10 Dollar WordPress Hosting Screenshot

A few days ago, I noticed a post made to the WordPress Hosting group on Facebook that said the following:

any shared hosting offer for 10$ at first year?

If you think this is a reasonable request, we need to have a good conversation about why this is a horrible idea.

Sure. It’s possible…

…if you don’t care at all about whatever you put on there. I could toss up a web server on a $25 Raspberry Pi right now, and charge you $10 a year. I could also do that to as many possible people that I can cram onto it, and just leave it running on a 15mbit upload cable connection while I go on vacation, not monitoring the server in the slightest.

If you’re only paying $10 a year, how many other people are doing the same just so that the web host can break even, let alone make a profit? When the server has an issue, who’s going to fix it? Nobody, because even my technology-deprived mother-in-law would cost more than $10 a year.

Let’s break down the cost.

For our hypothetical situation, first we need a server. Often times, this results in getting the hardware, getting it into a datacenter, etc. but for the purposes of this article, let’s just assume that a $10 DigitalOcean droplet is being used instead. At $10 a month, this puts the number of subscribers on a single server at 12 to break even.

Current number of subscribers to break even: 12.

Anyone who has ever had to fight with a server knows that you can’t just magically spin something up and have it work properly. There’s some tuning involved. I know quite a few people who can make a LAMP stack seem to appear out of thin air, but when running a web host, there’s quite a few unknown variables to account for. It’s hardly “set it and forget it”.

So in that case, you would need some sort of sysadmin to look things over, at least for a day or two (yes, many of you are chuckling right now while dreaming it was that easy). Let’s assume you’re an expert in everything hosting and development and you can do it yourself within a 40-hour work week. At the current minimum wage of $7.25/hour, that would amount to $290 that you’ve invested of your time. This adds another 29 subscribers to break even.

Current number of subscribers to break even: 41.

49 subscribers on a single DigitalOcean droplet is still entirely possible. With low-traffic sites, it could survive. But wait, if something happens on the server, what happens? Well, you now have to fix it. Let’s assume a part-time systems admin putting in 10 hours a week at minimum wage (if your sysadmin is working for minimum wage, be worried). That tosses another $72.50 per week into the mix. With 52 weeks per year, it adds another $3,770 in costs for the single server.

Hell, we’ll even assume that he can take care of all of your support tickets in that time-frame too.

Current number of subscribers to break even: 418.

Alright, now this is getting terrifying. What happens if that server suddenly explodes? You need a backup to fail over to. Duplicating the existing server, let’s add another $10 and another 2 hours of configuration to clone the server and set up automated backups to it once a day (once again, grossly underestimating here). That comes to another $134.50 for the labor and yearly server costs.

Current number of subscribers to break even: 432.

Ask any sysadmin if they can successfully handle a $10 DigitalOcean droplet with 432 shared hosting subscribers on it, and I can guarantee that they will look at you like you just grew a tail and started tossing around your own feces.

Going beyond breaking even.

Now, let’s assume that you’re going to try to live off of this venture and want to quit your full-time job flipping burgers at Mickey D’s. After all, you’re a world-class sysadmin!

The total cost of doing so, minus the week we already took away, would come to $15,790. That’s right; you guessed it. You would need to get another 1,579 users onto that $10 DigitalOcean VPS to handle that.

Number of subscribers to handle costs and pay yourself minimum wage: 1,997.

Yep, sysadmins across the world are now wondering who put the turd in their pants.

Hosting isn’t easy or cheap.

I’ve worked in the hosting industry and know it quite well. To put it as simply as possible, running a web host is ridiculously hard to do, and if you expect someone to do it for pennies a month, you’re either ill-informed or insane. Unfortunately, many of the budget hosts are in a race to the bottom in terms of low pricing, and anyone new to the market is forced to offer some sort of lower-priced hosting to the masses. Thanks to some of the awesome hosts out there (such as A2 Hosting, Pagely, and WPEngine), it seems that the trend might be changing, at least a bit.

Working for a major web host, I’ve seen a lot.  So much, in fact, that my entire perception on hosting has changed in the last 2 years that I have been at InMotion Hosting.  Many people simply see hosting a basic container for their content when behind the scenes, there’s a whole lot more going on.  With that, I feel that need that an honest, insider look on hosting as a whole is needed to clear up the fog.

Hosting isn’t just a few files on a server

In the time that I have been working at InMotion Hosting, I have spoken to tens of thousands of customers.  Of those customers, many seem to just simply think that a server is tossed up, and and operations begin.  In reality, there is a significant amount of work put in before any accounts hit the server, while they are there, and even some even after they are gone.

Prior to deployment

Before deployment of the server, it needs to be fully evaluated for performance and stability.  Without a full evaluation, a host could never know how reliable the server actually is.  Can the server support 10 users?  100?  1,000?  In this, every step of the puzzle is completely evaluated.  This includes both hardware and software and has an enormous amount of environmental variables to ensure that each and every customer has a positive experience.

On the live server

Imagine having a boss that tells you there’s someone you need to talk to in the office, but you have no clue who they are or what you need to talk to them about.  Then, you take 2 steps and someone else tells you there is yet another person you have to talk to.  By the time you get to the initial person, there are 5 others waiting.  Allow this to continue 24/7/365.  That’s what handling the uptime and stability of a web host is like, and that’s just one aspect of the situation.

Then comes security.  A host’s responsibility is also ensuring the safety of its users’ data.  Without it, any attacker could just waltz into the server and start destroying things.  If you’re not familiar with the aspects of security, there is an incredible amount of exploits that are found on a daily basis, many that directly affect any web host.  The host then has to keep an eagle eye on all possible security issues and ensure that they are not vulnerable.  This not only includes their hardware and software, but customer’s currently running software as well which cannot be as easily controlled.  Believe it or not, many customers leave their content wide open to security concerns and many are only saved because their host made changes on the server level to ensure that they are no longer vulnerable.

Lastly, you have the customer themselves.  This can range from an extremely polite, educated developer that just needs a simple change made or has a question about a server configuration, to the raging customer who doesn’t understand how to copy/paste and doesn’t understand why their email doesn’t work.  (Yes, that is actually a real-life example).  Regardless of the customer, support representatives are required to be the go-to for any questions a customer may have, or at least know where to direct you to for the answer.  They can take a severe beating by a customer and need to get up off the floor from that call, and pick up the next one as if they were having the best day of their life.

End of life

When a server is being upgraded, there are also several additional variables that exist to ensure everything is running smoothly.  This mostly consists of ensuring the happiness of all customers currently on that server that are being moved.  In this, every customer needs to be checked over thoroughly to ensure that they will not have any issues on the new server and includes things such as checking over every bit of software for compatibility with their site.

Your host cares about your website; maybe even more than you do

Many people think that their host couldn’t care less if their site is up or down.  That couldn’t be farther from the truth.

When someone’s website goes down, they call up their host and the host can sometimes either resolve it right away, or may take a couple of minutes to fix the issue.  The customer then curses the host and waits for their site to come back up.  Here’s what happens on the host’s side of things:

The most common issue would be shared servers with load spike issues.  Someone on the server did something to cause a large amount of resource usage, and the sysadmins are now tracking the issue to determine the appropriate action to take.  On a shared server, that means tracing activity through 100s or maybe even 1,000s of websites, locating the issue, and being able to determine the best course of action almost immediately.

While your sysadmins are calming down the server, support representatives have to be notified and readied to the amount of traffic they are about to receive.  Almost instantly, support calls, chats, and tickets will significantly jump and all need to be handled.  The absolute best case scenario is that you are able to tell the customer the issue, they are understanding, and they are quickly on to the customer with the next issue.  Sometimes, the result can be the customer calling in while yelling, threatening legal action, and not allowing you to get off the phone for a significant amount of time to try to help the next customer, and possibly canceling their account.

With all of the work involved with any amount of downtime, the host puts in a significant amount of man hours with just a single server and a couple of minutes of downtime.  Your host certainly does care about your website and sometimes, more than you do.

You use your tools by choice, we use your tools because you do

As you may know, there’s no shortage of tools to build a website.  You may spend time investigating some that interest you, and learn to master them, but hosts have to know what tools you use, before you do.

Software is released and changed every day and web hosts have to be able to know how it operates.  Nearly every piece of software in existence needs to be investigated to ensure stability and compatibility.  For example, if WordPress were to suddenly decide to switch over to Node.js, the hosting provider would need to know about that change far ahead of when any customers find out.  The host would then need to make sure that every server is properly equipped to handle the new changes.

This even goes as far back as Microsoft FrontPage.  Microsoft stopped supporting it many years ago, but hosts still have customers running it.  At that point, change after change has to be custom written to ensure that all customers running FrontPage are still able to publish their content and use FrontPage Extensions for as long as possible.

Conclusion

Hosts care a lot about what they do.  If they make one misstep, they are called out on it like they murdered your first-born.  Hosts are not just in the business of making money, but taking the responsibility of hosting the internet world on their shoulders.  Are there bad hosts out there?  Definitely.  Do the majority of web hosts care way more about your website and that you are happy than their bottom line?  Without a doubt.

Coming soon!

Soon, I will be taking an insider and unbiased look at some major hosting providers, and pitting them against each other.  We’ll see how they stack up.