we can physically locate a physical database server next to your physical web server network that you you gather directly at 400 gigabit per second on a 400 gigabit backbone. That’s, that’s, you know, that’s tied to, the best, data providers in the world and goes through Cloudflare with a direct fiber connection to Cloudflare’s network. So these are things that. Yeah, when you’re renting from somebody who’s also renting, when you’re subletting your server space, you they can’t do that. You know, these these companies can’t call Amazon and say, hey, hey, AWS team, can you physically put this server next to this server? Just not what it’s built for. That’s not how it works. So, you know, when you are running a premium business and you want premium digital real estate, you have to find a premium provider. And there are very few of us left that run our infrastructure.
Jessica: So it’s a great idea to look specifically out for partnerships that does provide this real estate. And as you said, it’s not just a vendor, but a real partnership because you, as a hosting company, just as invested in making this business that comes to you as a customer, help them also grow and get even better. Yeah. And that’s just like taking their money and they’re here are some servers. Have fun.
Zach: Yeah. I mean, it goes way beyond that. And, you know, if, if all you want is that transaction, that’s great. That’s fine. Yeah. I mean if that’s what you want to build your business around, more power to you. I did it. I did it for a long time. And I found hosting partners. I trusted. But it’s it’s just a lot of work, and that’s what you. If that’s what you love doing, if you love managing infrastructure, do it. But if you don’t want to, then find somebody you can trust to do it. And and there are a lot of great options out there for companies that are really good at doing this. Well, and now we’re not the only ones. Yeah. I’m a little partial because of where I work. Because I, I do think we’re pretty awesome, but, we’re not the only ones that do this. Well, and, you know, there are, there are benchmarks that come out every year that, are done by Gavin Ohashi from, Review Signal, the WordPress hosting benchmarks. And they’re a great indicator of where some of the performances from these, these companies, how can they handle load at scale? Yeah. So really good to look at as an independent, view into what these hosts and actually handle, and proud to say that after, you know, BigScoots has been around for 15 years, it’s only the last two that people have really been hearing the name because they grew by word of mouth for 13 years. And, now our chief marketing officer Tim started just over two years ago now. And so that’s when, you know, the push into the market has really started with actively marketing to the WordPress space. But they’ve been doing WordPress hosting for 13 years. And, and a lot of content creators, have been around for a long time. So, kind of cool to see, you know, the, the impact in the marketplace already. And, it’s just really, really neat to. Be a part of this journey and start to get, agencies to a point where they, you know, they have a partner, they can trust. So if you don’t have that yet, look me up. If, if you don’t want to work with us for some reason, look me up. I’ll. I’ll help you find somebody to you want to work with? Yeah, I, I have always been about serving the community, and that’s what I want to keep doing. So, you know, if we’re out, right, then I’ll help you find one.
Jessica: Yeah. That’s awesome. So I think we can wrap up here. I have one final question for you. And, this one is where do you think high volume WordPress hosting is headed next?
Zach: It’s interesting because one of the things we’ve seen really change in the last couple of years is, we’ve seen a lot of traffic coming in from AI consuming the content that we make. Right? And so, one of the largest traffic sources for sites is all of these bots now that are just seeing content. So I think that on one hand, one of the things we’re going to see high volume hosts doing is prioritizing, that traffic, prioritizing user traffic above these, these, and other services that are just consuming, consuming, consuming. And we’ll see, an increased focus on whether or not, those, those individual customers or companies want AI to be consuming their content. So, that’s, that’s an area of tuning that we’re really focused on right now is making sure that, you know, good bots can get in bad bots are not allowed, and that these, you know, really bad actors, because that some of these AI companies, they just don’t care how much traffic they throw at you. They, we’ve had customers go over bandwidth just based on, this AI bot traffic and logs are crazy. Luckily, we have a policy where we don’t paralyze customers for, you know, non sustained traffic like that. So we’ll help them block it at the edge that it stops happening. And we have a list of the ones that do bad things and you can automatically kind of add those in. Cloudflare has been doing a lot of work with, AI as well, trying to, you know, stem the tide of this traffic wave that’s happening. Because how do the point of AI systems is to consume all knowledge and be able to regurgitate it? Right. So, so yeah, that that’s the big thing is how do we how do we stem the tide of this AI traffic wave? That’s that’s one thing. Another thing we’re going to see, I think is an increased, focus on the way that performance occurs across the entire WordPress stack. So the WordPress team has a performance team the WooCommerce team has a performance team this year. Now hosting companies that are focused on performance optimization. And the hosting team inside make WordPress is been focus. You know, how can we create a stack that’s more performance, overall? And what can we do together to push the industry forward? So I think there’s this focus as an industry on supporting mid-market and our prize workloads is going to pay dividends across all sites, in the long run. And then I think we’ll see more specialty hosts that, that emerge that are really good at doing high scale WordPress. Before I was at BigScoots, I was Convecio CEO for 18 months, and now Convecio is containerized, WordPress built to scale horizontally. And why that’s. Well, that’s a necessary thing. You get to that level of, of scale if you can’t just deploy more hardware, which, you know, we can do really effectively, it’s one of the things I love. So you can deploy a 200 core server if we need to, you know, so that’s that’s kind of the other side of the coin is how where is that growth ceiling? How high can you go before you, you know, hit the, the limit of what’s what’s available. And with us, we can do a lot with a single server, with other companies, they can’t do as much. Or the single servers, they have to start to scale horizontally earlier. And so containerization helps a lot with that. So we’re going to see more of these specialty hosts that that are really good at handling specific workloads, at scale. And I think we’ll see the investments that are happening into things like WP cloud and yeah, WordPress VIP on the automatic side, really, cert to feed into the open source project more as well. Because they’re learning things that they maybe hadn’t known before as well, as they grow and scale across all of the automatic own hosting brands as well. So, yeah, I think we’re going to see, more growth than ever in high volume WordPress hosting and, I hope that we continue to see growth from the platform in general, providing, the content management system of choice to these large publications. You know, we’ve got a lot of large publications on us already. As far as WordPress as a platform. But I think there’s still more room for growth there. And yeah, with, with some of the tools that are out there, obviously, if you’re listening to Greyd Conversations, you’re aware of one of them that’s making publishing easier, for, for companies. But, there are a lot of people who are, are focused on this, this world of trying to enable, these larger scale businesses, franchises, other companies that have these enterprise needs to come in to the WordPress space and really thrive here. So those are the things that I think are coming, in the near future.
Jessica: Awesome. Thank you so much, Zach. So we’re at the end of our episode today. Thank you so much for taking the time and for this, wonderful conversation we had. And, yeah, everyone, see you in the next episode.
Music
Key Takeaways
Not all hosting is created equal – and cheap will cost you.
Ultra-low-cost hosting means hundreds of sites fighting for the same resources. At scale, that’s a ticking time bomb. If you plan to grow, your hosting partner must be ready to scale with you – not slow you down.Your hosting partner should care about your business, not just your invoice.
Sales reps are not support. If you’re still handling customer issues the hoster should be solving, it’s time to rethink the partnership. True partners invest in your success – and help reduce your workload, not increase it.Turn hosting from a cost into a growth engine.
When your host acts as a managed partner, infrastructure becomes leverage. Offload client management, automate maintenance, and free your team to focus on what truly scales – delivering results.WooCommerce can kill performance – fast.
It’s powerful, but resource-heavy. Every unnecessary plugin, slow query, or poor caching strategy adds up. You’ve got 3 seconds to make an impression. Lose that – and 45% of users might bounce.Caching isn’t optional – it’s survival.
For high-volume stores, caching is a layered strategy: object caching, page caching, edge/CDN caching. But remember: if your site only works with caching, you don’t have a fast site – you have a patched one.Deployment at scale needs grown-up workflows.
Version control, feature branches, and tagged releases aren’t “nice-to-haves” – they’re essential. When multiple devs work on the same codebase, conflicts are guaranteed. Staging-first deployment prevents disasters.You don’t know what you don’t know – until it breaks.
From missed core updates to poor SEO, slow response times, or inaccessible features – small oversights compound quickly. Surround yourself with experts who catch what you don’t see and make your team better.Guide your clients – don’t just follow.
Clients often want shiny features they saw elsewhere. Your job is to steer with care, confidence, and sometimes pushback. Be part of their process. Understand how they sell, operate, and scale – and build accordingly.Support that takes minutes, not days.
If you know more than your host’s support rep, you’ve got a problem. SLAs matter. Great hosting teams solve issues before they become your client’s problems. Sub-2-minute response times aren’t fantasy – they’re baseline.Hosting is digital real estate – own the lease.
Don’t build your business on shaky ground. Hosting isn’t just infrastructure – it’s a foundation. Agencies need hosting that’s stable, secure, scalable – and backed by real support. Not just a shiny landing page.The right tools turn chaos into control – especially at scale.
Running dozens or hundreds of sites? You need infrastructure that keeps up. Tools like Patchstack help you patch plugin vulnerabilities before they break production. ElasticPress supercharges search for large WooCommerce or content-heavy sites. Greyd.Hub lets you manage multiple projects, users, and updates centrally. Scaling isn’t just about servers – it’s about systems that work together.High-traffic WordPress is entering a new era – and bots are crashing the party.
AI tools and bots now generate a huge share of web traffic – and not all of it is welcome. Smart hosts are starting to prioritize real users over bots, throttle abusive AI scrapers, and help site owners control who gets to consume their content. At the same time, the WP ecosystem is doubling down on performance across the stack. WooCommerce, WordPress Core, and specialty hosts like BigScoots and Convesio are building for scale. Paired with tools like Greyd.Suite, high-volume WordPress is now ready for enterprise.





