Blog

Choosing a web host you can be with forever (spoiler: not possible)

Well, folks… the last few months have been one heck of a ride on the WordPress roller coaster! Now that we’re back to the loading dock and have that wonderful benefit of 20/20 retrospective vision, the biggest one of those loop-de-loops is still on our mind: moving all of our support client sites from WP Engine to Kinsta.

If you’re on our email list, you’re aware that a ton of changes have happened in the land of WordPress over the last 6 months (and if you’re not, you can sign up here!). WordPress core rolled out version 5.0 with the Gutenberg editor, all PHP versions 7.0 and below have reached the end of their life, BIPi is no longer supporting sites with iThemes Builder, we’ve moved to using a DNS management tool that’s separate from your domain registrar (thanks for being awesome, CloudFlare!)… and we are moving all of our clients over to Kinsta hosting.

This move has been a struggle, because while most of our support clients get hosting as a part of their support agreement, some still purchase hosting directly, which means that although we recommend moving hosts, it’s not covered under the support agreement and therefore results in an extra cost. Yesterday, after yet another disappointing interaction with a hosting company support tech about issues with a client site, we had to request permission to move their site to a new host, and the client asked:

“The only thing I’m not crazy about is yet another host switch… not because I don’t believe there is a need, but I’m concerned it could result in yet another switch later down the road because of the current pattern… Do you really have enough experience with this alternative to know it’s going to be a good switch this time?”

This wasn’t the first time a client has asked this question, and after responding, it seemed like it would be a good idea to put our answer out here for posterity. So, for anyone who’s frustrated with their host and wondering how to make the next hosting choice their last, we have bad news… it’s not possible.

How often will I have to move to a new host?

At Berry Interesting, our experience with hosting leads us to believe that we’ll need to be prepared to move hosts every 3-5 years (coincidentally, that’s about the lifespan of about half the sites we manage as well). There are outliers, but the more content and more traffic a site has, the more likely it is you’ll run into issues at some point.

Way back in 2006, we started out using GoDaddy, then moved to HostGator (2008), then to A2 hosting(2011), then to WP Engine(2014 – in fact, we wrote about this transition in mid-2013!), and now we’re moving to Kinsta. We do not take these changes lightly, because it results in 2+ hours of labor per site every time we have to move.

The biggest challenge of owning and maintaining websites is that the pace of technological development forces near-constant change as the status quo.

Isn’t there additional cost associated with moving to a new hosting company?

This is a large part of why we moved to including hosting in our support packages is so that we could manage cost:

  • cost of migration is less when you can do it in bulk;
  • cost of hosting itself is less when you do it in bulk;
  • cost of ongoing support is less because when everything is in the same place and the support or quality of hosting degrades, we can quickly move sites to a platform where the work that I and my devs do takes less time specifically because of the speed/availability/quality of the host server AND (the big kicker) the quality of the hosting support techs is such that it doesn’t take you days to get an answer.

Additionally – and it’s not a small thing – when you purchase a higher tier plan, the hosting company takes you more seriously, because you’re a bigger client. You’re essentially in a collective that has a stronger reputation than an individual.

The dollar amount isn’t the only reason to move hosting companies

That last piece is the biggest, most frustrating reason. It’s not just that it’s hard to justify to clients billing them for hours just for sitting on hold or wrestling with a bad support tech. It’s that having to deal with crummy support is a frustrating activity in and of itself. It’s the equivalent of sitting on hold with Comcast. You get transferred around umpteen times, your internet still doesn’t work, and every tech asks you to unplug and replug the modem when that is clearly not the issue. Meanwhile, you’re savvy enough to have not only already done the basic troubleshooting yourself, you find that you’re actually more equipped to solve the problem than most of the techs you’ve been talking to.

Avoiding that experience, and avoiding wasting the time that it takes (because it is a huge waste of time and money. Even when you’re getting paid to do it, you know that it’s doing nothing but costing someone money solely because of incompetence on the tech’s part and a lack of care on the part of the larger org b/c they hired that tech) is a big motivator. Those conversations with tech support are a supremely frustrating experience, because we’d all rather just have the basic tools in working order so that we can get down to the work we want to do and the work we can control – building and maintaining the site itself.

How can you know that the new host you choose won’t also be terrible?

To be completely frank – we do not have enough experience with our new choice to know for sure that it’ll be a good, lasting switch. There’s always a risk when moving to a new service provider, but we’ve done enough research to justify the switch.

The way we make our decisions about hosting is based on feedback from the WordPress communities we’re in, the developers we work with, and the trade publications we read. When everything went south with WP Engine last year, we didn’t immediately bail. It took us 6 months to get fed up enough with the degradation in quality (both of their servers and their techs) to look for another solution.

When Kinsta was mentioned by two different sources, we took a look, and we were impressed. We’ve moved around in their control panel and it’s intuitive (in a similar way to what WP Engine had before their redesign). We have been through a couple trouble ticket exchanges with their support staff, and even though their response time is longer, the conversations are waaaaay better – you can tell that you’re talking to someone who knows what they’re doing from the beginning. Additionally, they put a very high priority on uptime and security, something that all the previous hosts we’ve worked with seem to have de-prioritized as their companies grow. Kinsta’s servers have security protocols built in so that you don’t have to slow the site down by using heavy plugins to deal with performance and security issues.

We have considered several options, and the choice to go with Kinsta this time is a result of reaching a critical mass of justifications, which is no different from the way we’ve made the choice in year’s past.

Won’t I ever be able to just stick with one company for the whole life of my site?

Nope.

Ultimately, the thing about hosting is that if you have an active, growing site where you frequently post new content, get a lot of visitors, and regularly add new features, you’ll likely never find a company to be with for life. We really wanted that out of WP Engine, because our experience with them was so much better than every company we had used before them. Unfortunately, their company has grown in such a way that they are no longer worth it, which is exactly what happened for us with HostGator and A2.

The biggest challenge of owning and maintaining websites is that the pace of technological development forces near-constant change as the status quo. Like WordPress 5.0 and PHP 7.3, you’ll never outrun changes, and those changes can have a domino-like effect on everything. Agencies close, developers decide to run away to Arizona to brew beer instead of building websites (true story), companies get acquired and quickly shift priorities, a codebase is updated and that update makes other code obsolete so revenue streams disappear.

So, we can’t say that we won’t run into the same problem again. It’s almost guaranteed that we will. But if migrating a host takes 2 hours every 2-3 years, while staying with a host results in 2 or more extra hours and tons of frustration every time there’s an issue with a site (usually once every 6-8 weeks), it makes sense to be agile and open to migrating your site to greener pastures.

So, what’s the solution?

Whether you’re on a support plan with Berry Interesting or another company entirely (we love PressManaged, WebTegrity, and Chance Digital!), establishing an ongoing relationship with an agency or developer is absolutely key. The goal is to find people you can trust to be honest with you, who provide guidance and maintain their expertise so that you benefit from it without having to worry about wrapping your own head around making technical choices.


This post may contain affiliate links.

by | Last Updated: Jan 16, 2019 | shop talk, WordPress website advice