WPEngine review – after 1 month and 250k visitors, why exactly am I spending $249pm for WordPress hosting?

As you may know if you’re a regular reader, about a month ago we decided to switch our hosting from a dedicated server managed by yours truly to the WordPress hosting service WP Engine. Initially, I had expected to discover that it really wasn’t worth the money after the two-week free trial – but that time has well and truly passed, and I’m still with WPEngine. So, here’s a review – given they’re far from cheap, are they, in fact, the best WordPress hosting service ever?

This one’s off-topic for the Melting Pot, but as a lot of us are bloggers here, I figured it would be useful nonetheless. And I imagine you’ll be interested to know why I’ve voluntarily gone from paying $50pm for hosting to paying $249pm for what, at first glance, is the same service.

Quick note: the links to WPEngine here are affiliate links, meaning if you end up using WPEngine after clicking on one, I’ll get a small amount of money from that. However, I’m reviewing WPEngine because I’m impressed with their service, not for the monetary concern – their service has startled me by being well worth the fee, and I figured other bloggers would be interested to hear about it!

First Impressions

I first heard of WP Engine through a post on Kalsumeus, explaining why he made exactly the same move I subsequently would – from self-controlled dedicated server hosting to WP Engine’s managed service. I loved the idea right from the off – a host dedicated to nothing but WordPress, who could solve all the finicky little things that I couldn’t figure out (and with WordPress, there are a lot of finicky things that crop up over the months), and who knew exactly how to tune a site for speed and performance.

However, a quick look at WPEngine’s site made me think we couldn’t afford them. Reading through their prices, the Melting Pot would cost a minimum of $249 a month – and when we had a burst month and hit more than 400,000 visitors, as we do from time to time, it seemed we’d be moved up to the “if you have to ask you can’t afford it” top tier of hosting.

I said as much in the comments, because I can be kind of a dick sometimes. :)

A few days later, I wandered back to the same article, and lo and behold, there was a reply to my comment – from the CEO of WPEngine, Jason Cohen. And he made a hell of a lot of sense – and made the key point that unless we stayed above 400,000 visitors/month consistently, they’d charge us at the lower price.

About now, I realised they had a free trial (which has since been replaced by the rather better offer of a 60-day money back offer).

So I figured “what the hell”, and signed up for two weeks – fully expecting that I’d conclude their services were nice, but not worth the money.

Why even move in the first place?

The Melting Pot had started to suffer from a number of problems that lower-traffic WordPress sites don’t encounter.

Notably, WordPress is, to use the technical term, a right bastard to configure for speed under high load. I’m highly technically adept for someone who isn’t a full-time sysadmin, and I’d consulted with my hosts, Bytemark (whom I still highly recommend for non-WP services), who are very good indeed, but still, we couldn’t get load times stable and fast enough for my liking. Research shows that the amount of time a viewer will stay on your site, and the chances of them bookmarking or subscribing to your site, is/are highly affected by page load speed – and at peak times, despite my best efforts, the Pot moved like a hamster towing a truck.

WPEngine promised the fastest WordPress load times known to man. I liked the sound of that.

At the same time, my WordPress installation was starting to get a bit… clunky. We didn’t have automated upgrades installed, because it was a headache to do so, and so an increasing number of my plugins, not to mention my core WP install, were starting to get decidedly out of date. Out of date WordPress servers get hacked. And it was getting to the point where I was worrying about that fact every day, but still wasn’t willing to undergo the pain of a full manual upgrade.

Little things were starting to break, too. I couldn’t drag widgets around for some unfathomable reason, despite spending a day trying to bugfix that problem. I couldn’t Quick Edit posts. Scheduling would occasionally… not.

And I was getting increasingly sick of getting up at 4am to restart servers when my phone beeped to tell me the Pot had burned down, fallen over, and sunk into the swamp.

All of this pain would, in theory, be solved by moving to a managed WordPress solution like WP Engine – whilst still giving me the control of having my own self-installed WordPress.

Hence, the trial.

Migrating to WPEngine and the first few days

Migrating from the Melting Pot to WPEngine was a joy.

No, wait, not a joy. The other thing.

My initial migration was something of a nightmare. Despite having some instructions on the WPEngine site, nothing worked first time. I couldn’t even log in to the bloody site. Then I couldn’t log into the database. Then it wouldn’t import. Then it did import – and the Melting Pot appeared entirely in 64-point text with no images. Then my theme, the usually-excellent Thesis, decided it needed to be upgraded – and then upgraded again, and again, and so on.

I very nearly gave up on WPEngine as a bad job in the first three days.

However, their support was uniformly excellent. Alexander, one of their technicians, responded patiently to email after email after email, and did an excellent job of calming me down as I became increasingly shirty and irritable. It’s safe to say that without his support, I wouldn’t have even made it to Day 3.

And now we hit one of the major reasons I’m still with WPEngine. Their support is bloody marvellous. I’ve pinged them almost every day since I joined up with one problem or another, and in all cases, it’s either been sorted out or is still in the process of being sorted out. Technically, they don’t provide support for plugins, but in practise I’ve found that even with plugin problems they’re willing to pitch in and help out, and usually know more than I do.

(This post is being written with Markdown rather than Textile after I asked a support question about my textile plugin, and Jason, the WPEngine CEO, came back with a super-informed answer. Yes, their CEO often answers support queries.)

Eventually, the Melting Pot was online. And somewhat after eventually, DNS had actually changed, and we were settled in.

Running a Blog On WP-Engine Day To Day

Once you’re up and running, most of the time running a site through WPEngine is similar to running it on a self-hosted blog – although with decidedly less of the irritating niggles that WordPress tends to develop with age.

I log in – first time, now, since moving to WPEngine solved the irritating redirect loop my self-hosted solution had somehow gotten itself into. I edit posts – in an interface which not only works perfectly but also runs noticably faster than it used to. I manage comments – again, seamlessly, with no sudden hangs, slowdowns, or wierd errors.

What became noticable after a couple of weeks, though, and what convinced me to stay with WordPress, was what I wasn’t doing. Since I’ve moved, I’ve never had to

  • Delete a sudden onrush of spam. WPEngine run some kind of ninja anti-exploit software. I don’t know what it is or how it works, but spam has dropped to close to zero.
  • SSH into my server and run “top” to try and figure out why the Melting Pot is running like a dog. It’s always smooth and fast – and I know that if it isn’t, I can just email WPEngine and suddenly that’s their problem, not mine. (They guarantee 99.5% uptime for the server. If they go beyond that, they refund 5% of my monthly hosting PER HOUR.)
  • Restart the server because it’s crashed. See above.
  • Panic at an attempted hack on the server. Again, it’s their problem, and I know they’ve forgotten more about security than I know. ( Various reports from around the ‘net plus the CEO’s blog verify they’re the real deal.)
  • Spend a day trying to figure out why something bizarre isn’t working in WordPress. Again, I just email support and it’s their problem. Right now they’re tracking down the bugs the Melting Pot seems to have with trackbacks – something that I’ve previously and unsuccessfully spent large chunks of a day on.

Speed is definitely faster – not MUCH faster, but Pingdom reports a modest increase. More importantly, since we switched over our site speed has been a lot less spikey – previously, the Melting Pot was often fast but sometimes very, very slow – and that would always be at peak times when a lot of people were trying to access our site. Since the WPEngine shift, our load times don’t change with the number of people hitting us, and I’ve noticed we’re regularly serving 3 figures of simultaneous users. I’m looking forward to seeing how the site handles a real traffic spike.

So why DID I stay with WPEngine as a WordPress host?

It’s not the speed, at the end of the day. I thought it would be that if anything, but I was wrong.

I’ve decided to commit to staying with WPEngine, despite the steep rise in hosting fees, because it turns out to be a cheap way to buy myself a load more hours in the day. Essentially, I’ve outsourced all the WordPress and server maintainance tasks that were previously interrupting me or taking my time up – stuff that was slowing me down, but that I couldn’t have afforded to hire someone to deal with.

In addition, I also buy myself time because everything works. I don’t have to use arcane workarounds to deal with the fact various bits of WordPress aren’t working properly, because, well, they all are. I don’t have to log in 5 times because there’s something screwy with my self-hosted DNS. I estimate that the shift to WPEngine has saved me 10-20 minutes a day of messing around – and that adds up fast.

And I don’t have to spend time learning about stuff that doesn’t directly benefit the Pot. I don’t need to spend hours reading about the difference between Hypercache and WP Total Cache, or whether I should use Memcached, Varnish, a reverse proxied Nginx server, or all of the above. All that stuff’s interesting, but it’s not the best use of my time.

Basically, I’ve decided to stay with WPEngine not because they provide very expensive hosting – although it’s excellent – but because they provide a very cheap, very good, very agreeable full-time dedicated sysadmin for my blog.

Which means I can concentrate on, you know, writing it.

You can find out more about WP Engine here.