In part two of this conversation, Jesse Friedman and Wes Tatters dive into performance benchmarking and WordPress ecosystem challenges. Wes explains Rapyd Cloud’s sponsorship of Orderly Apes, Kevin Ohashi’s open source fork of K6 performance testing tools, created after Grafana commercialized the dashboard and made ongoing WordPress hosting benchmarks prohibitively expensive. This project exemplifies open source values by returning critical testing infrastructure to the community.
The discussion explores why successful community and learning sites require fundamentally different hosting approaches than traditional shared hosting. Wes describes customer journeys where sites get churned multiple times as hosts try solving concurrency problems with bigger plans rather than optimized infrastructure. He explains PHP worker limitations, the dangers of legacy hardware marketed with correct specifications but outdated performance, and why customers want simple dashboards with enterprise capabilities rather than complex cPanel configurations.
Jesse and Wes examine uptime expectations for membership communities where thousands of paying customers depend on platform availability. Unlike e-commerce sites where downtime means missed sales, community platforms create direct customer satisfaction crises when unavailable. They discuss increasing AI-generated security threats, including plugins with vulnerabilities baked into AI-generated code, and why RASP tools and behavioral monitoring outperform signature-based malware detection.
The episode tackles WordPress’s competitive positioning against SaaS platforms. Wes describes how agencies must pitch WordPress benefits before pitching their services, while SaaS competitors skip directly to feature discussions. They explore WordPress advantages including data ownership, infinite customizability, plugin ecosystem interoperability, and cost structures that scale favorably compared to percentage-based SaaS pricing.
The conversation concludes with WordPress scalability challenges including database architecture limitations, unindexed meta value columns, and dropdown performance with large user bases. Wes explains his active involvement in WordPress make hosting team, the new test runner project, and efforts to create hosting directory metrics. Both emphasize how contributing to WordPress core creates holistic benefits beyond recognition, improving customer experiences while strengthening the entire ecosystem.
Links:
Performance and Testing Tools
- Orderly Apes – Open source performance testing dashboard forked from K6
- K6 – Performance testing tool acquired by Grafana
- Grafana – Monitoring and observability platform
- WP Hosting Benchmarks – Kevin Ohashi’s WordPress hosting performance reviews
- Review Signal – Web hosting reviews and benchmarks
Platforms and Products
- BuddyBoss – WordPress community platform and social networking plugin
- The Happy Pear – Vegan lifestyle community that started on BuddyBoss
- Fluent Communities – WordPress community platform with BuddyBoss importer
- SureCart – WordPress e-commerce plugin
- Fluent Cart – WooCommerce alternative
- LearnDash – WordPress LMS
- Tutor LMS – WordPress learning management system
- LifterLMS – WordPress LMS platform
- Elementor – WordPress page builder
- Advanced Custom Fields – WordPress custom fields plugin
WordPress Community Resources
- WordPress Make Hosting Team – WordPress hosting contributor team
- WordPress Make Slack – WordPress contributor Slack channels
- WordPress TV – WordCamp presentations and talks
- Five for the Future – WordPress contribution initiative
- WordCamp Europe
- WordCamp Brisbane
- WordCamp US
- CloudFest
Security Tools
- WP Scan – WordPress security vulnerability database
- PatchStack – WordPress vulnerability protection (mentioned in previous episode)
Transcript
Wes Tatters: There might have been one person visiting your site during that one minute window when we had to reboot your server. In our space, customers are expecting a lot more, which means white glove everything.
Jesse Friedman: We actually have really high integrity hosting benchmarks so that you can get an idea of exactly who’s providing some of the best hosting, specifically around performance.
Wes Tatters: The more users that come on, it grows and it grows. And you get this quite rapid escalation from a point where, oh, we were holding okay at 25 users and we hit 50 users, and the performance, and then you hit and things are starting to fall apart.
Jesse Friedman: It’s not just the hosting. You can take your entire platform and you can move it from one host to another, but you’re also diversifying the code that’s behind it, the technological teams that are behind it, all these things so that you’re not relying on this one place. And then if God forbid something does happen with that SaaS product, I don’t know what that even looks like, getting back out of it again. Welcome to Impressive Hosting, where we uncover the core tenets of great WordPress hosting. I’m your host, Jesse Friedman. Before we dive in, remember to check out impressive host. It’s where you can comment on episodes, ask follow-up questions, submit questions for upcoming guests. You’ll also find all our links for you to like, subscribe, and follow wherever you listen. With me today is part two of our conversation with Wes Tatters. As the managing director of Rapid Cloud, we are having such a great conversation that we needed to continue this and keep going. We’re talking about performance and concurrency for really complex sites, the way in which Rapid Cloud is helping to solve those problems. Before we jump in and pick up where we left off, one of the things that I wanted to ask you is, are you aware of the Orderly Apes product built by Kevin Ohashi to help with performance testing? I think you are. You said you actually sponsored it.
Wes Tatters: Yeah.
Jesse Friedman: Tell me, you know, what do you know about that product and why should we be paying attention to it?
Wes Tatters: The biggest product in the performance testing space is K6. It’s a really mature software suite that was acquired by Grafana. And the entire product was exported into the Grafana space. Which meant for people like Kevin who do things like the WordPress performance review…
Jesse Friedman: Benchmarks and he runs Review Signal.
Wes Tatters: Yeah. For people like him, it’s gonna cost hundreds of thousands of dollars to use the new Grafana tools inside K6 to generate one year worth of reporting. And a lot of that was gonna come out of Kevin’s pocket.
Jesse Friedman: Right.
Wes Tatters: We had a number of conversations with Kevin. Obviously I’m very passionate about the performance space, as is Kevin. And we had a number of talks about it and he was going, I’ve got an idea for us to fork K6 and create an open source version of the dashboard. And that’s really what the Grafana component is, the dashboard. Orderly Apes was the name that he chose. And he’s been working with a team on that for about 12 months now, I think since CloudFest last year, 2023. He’s been working on that product and announced its official release, I think last week or the week before. What that means now is that there is a truly open source version of K6 available. People that are serious about doing their own hosting platforms can launch the equivalent of K6 Grafana on their own infrastructure, without the cost, overhead and burdens of the Grafana platform. And look, it really is what open source is about. I mean, this is it. K6 was an open source product. Grafana created a really nice set of dashboard metrics for it, but then chose to commercialize those. The opportunity with open source if a team of people are willing to step up. And in this case, Kevin and people like ourselves who sponsored that product, can re-fork and re-release that product back into the open source community. We are very proud to be a part of that.
Jesse Friedman: Yeah, I think that’s really great. Really awesome work that Kevin’s doing to make sure that we actually have really high integrity hosting benchmarks so that you can get an idea of exactly who’s providing some of the best hosting, specifically around performance. I think one of the great things about Kevin is his level of integrity. He’s very stringent about the way in which these tests run. And he does it to make sure that he’s doing it in a way that, you know, because if you look up online, right? You go to Google and you type in like who are the top 10 best hosting companies? It’s really a top 10 list of like the best affiliate payout…
Wes Tatters: Yeah. Or who’s bought the most ads this month or who’s pushed the most leads to G2 or Trustpilot or any of those sort of things. And there’s a lot of misconceptions about performance in the WordPress space.
Jesse Friedman: Yeah, and I think that Kevin really tries to paint a very clear picture there. So, if anybody’s interested, go to wphostingbenchmarks.com/orderly-ape, and take a look at what he’s building there. I think it’s something that’s really valuable to any hosting company or anybody who’s looking to evaluate their current performance. So, let’s dig into where we left off. Wes, one of the things that we touched on here is that you’re providing in a way niche hosting that is specific to these concurrent type sites, but it’s not just from an infrastructure standpoint, it’s also from a marketing standpoint. If you go to the Rapid Cloud website, you actually have landing pages designed to teach people specifically the offerings that you have for these very specific use cases. And so I’m curious about how you are differentiating yourself from a marketing perspective, from other hosting companies and how those messages are resonating with your customers. Does it feel like they’re shopping differently for hosting these days and that they’re looking for much more specific solutions?
#### What’s the split between…
Wes Tatters: I think at the point where customers are getting into this success space, two things are coming into play. The first is their cost metrics are going up significantly. They are getting churned every month. They’re getting customer success calls going, guys, too many Ajax requests, or, hey, too many REST calls or, we need to look at moving you up to another plan. You know, repositioning you within a different product space in our platform. So that obviously has an impact on their commercial economics. But what they start to find in many cases is that they’ll churn a couple of times. Where they’ll go, oh yeah, okay, we understand that. But they tend to find that the churn isn’t solving the problem, because the underlying infrastructures are not specifically angled or designed. One of the metrics that we look at a lot obviously is this concept of PHP workers. You know, the most misunderstood concept in the WordPress space. And lots of hosts will, when you’re on a shared host, you’re lucky if you’ve got two or three.
Jesse Friedman: Yeah, sure.
Wes Tatters: But even on higher plans on a number of managed WordPress hosts that are out there, 40 to 60 workers is, we’re starting to get, that’s at the business end of their product space. But when you realize that a PHP worker has to be exclusively used for every page request. And that can also be a REST request, an Ajax request or anything else. And you’ve got a hundred people concurrently on your website, 50 PHP workers get eaten quite quickly, especially if those PHP workers are on a slower or older data center. Because taking longer to execute, which means that the window of time that that PHP worker has holding memory exclusively and holding a CPU exclusively grows. And the more users that come on, it grows and it grows. And you get this quite rapid escalation from a point where, oh, we were holding okay at 25 users and we hit 50 users, and things are starting to fall apart.
Jesse Friedman: It’s not the best hockey stick you want to see. You know it’s not.
Wes Tatters: Exactly. You know, J-curves are great if it’s revenue, they’re not as great if it’s customer satisfaction. So what they’re finding is we’re getting churned again, but if they’re gonna be churned and it’s gonna cost a thousand dollars a month and the customer satisfaction is gonna go up, great. But what they’re finding is it’s not doing that. The customer satisfaction unfortunately often starts to still fall off. So, that’s the journey that we’re seeing with many hosts that are coming into us and their customers have even done the right things. You know, BuddyBoss will do things like say, okay, you need an 8-CPU core 32 gig of RAM, you know, box as a baseline for a BuddyBoss site or a BuddyBoss app, you might need 16 CPU cores. So then they’ll go out and they’ll find a host somewhere that’ll sell them 8 CPU cores and 32 gig of RAM. Unfortunately, it’s 8 CPU cores of processing power in a CPU that was built by Intel 15 years ago. It is 8 CPU cores, it is 32 gig of RAM, but those 8 CPU cores are statistically a tenth of the speed of a modern CPU core. The memory
Wes Tatters: at 32 gigabytes of memory is probably DDR2 or DDR3 memory, statistically a quarter or a fifth of the speed of a modern memory core, say on a Zen4 or even a Zen5 DDR5 type memory component.
Jesse Friedman: And those nuances can become completely overwhelming and you end up getting analysis paralysis and you don’t know how to make a decision. It’s very much like I was trying to buy a gaming computer for my daughter, and it’s been a long time since I’ve had to do that. And so I was looking at like GPUs and the way in which they work, I just wanted an answer. I just wanted someone to tell me can I play this game? Right? And so it’s very much the same thing. It’s like, can I run BuddyBoss in this way? Yes or no? And so when you start shopping for the intricate details of the way in which your server’s gonna run, maybe it’s not giving you the exact answer. So maybe that’s where the benefit comes in, in you having solutions for those use cases.
Wes Tatters: It’s a specific solution. And I guess the other challenge is that when they get to that space where they’re talking dedicated servers and VPSs, the dashboards and the interfaces that are out there are in themselves quite daunting. Most of our people are there to run a community. They don’t wanna know that they’ve gotta go 45 tabs down in a cPanel to turn on this setting to fix this problem. They want very neat, very clean dashboards. Respectfully, they want the GoDaddy dashboard, which are clean one-click WordPress, give me a WordPress site, please. They want that dashboard. But with 64 gig CPU or in our place, or same as WP Cloud, dynamically scaling. They want all that magic to just happen and they want that very simple dashboard. Of course, their job is to run their community or their e-learning experience or their WooCommerce site. Their job is not to be a SysOp or a DevOp.
Jesse Friedman: Yeah. It’s so funny because you’re touching on another recurring theme in here. Now, we’ve had this conversation previously from a shared hosting perspective that you’re a small business owner. You wanna get back to knitting sweaters or teaching yoga or whatever it is that you’re trying to do. And being a web designer is not your core passion. So that’s not where you wanna spend your time. This is even at a deeper level here, you’re talking about running an entire social media community. Obviously being a SysAdmin is not necessarily something that’s at the top of the list. So even as they expand outside of that shared hosting world and into a much higher enterprise-level hosting plan, they still don’t want to be running the minutia of all this stuff. They want it all working. And you mentioned WP Cloud, one of the things that we’re very proud of is that we have auto-vertical scaling to compensate for these things, right? And we have something called bursting or PHP worker and CPU bursting to kind of compensate in real time for those things. It allows for us to create features for end customers so that they can feel very good about sleeping at night and not worrying about it. We’re gonna take care of it. We’re gonna manage it for you in the backend, and you just can feel good about knowing that your site’s gonna be up and running.
Wes Tatters: They want that bursting, they want that flexibility. Because what we discover is that these sites are incredibly heavily shaped.
Jesse Friedman: Yeah.
Wes Tatters: We’ve got sites that have got no one on them for 18 hours a day, but then for two hours in the morning between 7 AM and 9 AM there can be a hundred people a minute.
Jesse Friedman: This is the time they get up to date on things before they get into work. Yeah.
Wes Tatters: And their window, and everyone’s in the site during that time, and they chat, discuss, and boom, boom, they’re gone. And they’re off to their day. We’ve got a vegan lifestyle community out of Dublin, Ireland, called the Happy Pear, which grew out of a BuddyBoss site. And it’s now in the high street. They’re in shops, they got restaurants, they got TV shows that they’re on, but the guys started on a BuddyBoss site that was basically a recipe booklet of how to cook vegan food. It’s now an e-learning experience, how to learn, yoga courses, there’s people, but people log on every morning to download their recipe recommendations for the day. And then they’re not there and they’re not there again until they get home at night. And they might do a yoga course or take an e-learning experience. So the sites do this. And the traditional hosting model, you basically have to buy the biggest you need for your peak in the VPS model and in the dedicated model. So yeah. Oh, well, you peak at, you need 16 cores and 32 gig RAM, or that’s your peak, and so you pay for that.
Jesse Friedman: And you’re paying for that, even if it’s dormant for 80% of the time.
Wes Tatters: 90% of the time. Got a site in Europe, which is an online dating community. They pay real money to be in that dating community. It’s got a maximum number of customers. There’s only this many members in the community any one time. And it’s that 9 o’clock to midnight, but again, they’re in that same boat. They would have to, in a traditional hosting model, buy that bigger server they needed for that window in time. And they don’t have the brain space themselves to build clusters or auto-scaling clusters. And it’s a full-time job for a SysOp. They just wanna run their community. So yes, these tools that can take that load, but they want it even beyond that. They want people that go, your site’s down, we’ll fix it for you.
Jesse Friedman: Yeah. Right, right.
Wes Tatters: It’s a 500 error. All right? They’ve had a malware attack. We’ll fix it for you. They just wanna wake up and go, oh, you caught that. Thanks. And here’s the certificate that the malware’s being addressed. One of the unfortunate natures of WordPress right now is the zero-day attacks are increasing, because AI is now really good at rewriting those malware attack vectors so that they don’t look like the signatures that are in the traditional malware signature database. So you need RASP tools, and you need AI defensive tools that are actually monitoring your server and going, I don’t think that particular call should be writing that sort of obfuscated code to a temporary folder with root access. As opposed to, oh, we recognize that signature. It’s bad. Of course that signature is being changed by script kiddies with AI tools on a regular basis. And even more scary, I was talking to one of the plugin review team last week. They’re getting new plugins, plugin uptick at the moment, and they’re seeing AI-generated plugins being submitted to the repository. I went, how do you know they’re AI ones? And he went, oh, I got comments. Really good commenting.
Jesse Friedman: They’re actually taking best practices seriously. So it must be AI.
Wes Tatters: Yeah, but it must be AI. The problem is there are vulnerabilities creeping into that AI code. Because how does the AI code generate its knowledge base? Well, it read and sucked in the data of every PHP plugin it could ever find. And that means that some of those vulnerabilities were sure as little green eggs in those plugins and it’s now a pattern. And you got AI plugins with the vulnerability baked in, not necessarily the best look. And as you mentioned, Oliver Sack, and I know the WP Scan guys and the WordPress security team are just seeing upticks in security vulnerabilities that are doing this. So our customers in this space, they can’t afford, so let’s say they’ve got 5,000 customers and they’ve all paid them $10 a month to be a member. They’ve got a commitment obligation to those 5,000 customers to be up. It’s not like a WooCommerce shop where, oh, the site was offline. Oh, well, we might have missed a couple of sales. Or someone’s just bought a LearnDash, a thousand dollars course on a LearnDash product, and they’re ready to do it at 8 AM. There isn’t an option that says, oh, the server’s down, or there’s been an attack. So our customers are expecting us, we all talk about 99.999% uptime and all that sort of thing. But in our end of the space, that’s gotta be real as opposed to the traditional oh, well, yeah. Uptime is statistically, there might have been one person visiting your site during that one minute window when we had to reboot your server. In our space, customers are expecting a lot more, which means white glove everything.
Jesse Friedman: Yeah, that’s a really interesting concept because if it is an e-commerce customer, you’re probably thinking to yourself, oh, I’m really frustrated that this site was down. I wanted to buy this product, but whatever. I’ll go on with my day. I’ll find it somewhere else. They’re not angry at that situation maybe as much. But if you’re a paying customer and you have this expectation of things being up, now you’re creating a different set of expectations and you have a very real situation in your hands when a lot of them can’t get on. And I would imagine a lot of them do what a lot of us did yesterday. This is gonna actually really kind of tell us where we are in terms of how fast these podcast episodes are coming out. But yesterday there was a big outage with Google Cloud. And so a lot of times, you end up seeing a congregation of people joining another network somewhere or going on Slack or a community or whatever and saying, hey, are you experiencing outages? We are too, da da da. And so all of a sudden it becomes a very real brand issue, because you have all these people talking about how poor the situation is or how
Wes Tatters: When the outage reaches CNN and MSNBC at the top of the hour, there’s a big problem. But for a small community with a thousand people who all know each other, it’s just as big for them. I’m paying good money for this, and respectfully, they are.
Jesse Friedman: That’s their hard-earned money. It’s not for us to say that $10 is any different for them per month than someone who’s paying thousands of dollars. They’re investing not only the money into it, but the loyalty, the cognitive overhead, everything that goes into becoming a user of this. And we all know this too. The experiences that we feel when a community or something that we’re using every day is suddenly not there, it’s that feeling of knowing that you have something that just works is always something we end up taking for granted. And then when it’s suddenly not there, it’s anger inducing.
Wes Tatters: And people are very quick to take to…
Jesse Friedman: Other social media platforms, which is exactly what these people don’t want.
Wes Tatters: X or Reddit and an angry post or the standard tool of the trade. I’m writing a Trustpilot review or a G2 review. These are real issues at this end of the WordPress space. And part of the reason frustratingly why there is that drift away in perception by people at this level who then get sold on, oh, well, a SaaS model is much more stable or secure. And I’m not necessarily sure that that’s the case. But they have better language around that than a historic set of language in the WordPress space, which says WordPress is slow, WordPress doesn’t perform well at load and all those things.
Jesse Friedman: Well, of course they’re gonna pivot on top of any kind of thing that’s an opportunity to kind of say, hey, they’re not doing as good of a job. And here’s the magic of our ability to segue here. We’re talking about defending sites from downtime and other things. Let’s segue into you defending leakage from the WordPress community. So one of the things that you’re focused on is helping to identify sites that are basically being targeted by these SaaS offerings or maybe are considering them. And they’re running into these issues where there’s language or landing pages or marketing material or white papers saying, hey, we’re gonna do so much better than WordPress. And so you’re coming in and you’re offering a solution that’s not only beneficial to RapidCloud, as in you’re getting a new customer and helping them to stay inside of your platform, but you’re also helping them to stay within the WordPress ecosystem. And I think there’s something really noble about that as well as it’s in alignment with your business. So we’re talking about you being aligned with customer success. Now we’re talking about you being aligned with WordPress’s success. How does that story go? How do you actually go into a conversation there and what are the topics of conversation that you have to have? What do you have to defend in terms of WordPress? I know that’s a lot of questions, but let me just ask one more thing on top of all that. In the old days when I was selling WordPress, it was very much like having conversations about open source technology as a whole. Who’s updating it, who’s in charge of it? Who do you call when something goes wrong or who’s broken? I feel like I’m not having those conversations as much anymore. So I’m curious what your take is on all that.
Wes Tatters: The language around WordPress. And there was a conversation by one of the XWP team at WordCamp US last year, I think, around this exact problem. At Enterprise, even when they go into a meeting, they have to do two pitches where the SaaS guys are doing one. Their first pitch is around WordPress. First pitch is around the history of WordPress, its legacy, and all its upsides and positives and benefits. The benefit of open source, the benefits of a cohesive but equally diverse platform of plugins and themes and systems. And most importantly, about the privacy aspects and the data ownership and the IP retention and the control. You head into a SaaS model, look, none of the SaaS guys are saying they’re cheaper than us. They’re quoting prices often higher.
Jesse Friedman: Sure.
Wes Tatters: But they’re arguing. And what they’re not explaining in a lot of cases is the loss of control and the loss of data integrity or data ownership. WordPress is infinitely modifiable. WordPress is infinitely customizable. There is a reason why BuddyBoss deployed what is effectively Facebook on a blogging platform because the underlying WordPress architecture, the underlying structure of WordPress itself allows that and permits it. Beyond that, BuddyBoss can release a theme or Tutor LMS or LifterLMS or Elementor or LearnDash can release a product and a guy down the road can go, that’s a great product, but I’ve got an idea. Create something new and be a LearnDash plugin developer releasing a product into that space that solves another problem. That’s not possible.
Jesse Friedman: It’s a physical example of this idea. I remember Jason Sylvie’s, like a futurist, he talks about all this stuff, but he talks about this idea that ideas can procreate and that they can, you share an idea with me, you have a slightly different take on it. But WordPress is a physical manifestation of that because you can actually see how people are writing code. You can fork it, you can change it, you can adapt to it. And I think one of the things that you’re touching on there too is that when you make the decision to move from WordPress to a closed source SaaS based system, you’re creating a single point of failure. Whereas with WordPress, you’re diversifying and you’re creating ways in which you can pivot on many different areas. It’s not just the hosting because you can take your entire platform and you can move it from one host to another, but you’re also diversifying the code that’s behind it, the technological teams that are behind it, all these things so that you’re not relying on this one place. And then if God forbid, something does happen with that SaaS product, I don’t know what that even looks like, getting back out of it again. It’s incredibly expensive, I would imagine.
Wes Tatters: Well, the data’s locked.
Jesse Friedman: Yeah.
Wes Tatters: I’ll give you an interesting example. Fluent Communities has been released into the WordPress space recently by Jewel and the team there. Good product. As I said, I’m related to BuddyBoss, but I’m still WordPress agnostic. Jewel and his team have worked on an importer. If a person is on BuddyBoss and decides they don’t like it, you could literally import your entire BuddyBoss community into their platform. Now that’s just a part of WordPress. I can tell you right now that if you went to Circle or Mighty Networks, there’s no migrator available to migrate between Circle and Mighty Networks. But the nature of WordPress says that you can. We have two new WooCommerce products coming into the space at the moment, SureCart and Fluent Cart. And SureCart obviously is already there as well. All those actually have exchangeable data. It’s a massage, but you could, at any given point in time, massage your data, switch plugins, or we can change our themes, we can change our looks. In the SaaS platform world, those are not options for you. Or if there is an option, oh, well there is a developer module, that’s an extra hundred thousand dollars and you need to onboard a team of DevOps or SysOps and a systems engineer. Or whatever the obligation requirements are. In the WordPress space, literally go to ChatGPT whether we like it or not and go, hey ChatGPT, could you help me build a plugin that does this?
Jesse Friedman: Yeah. I would imagine though you’re having very similar conversations to the ones that we had in the first episode around the fact that you are owning your influence, you’re making a decision to build a social platform on something like BuddyBoss because you want to have control over it. You’re not gonna be beholden to some algorithm that you have no control over. The same conversation kinda happens here. The SaaS company, they may be a very well-built company. They might have everything, but who knows if they’re gonna be around in a year or two or three or five. And so now you’ve built your entire business on top of this platform that you’re then creating this single point of failure again, and you’re not even paying attention to the fact that you literally chose not to build this on Facebook or Instagram or whatever because of those same exact reasons.
Wes Tatters: And they often have a sting in the tail. Shopify’s a good example of that. As you get successful on Shopify, Shopify starts to look like an expensive business proposition from a total cost of ownership model when you’re looking at unit metrics. What seems to start happening is it gets more expensive, not less expensive. And they’re taking X percent doesn’t feel bad when you’re only selling a couple of products. But when you’re, we just sold a million products and we gave how much to Shopify. If we’d sold that million products on WordPress, we would’ve retained a bucket load of…
Wes Tatters: And they often have a sting in the tail. Shopify’s a good example of that. As you get successful on Shopify, Shopify starts to look like an expensive business proposition from a total cost of ownership model when you’re looking at unit metrics. What seems to start happening is it gets more expensive, not less expensive. And they’re taking X percent doesn’t feel bad when you’re only selling a couple of products. But when you’re, we just sold a million products and we gave how much to Shopify. If we’d sold that million products on WordPress, we would’ve retained a bucket load of
Jesse Friedman: Right.
Wes Tatters: So there are those conversations to be had.
Jesse Friedman:
Wes Tatters: But they have to be tempered right now by hosting challenges and scalability challenges in the WordPress space. There are some structural problems in WordPress, and they’re part of WordPress’s own success. You know, the database, the nature of the WP Post, WP Meta, post meta tables, allow WordPress to do what it does, but they’re also its worst enemy when you start to look at scalability and especially the databases. We don’t handle a hundred thousand users very well in our platform because our platform’s not optimized to handle those sort of things. You know, we’ve got little dropdowns that are trying to load a hundred thousand users into a select the user you want dropdown, it’s like. Yeah, we need to rethink some of that stuff. We’ve got databases, you know, the meta value columns, that aren’t indexed in any way, shape or form, but, Advanced Custom Fields users who are going, Hey, I’ve added a new custom field, which defines the hair color of every one of my users. And now I want to filter on it.
Jesse Friedman: Right.
Wes Tatters: And I’ve only got three redheads in the group and a hundred thousand users, and I’m now doing a hundred thousand record table walks every time someone says, and who’s the redhead? Because we don’t have indexing built into WordPress on those columns. And that’s the legacy thing technically. And these are the sort of things we do with our customers. We actually optimize those index and can add indexing to value columns because of, you know, subfield indexes and things like that. But it’s not in WordPress itself.
Jesse Friedman: How do we take the learnings and the things that you know, you are working on that companies like yourself are doing to scale WordPress and how do we give that back to the WordPress core team to improve?
Wes Tatters: For me, I’m actively involved in make WordPress.
Jesse Friedman: That’s great.
Wes Tatters: I’m a member of the hosting team.
Jesse Friedman:
Wes Tatters: The creation of the hosting performance communication channel between the hosting and performance teams. We’re working actively in that hosting group. So how do we talk? How do we do that? We get I love that. There are opportunities for everyone in this space and forget, fight for the future and obligations to contribute. There are many more benefits to contributing than thoughts
Jesse Friedman: Than purely just having your name listed in that group, for sure. Yeah.
Wes Tatters: We can be there active. You know, I chaired the fight for the future round table at WordCamp Europe. Because it needed to be done.
Jesse Friedman: Yeah.
Wes Tatters: And they required a contribution. Actively, we’re actively involved in the hosting community channels. We’re part of the team championing the new test runner project, that, you know, has been running for a long time, but we’re repositioning it. We’re also looking at extending that test run project to an actual directory inside WordPress that can show metrics about hosts in a very orthodox way, that helps people have some sort of an understanding of what hosts are offering in these spaces, the ones that are getting involved
Jesse Friedman: Yeah.
Wes Tatters: So they’re the options, you know, to actually get involved in the community.
Jesse Friedman: I think that’s really great because you are identifying some critical aspects that need to be maybe improved within WordPress. And it’s very much like, when we think about ourselves as members of the WordPress community, it’s very much like you have a sibling or you know, a younger brother or sister. You can pick on them, you can tell them what needs to get improved. But it’s like when someone outside, like some kid down the street starts picking on your brother. Yeah. You stand in and defend them. And it’s very much the same way, right? Like you are coming from this from a perspective of someone who’s actually dedicating the time to improve WordPress and to give back. And not doing it purely just because you want to have, you know, a contribution listed somewhere. But because you actually identify with the fact that like you and your contributions can make WordPress better. Helps you to have better customer experiences, helps you to make more money, helps you. It is a holistic view and I think that’s something that every hosting company can benefit from looking into that and thinking about how they can give back.
Wes Tatters: 100%.
Jesse Friedman: Yeah. Yeah. That’s great. Well, I think that’s an excellent place to take a, you know, to pause here, another awesome episode. It’s so funny how fast.
Wes Tatters:
Jesse Friedman: 30 minutes goes by Wes, but it was a real pleasure talking to you. I would love to have you back on the show at some point. Maybe we can talk about some more things. You know, we didn’t get to talk about like ways in which people at home can actually start to, you know, identify where and what’s going wrong with their sites. So I think it would be really great to bring you back on and say maybe we can have a conversation and talk specifically maybe about like some specific sites that are struggling and how to identify
Wes Tatters:
Jesse Friedman: You know, what it is that they need as an improvement or something like that.
Wes Tatters: Exactly. I’m at WordCamp Brisbane this weekend doing a 45 minute presentation on that exact topic we’re
Jesse Friedman: Oh, that’s fantastic. All right.
Wes Tatters: And we are literally discussing not the hosting perspective, but how customers with websites can spend the time and identify areas that they themselves can address or at least talk to their hosts about things like object caching and op caching and all those sort of things, for many customers aren’t even on their radar. So yeah,
Jesse Friedman: That’d be great. Yeah. So I’ll look for your presentation. I think almost all WordCamp talks now end up on WordPress TV. So Yeah, so we can watch for that at home. And in the meantime, if anybody wants to get in touch with you or reach out, where’s the best place to do that?
Wes Tatters: Wes Rapid at Makers Mate or LinkedIn? Wes TA is at LinkedIn, but, we rapid at, you know, make.webpress.org is where people reach out to me on a regular basis.
Jesse Friedman: All right, well, thanks for,
Wes Tatters: Or host, join us in the hosting team.
Jesse Friedman: Yeah, that’s some great advice there. If you are a hosting company and you’re listening to this show, take some time to go onto the make WordPress slack and join the conversation.
Wes Tatters: Hundred
Jesse Friedman: Awesome. Thanks so much Wes. Thanks for joining us on another episode of Impressive Hosting, where we uncover the core tenets of great WordPress hosting. Do you have a follow-up question for today’s guest thought or comment on anything? We talked about a future guest suggestion, a hosting horror story. What do you think makes great WordPress hosting all your comments? Shape the show. Drop them on impressive dot host. We also appreciate you following us on social media and subscribing to the podcast on your favorite platform. Finally, do check out our list of open source projects that need support at impressive dot host. Whether it’s code, community, or cash, you can make a difference. See you next time.





Leave a Reply