Just when the Web hosting industry seemed more stable and entrenched than ever, announcements last week by Google, Microsoft and Amazon gave an official stamp of approval to the technological shift that is about to send it all swirling down the drain, and the specialist WordPress hosts may be the first disappear.
Web hosting has always been the bastard child of the I.T. industry. In the early days, any idiot sitting in his bedroom with a computer and a dial-up Internet connection could copy and paste together a professional-looking website featuring photos of tall server racks, arranged in reassuring rows, in a mysteriously white room, possibly a nuclear bunker.
Even if he was the smallest link in the long chain of guys slicing, dicing and reselling server space, stretching all the way up to the guy who had actually rented the server from the datacenter, the art of being a Web host was to present yourself to the world as if you were IBM.
As soon as you got your tiny slice of space, you could divide that sucker up and sell the fragments on to as many customers as possible. That was always the key point: selling. Regardless of their actual capacity, every Web host would offer miraculous $10 plans which included unlimited resources and, if more customers turned up than expected, you simply crammed them all into the same space. If anyone complained you would kick them out: even vaguely knowledgeable customers were simply not worth the trouble.
Did it matter if you knew nothing about online security or customer service? It did not. All that mattered was that your website contained the magical phrases – such as “99.99999% uptime” – which would give customers the strong impression that you did. In reality, nobody knew anything and the standard support response when things did go wrong was to stop replying to their emails and wonder how long it would take them to cancel their subscription.
What if customers discovered you were a sham, a fraud?
What if … what if people got angry with you?
Well … so what!
Those bozos might not like you, but the imaginary people giving the fake testimonials on your website LOVED you and, with the Web booming, more customers would keep flowing in. Talk to anyone who ran a website in the first decade or so of the Web and you will hear the same nightmare tales of wandering from one lousy host to the next.
Things are somewhat better today, if only because the software most customers use (including WordPress) has matured, the software used to manage Web hosting is idiot-proof and the cost of the servers themselves has plummeted to insanely low levels. It is still, however, a filthy, grimy industry, where marketing remains the overriding concern and profits are built upon the fundamental ignorance of customers. It is almost impossible for the average consumer to research good providers without becoming entangled in supposedly unbiased articles pushing whichever host pays the highest affiliate fee. Some companies pay bloggers hundreds of dollars for a single referral, so, it is hardly surprising that there are so many enthusiastic articles about the same small handful of hosts.
As the industry has consolidated around big players and mainstream Web hosting has become commoditized, we have seen the emergence of specialized WordPress hosting as a significant value-add. These companies have been able to ring-fence their pricing and, by using non-standard resource metrics such as page-views per month rather than bandwidth, prevent customers from realising how much of a premium they are actually paying.
This works well because the host benefits from the customer’s warm feeling that a team of experts are watching over it 24/7, warding off hackers, drug dealers and terrorists. With the lingering perception that this stuff must be difficult, and the customer’s delight with all that WordPress can do these days, the specialist WordPress hosts get a lot more credit than they deserve.
The problem is that modern technologies are rapidly rendering their role unnecessary. Anyone can now go to Linode or Digital Ocean and fire up their own VPS (Virtual Private Server) within half a minute, for a fraction of the approximately $30 cost of the most basic, one-WordPress-installation package from a specialist WordPress host. The VPS user can then use Docker or a similar technology to instantly install all they need for any number of secure, optimized installations of WordPress, all running with more memory, storage and bandwidth. The dirty secret of the specialist WordPress hosting industry: this is exactly what most of them are doing themselves.
Of course, using a VPS and playing around with installation images requires a willingness to follow a few tutorials, which is precisely what most customers are paying to avoid. That, combined with simple inertia, will keep the specialist WordPress hosts profitable for a few years to come, but the writing is on the wall.
The VPS providers are already evolving rapidly, improving their interfaces and bringing us ever closer to the day when running a server will be as intuitive as using WordPress. Meanwhile, a second layer of service provider is emerging, providing a simple dashboard that you can use instead of the native Digital Ocean or Amazon AWS or Google Compute Engine dashboards, and which also provide secure, optimized installations of WordPress at the push of a button, and without having to read any damn tutorials.
What you should really watch, however, is how the giants such as Amazon, Google and Microsoft build out their infrastructure. They are ALL brutally aware that the difference between success and failure will lie in the interface they provide, and the extent to which they enable ordinary people, not just techies, to harness that power.
That means a frantic race to provide an increasingly rich “out-the-box” experience, with server applications being the key battleground. Just last Friday, Google announced Google Cloud Launcher, enabling push-button deployment of over 120 top Open Source applications, including WordPress, to the Google Cloud Platform. Just two days before that, Microsoft launched the Azure App Service, which does essentially the same thing from a slightly different angle. Meanwhile, throughout the week, Amazon announced the latest pushes in the remorseless advance of their AWS cloud platform, which has long integrated Bitnami’s version of WordPress.
As the finer details of deployment, optimization, backing up, restoration and monitoring are all improved, and as the stability and resilience of the WordPress experience provided on virtual servers that cost pennies a day improves, how will the specialist WordPress hosts manage to hold onto their customers?
This disruption will hit specialist WordPress hosting first, but will continue on to wipe out the entire Web hosting industry as we know it today. Within five years, today’s mainstream Web hosting customers will be more comfortable with rich, well-designed interfaces that allow them to do more themselves, rather than rely on the supposed “expert support” in the traditional hosting companies, and this will collapse the industry down to a handful of providers with excellent software teams and massive infrastructure. I would not be surprised to see, before the end of this decade, SquareSpace or, even, Automattic acquired by one of the cloud giants as they seek to bolster their position.
Looking slightly further ahead, within a decade from now we are likely to see cutting-edge hosting, with deeply integrated applications, provided as part of Amazon Prime or Google accounts. The cost of all website hosting will be free or insignificant, and all hosting will be of more of less the same speed and reliability. The future will look back at today’s industry, and the time we spend dealing with hosting issues and optimization, as charmingly antiquated.
Looking ahead two decades, we will all be replaced by robots that look like Matt Mullenweg and Sergey Brin, while drones modelled after the laughing head of Jeff Bezos hover menacingly overhead.
72 Responses
I recently came across another article that also pronounces web hosting is dead; it looks like you were quite right in his arguments when you wrote this piece here a few years ago. Check it out:
Very interesting, thanks Jean.
I am not sure that I followed all his swings in logic but I agree with his observations that the hosting market has been rapidly consolidating and that generally improving tech is reducing the ability to differentiate on performance/reliability alone.
His proposed remedy – that small hosts should find more niche-specific, high-touch, unscalable services that they can provide in addition to hosting – is, I think, the clearest sign that hosting itself is (still) dying.
The logical conclusion would be that, as the value evaporates out of the hosting layer, such companies should simply provide those services as an add-on to the fully-automated, commoditized hosting provided by Google, Amazon or Microsoft. Why confuse the issue by continuing to think of themselves as hosts?
I do not agree with his contention that such services should necessarily be unscalable. It might turn out to be unavoidable but it certainly should not be part of the initial plan.
My current thinking is that much of the need for niche functionality will be satisfied by a secure plugin architecture/marketplace within whatever CMS is provided by the giant in question (that CMS could well be WordPress).
So, the hosting itself will be fully-automated and commodity-priced or free, but any specialized functionality, such as booking, would be provided and supported by an individual plugin company. Perhaps that is the direction in which stagnant hosting companies should be starting to think.
Hi Jean, I actually think that article shows a bit of the opposite. Here are a few quotes from it:
“If you look for epic success stories, the core of that message is always about care, not technology.
…
Because you can’t scale caring.
…
But you know what will still be in demand?
Human empathy and patience.”
Maybe the ‘hosting’ aspect is becoming more invisible to the end-user, but it isn’t going away. AI isn’t coming to save the day. I’m aware of a few hosts built on AWS, Google, Microsoft, but the majority still are not.
And, as for the end users…. if they think ‘traditional’ hosting is complex (i.e.: cPanel and being their own security/server admin) then I sure wouldn’t point them towards AWS, Google, etc.
While I’ll agree is (hopefully!) going away is the cheap shared hosting that never should have existed in the first place. What seems to be replacing it is higher-touch services that are priced more adequately to sustain these businesses, and actually do more of what the customer needs.
I wouldn’t call that the death of hosting, I’d just call it maturity.
Sorry to butt in Steve, but the “premium web hosting” you provide at your company IS based on Google’s Cloud Platform.
You clearly did not realize that, but it is just one of the shifts in the hosting industry since I wrote my predictions 3 years ago: WP Engine moved to Google. That is precisely the sort of flattening that my article was about. My 5 and 10-year predictions are well on track.
Your role is reseller, leveraging your relationship with your design clients to sell them WP Engine’s service at a fat markup. Good luck with that, but you are not adding any value. Your customers would undoubtedly be better off receiving any “care” they need directly from WP Engine’s 24-hour support rather than waiting for the office hours of your one-man operation.
You can continuing mining your unfortunate clients for years to come, but the industry as a whole is changing, with regular folks more aware than ever that they do not need to be at the mercy of expensive middlemen.
Hi Donnacha. Interested to hear any more thoughts you may have now, over a year since your last comment. I’ve read the entire post and all the comments here.
The current managed hosting services industry in its current form still seems to be growing… When and how quickly do you think companies like EIG will really start to suffer? When will the “old guard” start losing revenues to the new world order?
Hi Matthew.
A year on from my last comment, and two years on from the original article, we are still on course for my five and ten year predictions. What I wrote is a now far less controversial, and probably even seems quite mundane, because the trends I identified now loom over other areas of life and have entered mainstream awareness.
By growth, I presume you mean the publicly visible kind: that more competitors are entering the managed hosting industry and marketing like crazy. I believe that mainly indicates that the barrier to entry, in terms of both skills and capital, is continuing to plummet.
Right now, anyone, in any almost any country, can pay around $50 per month, with no long-term commitment, for a server hefty enough to comfortably host over a thousand clients, manage that server by paying $10 to a service such as ServerPilot and bulk-manage the WordPress sites by using software such as InfiniteWP, giving them the full array of “advanced” features such as automated backups and weekly client reports. I am aware of several people, none of whom I would describe as being technically oriented, who are single-handedly running successful managed WordPress hosting businesses in this way. Their only real cost is marketing.
Other people are starting with no capital at all by rebranding services such as Pressable, at a cost of around $3 per WordPress site, they can even handle the client billing. When you realize that tiny teams like Pressable, with just a couple of technical guys, provide the entire technical, support and billing backend for hundreds of other “managed WordPress hosting” businesses, you begin to grasp just how deeply automation has gone, and how reliably it can scale. Again, for businesses based upon Pressable, their only real cost is marketing.
So, the technical cost is minimal, the additional costs per additional customer are insignificant, and scaling is easy, you no longer need to hire a load of “WordPress experts”. The whole game is customer acquisition, and that is why we are relentlessly bombarded with adverts and marketing for managed WordPress hosting. It does not necessarily mean that these new businesses are making a lot of money, it just means that the profit margins are huge and each customer is worth fighting for.
Meanwhile, the more traditional hosting industry continues to consolidate, driven by simple math that one host can absorb another host’s entire customer base while retaining just a few of their staff, mainly support. We know that, behind the scenes, the big hosts are in an arms race to deploy any form of automation that will lower their per customer cost, reduce the events that cause customers to leave, and provide increasingly sophisticated software so that they can flood into more niches. Godaddy’s acquisition of ManageWP is a fine example of using software to target the managed WordPress niche.
Most of the people who expressed anger at my predictions, both in the comments here and the four other WordPress news sites that reported on it, were well-meaning guys who provide hosting to a small number of clients, probably alongside various web design tasks, and they proudly do all the technical stuff “by hand”. They are all dimly aware that automation is happening, somewhere out there, and understand that it will affect jobs but, in their heart of hearts, no white collar worker really believes that his work, with all its peculiar intricacies, could possibly be performed by an algorithm.
They are wrong. Their clients will not disappear overnight, but that portion of their livelihood which comes from managing WordPress hosting by hand will represent increasingly bad value, more of a tax levied upon clients too loyal or lazy to investigate other options. That is not sustainable and being angry about that reality does nothing to change it.
Interesting, thanks. I meant growth in terms of actual revenues, based on a little preliminary research. Of course it’s difficult to make accurate macro estimates, since industry consolidation produces quite a large distortionary effect (organic versus inorganic growth).
Still, it looks like the industry is still “working”. EIG is not performing especially well, and even got caught up a while back, incidentally I think in April 2015 when you first published this article, in a case that involved a dispute regarding its organic growth figures.
I’m just wondering when we’re suddenly going to see all of these VC and IPO-backed web hosting companies (read: marketing companies) start to lose out to the bigger players.
It’ll be interesting to see whether the larger players will subsume (through acquisitions) the thin existing “service layer” that companies like WP Engine provide, or whether they will create their own, new platforms that actually steal market share rapidly and evaporate the current industry (like your super bowl reference: AWS going mainstream).
And if/when mainstream web hosting is a commoditized utility, and when the “service layer” is also Apple-like in its simplicity and accessibility, where will the opportunities lie? Will there be any niche opportunities left in the industry, or will all the “high niche margins” float to creative design and application development, etc. (i.e. the product itself, rather than the infrastructure).
GoDaddy, as you mention, seems to be doing well for now. The stock is at all-time highs, revenues are set to reach $2bn+.
Thanks for your insights Matthew, your phrase “service layer” is exactly right, wish I’d thought of that!
That’s the bit which, already largely automated, will continue to be eaten by software. The specialist brands are already vehicles to accumulate customers. In most cases, they will get bought out based upon the value of those customers, and the brand either disappears or lives on but with their backend swapped out for the purchaser’s (EIG-style).
In the most lucky case, a giant will buy them with the intention of throwing their entire content management offering behind that existing brand. I find it hard, however, to imagine Google, Amazon or Facebook buying, say, WP Engine for the brand, they almost always launch their own brands. None of the big names in CMS hosting even sound particularly good.
The one exception is WordPress itself, which we could consider to be a super-brand due to its foundational importance in the CMS niche, its numerical dominance and, perhaps most importantly, the undeniable coolness of that name. Then again, Automattic would be an incredibly difficult company to integrate into any other, and it is almost certain that there would be a user backlash against the buyer.
Politically, Facebook could not do it even if they wanted to, while I sense that Google would turn their nose up at both the technology and the organizational structure. Microsoft buying WordPress would cause quite a stir but it would be good for the ongoing rehabilitation of their reputation, and a good match for last year’s acquisition of LinkedIn. Amazon has sufficient credibility to avoid a serious backlash, in the sense that they are already the backend for much of the Internet, and have the ambition, expertise and scale for the purchase to make sense.
The commoditization of advanced, managed CMS/app hosting will be just another step in the continuous lowering of barriers, allowing ever-smaller groups, with fewer technical skills and lower budgets, to reach an expanding worldwide audience. The opportunity moves to content, trade, curation and services, all iterating and evolving under brutal competitive pressure because the gap between idea and execution will be narrower than ever.
Great response and analysis, thanks.
Hi,
Your “Hosting Hell” website idea seems interesting 🙂
Yes, every development I have observed in the ten months since I wrote this article is moving in the general prediction I predicted, I am confident that people who come across this article by the time we hit the 5 and 10 year marks will wonder why anything I said was so controversial.
In the same way that the perceived value of Uber is not about their current model, relying upon human drivers, but, rather, about the value that gets unleashed when autonomous vehicles become a practical reality, we can see that the big players are positioning themselves for a future when it will be possible to provide complex Web hosting, at massive scale, with a minimal staff.
Off the top of my head, and just as a quick response to your question, look at how one of the companies I mentioned in my article, Automattic, are continuing to use Jetpack to pull self-hosted WordPress sites deeper into their orbit. The post and page creation pages now prominently display an invitation to edit via WordPress.com instead.
Offering this service, and so many others, to tens of millions of self-hosted sites does nothing but cost them money today, but they know that being in a service relationship with all 100 million WordPress sites, not just the WordPress.com ones, will be a vital competitive advantage for them as the range of services that can be provided automatically explodes in the coming years.
The race has already started, all the big players know the destination, that so many people providing ah hoc, small-scale WordPress hosting today are in denial about it is unfortunate.
With regard to EIG, yes, they are not worried about any of this because their business plan is very simple and far more short-term. They tap into the vast over-supply of money currently coursing through the World’s financial systems and use it to buy up hosting companies, strip out costs and wring as much money as possible out of the existing customers.
Looking at the list of EIG purchases, I see quite a few names that were once highly respected within the hosting industry, such as A Small Orange. I presume they are now just a ghost, just another front-end for the same rotten EIG service and this is the unavoidable problem with tech companies whose value is built around the human touch they provide: enthusiasm wanes, key employees move on, owners get tired, companies get bought, loyal customers get fucked.
The hosting giants won’t have to provide an automated service that is better than the service provided by humans, it just has to be more consistent and improve over time rather than deteriorate.
Very interesting to read this article about 1 year after publication. Are you still standing with the predictions you made? I’d like to believe in a brighter future with less (or even no) dependence on a traditional web host. Unfortunately, I don’t see it happen in the next couple of years. Big players like EIG are having a grip on the industry by acquiring a whole portfolio of hosting brands, but intentionally not consolidating them. In this way they don’t have to worry about quality and customer service. As when a customer is fed up with the service of one brand, chances are big that he will unknowingly switch to one of their other brands. These umbrella companies seem still fearless of a technological shift at any time soon. I would love to hear your thoughts!
Who is this anonymous fellow “kingofnyct” who, six months after my article was published, turns up to post a rant in the WPTavern discussion and THREE increasingly deranged rants in the WPMayor discussion?
I mean, seriously, what extremes of loneliness, procrastination and mania would have to possess a man to cause him to waste hours of his life, his precious irreplaceable life, tapping the same misinformed drivel into multiple websites?
There are many angles that one could take to question a prediction looking ten years ahead, and many here made good points (six months ago), but it is hilarious that this moron doesn’t realise that most of the major WordPress hosts are actually built on the cloud he considers so unsuitable 😀
Also hilarious that he thinks Amazon is about to close down AWS, their most profitable and fastest-growing division with over $1 billion in annual profit.
All of that pales into insignificance, however, when set against the personal tragedy of this poor, bewildered man, driven to pontificate but unable to squeeze out rational thoughts without spasmodically adding completely irrelevant buzzwords such as IOT and big data.
Initially funny but, ultimately, a bleak warning of the perils of not having friends to hang out with on a Friday night.
Who is this guy who decides to respond to the late, crazy commenter on his blog, takes notice of him on other sites, and then takes six paragraphs trying to find a psychologically vulnerable human to pour contempt onto?
Too bad Jetpack subscriptions don’t auto-expire after a few months.
@Dan Knauss, get off your high horse. As author of the post, I am notified when comments are made. I am also notified when comments are made on any of the websites where this article spawned discussions. All told, I wrote well over two hundred responses.
I felt I had a responsibility to take the time to respond to all the comments made, regardless of site. I engaged, in a civil manner, with everyone, including you, who made their points in a civil manner. You surely noticed that.
I am, however, under no obligation to keep on the kid gloves when some asshat starts his comment with the line “This has got to be the dumbest article i have ever read” and continues in a similarly abusive manner. A psychologically vulnerable asshat is still an asshat and, yes, I have utter contempt for him or anyone else who trolls discussions.
I understand that, as a Midwesterner, you may have an exaggerated sense of politeness but I feel that more people should take a stand when these nutters hijack our discussions. They are not simply howling into a void, they are venting their insanity into otherwise valuable conversations and stealing the attention of real people.
Jetpack subscriptions probably should auto-expire after a few months, or comments should auto-close, I certainly would not object. As it stands, however, if I come across one of these keyboard vandals I will take the opportunity to remind them that actions have consequences, that the world is not their punching bag, that people will treat you as you treat them.
Have you considered that I might actually be doing him a favor? Have you considered that, by being absolutely clear that his behavior is deranged and unacceptable, I might be giving him a rare chance to think about what he is doing and find a more productive, less negative use of this time?
Google deeper. I’m not a midwesterner, I’ve lived all over the place. Originally from southeastern New York State. Do you habitually google people and make assumptions about them you can use to explain them to yourself (and them) in terms of their perceived defects? Does that seem like a good use of time and higher brain functions? You are not obligated to treat others as rudely as they treat you; in fact there is a preferred version of your ethical axiom that says treat others as you’d like to be treated and do not treat them as you would not like to be treated.
Dan, I did not Google you. I followed the link that YOU provided at the head of your comment. That is what those links are for, for the commenter optionally provide context about himself. YOU chose to link to your “About Me” page, a page that specifically YOU created to publicly provide information about yourself.
In that page, literally half of the information points are Midwestern, so, yes, I took a guess that your bizarrely combative crusade for politeness towards deranged anonymous trolls was a regional quirk.
All of that is, however, largely irrelevant to the actual situation:
1. An anonymous idiot posted some deranged drive-by comments on an old article.
2. As the bartender for this article, I called him out for unacceptable behaviour but left his comment in place for entertainment value.
3. You butted in to call me out for calling him out too harshly.
4. I argued that harshness was precisely what was called for.
At this point, let’s just agree to disagree, I’ll hang on to my preferred ethical axiom, you hang on to yours, and both of us can get back to what we were doing before this anonymous troll sprayed his nonsense in our direction.
Cloud computig has no controls over performance because ultimately they do not co trol the network which is 99.9999% what good hosting and mobile computing is about. So this hypothesis is seriously flawed. Lastly there are plenty of hosting company applications that can totally destroy or outcompete cloud providers such as OnApp. These large cloud providers cannot compete with companies that specialize in their business nor do they want to.. they just want to offer mediocre services to get more advertising eyeballs or user coverage as much as possible without a concern for quality or quality controls. Then what about IOT and big data? That cant be hosted in the cloud its not powerful enough by its very nature of its infrastructure
Silly article…and naive
This has got to be the dumbest article i have ever read. Sorry this is pipe dreaming. The big cloud players do not have the infrastructure or capabilities to support web hosting on a large scale. Cloud systems cannot operate mobile applications which are consuming the world. In fact mobile computing will make the large cloud players like Amazon or Google obsolete because the latency is too high. Saas is already showing its cracks and failing in the current mobile and edge computing market. If the cloud providers attempt to sell hosting it will not work because hosting companies own entire backbone networks and infrastructure designed to host. This goes up to the core of the internet itself.
This has got to be the dumbest article i have ever read. Sorry this is pipe dreaming. The big cloud players do not have the infrastructure or capabilities to support web hosting on a large scale. Cloud systems cannot operate mobile applications which are consuming the world. In fact mobile computing will make the large cloud players like Amazon or Google obsolete because the latency is too high. Saas is already showing its cracks and failing in the current mobile and edge computing market. If the cloud providers attempt to sell hosting it will not work because hosting companies own entire backbone networks and infrastructure designed to host. This goes up to the core of the internet itself. Cloud computig has no controls over performance because ultimately they do not co trol the network which is 99.9999% what good hosting and mobile computing is about. So this hypothesis is seriously flawed. Lastly there are plenty of hosting company applications that can totally destroy or outcompete cloud providers such as OnApp. These large cloud providers cannot compete with companies that specialize in their business nor do they want to.. they just want to offer mediocre services to get more advertising eyeballs or user coverage as much as possible without a concern for quality or quality controls. Then what about IOT and big data? That cant be hosted in the cloud its not powerful enough by its very nature of its infrastructure
Silly article…and naive
The world is going to the one web , one log in , one hosting , one world. After few years we will log in a website and no need to log in anywhere , only one host will be there where we can host – Google ,Microsoft ,Facebook will be dominate all of this . Other will be either their own site or will support them like wpmayor – Things are changing very quickly and moving towards one world .Nothing to do ..
Nice insightful article an interesting read. I disagree about the role of a web hosting company disappearing. We have our own managed dedicated server and the majority of our customers are just not interested in the details of running their own account they just want their email and website to work and to have someone at the end of the phone should they need it or if there are any issues. A lot of our clients stay with us because of the service we provide as well as ensuring their email and website is working, something you certainly wouldn’t get from a huge company like Google on their ivory tower.
James, thanks for taking the time to comment.
I agree that the vast majority of clients are not interested in even the slightest hint of complexity but, within the ten year time span I am thinking about, both technological capabilities and customer expectations are likely to shift significantly.
With regard to Google’s ivory tower, I recently became a Google Apps for Work subscriber and my impression is that, when you are paying, their attitude towards support is markedly different. I even have a phone number I can call, presumably connecting to some sort of actual human, but so far I have not had any reason to call because their automated systems are performing flawlessly.
I also use Google Business Apps and have on a few occasions used Enterprise Support from Google. They are amazing and very very helpful.
Thanks for taking the time to comment tziady, that is good to know.
I agree then in 10 years things can change a lot, especially where technology is concerned so I guess on time will tell.
I also use Google Apps for my mail not had to speak to their support as yet but good to know they are efficient.
I have tried the Google platform and can confirm it’s really fast. I write automated tests for web applications and with what I experienced, I reckon most companies will go for this. Running a test script on local box took about 1m45s, on Google’s LAMP VM I created, it took just 0m2.261s. That’s a massive cut in time. I have been using various hosting services since 2005 and can tell you, this is the beginning of the end for specialised hosting services. The real gap between these “giant cloud” service providers and normal hosting companies is the ability to use new “Big Data” technologies and remote connection(ssh) without having to supply new forms of identification as they already know you. For me, deploying a comparison engine coupled with Elasticsearch and Apache Hadoop + Spark doing the clever stuff behind the scene just got a lot easier. Welcome to the present, the future is bright!
Yup 🙂
At this stage, while the cloud giants are not actively pursuing the mainstream market, the specialised hosts will retain most of their customers because, for most people, support is more important than raw performance, but we are already seeing a wave of companies who provide WordPress support as a thin layer above cloud hosting, that trend will grow over the coming years.
The real disruption will happen, a few more years ahead, when the cloud giants feel that the time is right, and the technology is in place, for them to cut out the middleman and directly chase mainstream customers.
That ‘middleman’ is the services the giants are unable/unwilling to deliver. So, unless you think the above Terminator robots are going to be able to do it, I’m just not seeing how your thesis is likely, if even possible.
What were the details of that local box? Was it a Raspberry Pi? 😉 Just kidding. Yes, some of these places offer some pretty fast raw hosting performance, but it’s about the service IMO, not the performance for most people. As I mentioned in another thread, I moved from free hosting to fairly expensive hosting because of the service. My backend speed got slower and I only gained maybe a second on my front-end speed (i.e.: 3 sec -> 2 sec type stuff). It’s all about the services unless you’re an IT-masochist or a newbie (or business is Web-hosting).
Very good insight and breakdown of future approaches. Don’t completely agree with everything he says. Nice insight; but the reality of it; is that more and more clients are coming on board that do not want the hassle of setting up and dealing with these issues. The key is service for many of our clients that do not want to be bothered by the minutia of hosting a website.
Thanks for the comment Tamer. Support is certainly important for most people at this stage in the evolution of hosting technology.
And support isn’t going to be important in the future because….?
Exactly… and it is important even for folks like me who know a good deal about what we’re doing. A couple of years ago, I transitioned my clients from *free* hosting (where I helped manage the server) to some fairly expensive managed WP hosting, just because of the services they offer. While I might user Google, Amazon, etc. for some aspects of my business, they simply don’t provide the services which are valuable to me, and I don’t expect they ever will.
This interesting article reminded me somewhat of Chris Lema’s outlook for the Managed WordPress hosting industry, with service providers becoming more and more just specialty VPS image providers that can be used on a wide array of inexpensive VPS hardware.
From the little I understand, Kinsta is actually in a way advancing in this direction, since they can and do deploy their (really nice, BTW) WP stack on Linode and Vultr wherever one asks for, and possibly would and could do the same with many more good-quality VPS providers. They understandably are not quite ready to do this for as little money as ServerPilot, for example, but will or would be well-prepared when the high-margin WP hosting business begins to falter.
Possibly this could then be a step inbetween today’s highly specialized and thus pretty expensive WP hosting industry (GoDaddy being the exception – they probably need to get as much market share as possible before their private equity investors can cash out) and a possible fully commodized future of hosting run by the industry giants* for free, as envisioned by @donnacha.
* this is not something I necessarily look forward to, actually – I’m old school enough to quite frankly fear the Google/Amazon/Microsoft-run world.
Hi Phil, thanks for commenting.
Your prediction about providers such as Kinsta transitioning to running their stack on the customer’s choice of cloud is spot-on, that is almost certainly one of the trends we will see emerging.
My hunch is that we will see stack “recipes”, from a wide range of individals and companies, becoming available as one-click deployments on all the clouds, and competing for popularity in much the same way as apps do on today’s app stores. Just like apps, I suspect these recipes will be mostly free, with companies like Kinsta making money by integrating the sale of support, training and preferred premium themes and plugins into their recipes.
I’m actually not worried about Google, Amazon and Microsoft, because competition will force them to become interchangeable utilities, switching providers will be far easier than it is today.
Heh, I read this and then realized it was posted on April 1st… please tell me this was an April’s fool article.
If not, then LOL and this guy is probably one of the folks predicting the end of truck drivers because of automated cars, and fittingly (based on the choice of photos), is awaiting the robots who ‘decide’ to end humanity. And, he probably doesn’t even realize why such a thing is so silly. Oh well…
Hi Steve, thanks for taking the time to comment.
Just to confirm, yes, if we are talking about the same ten year timeline as my hosting prediction, I reckon there is a high probability that autonomous vehicles will impact the trucking business by 2025 – not all of it, not everywhere, but they will have their place.
As for robots ending humanity, I figure humanity can probably handle that job just fine on our own.
Agreed on your final point. 🙂
Various kinds of automation and assist technologies will certainly impact trucking, as well as the average daily commute for the rest of us. But, so long as there are human drivers on the road, autonomous vehicles will be a disaster waiting to happen. That said, it might be tried, because too many people have sci-fi visions of the future rather than understanding the real implications of ‘A’ in AI.
If we have anything to fear from robots, it will only be in what devious humans might program them to do… despite recent predictions from some otherwise really bright folks who should have taken a philosophy class or two along with all their tech.
I am not so sure that the giants will kill off the specialist, as Google, Amazon etc, could never offer the support that the niche hosting suppliers offer.
Many WordPress bloggers, do it for love, and have very little technical knowledge, and when their site is compromised, or something just does not work as they want it too, they look for personal support rather than spending hours on a forum to try and find help.
Well, April 1st and all that, but still an interesting thing to consider. Your proposition might hold water if one is working for a ridiculously low wage. The $30 you are tightly clutching will cover about 10 minutes of my time. So I need to figure how much the hosting service saves. Is it more than 10 minutes? Certainly is. As for Google Cloud Launcher, Google estimates the monthly charge will run about $36. So the hosting service saves me $6 and eliminates some mighty boring maintenance work. I’ll happily pay the $30.
I’m scared to ask what sort of services you provide for $180 an hour … but do I get a discount if you stay all night?
The important point is that there is no form of hosting that requires employees who are paid the $360,000 a year you earn (before overtime). All the technologies involved are relatively easy, the fall-back procedures are idiot-proof and increasingly intelligent automation means that many companies have replaced their inhouse server and network admins with a contract for remote management. I am not saying this is a good thing, I am not saying that it is a bad thing … just that it is a thing.
The cloud giants are already gearing up to provide smart hosting as a utility and many customers will embrace that, especially the ones who cannot afford to hire you. It means that hosting itself becomes a simpler, more predictable part of the equation, and that people with talent and skills will now be able apply that effort and money to the other parts of their projects.
Why do you think you should be charging less per hour than a plumber or a carpenter? Are your “professional” and “specialized” skills worth so little?
Using 2000 hours as a multiplier is appropriate for an employee, not an independent consultant. Also “overtime” is a concept only appropriate for employees. Billable hours for an independent consultant are much less and must cover many expenses not faced by an employee. No fooling.
Still, why should I pay $36 and take on the extra work when I can pay $30 to have somebody else do it for me? Looks like faulty math to me.
Very well said!
The Google comment is especially spot-on. Google doesn’t even know what customer service is. They are service provider who makes some great tools for people who can use them as is, or are technical enough to dig in.
“we will all be replaced by robots that look like Matt Mullenweg” Something is going wrong here, Donnacha, I don´t want a Matt replica replacing me. Can´t I choose a Groucho March clone instead?
Don’t worry Luis, the good news is that WordPress.com VIP customers will be replaced by robots that look like Beyonce.
This article is understandable in a WordPress or a small to medium sized website context, but does not apply at all to a large segment of the hosting market. By large segment I don’t mean in terms of “website” volume, but in terms of revenue. Large and enterprise level clients have very specific demands that require internal IT or rock solid 24/7 management from the host, and that will never go away regardless of the big players.
It should be named “Hopefully an end to crappy shared website hosting and marketing savvy niche hosts who don’t really do anything different…. is near”
Most of us in the WordPress world are unaware of what is happening in the Enterprise but a major theme over the past few years has been the phasing out of dedicated, inhouse admins for servers and networks in favor of remote management. That would have been unthinkable just a few years ago, but it is happening.
Some big companies will always have their own inhouse Web tech team, some will have a combination of inhouse and expensive outside agencies, but my hunch is that, slowly at first, parts of that market will break away, greatly accelerating towards the end of this decade.
The title … well, you know, readers like snappy titles … that is one industry truth I am unable to disrupt at this time 😀
Ya it’s all changing, many enterprises are moving to the cloud to reduce IT costs and there is a market surge of software across all industries gaining traction because they leverage the cloud and can adapt quicker, SaaS is basically swallowing up hosting part because no one wants to pay or deal with IT, In some markets this will take decades. In the WP world though a barrier of entry for different frameworks which are typically more difficult to install (laravel, node, etc) will certainly be a factor.
The title talks about the end of the webhosting industry as a whole, but the body talks about the end of premium wordpress hosts.
Ah … the notorious “SomeDude” 🙂
The article is, as the title suggests, about how this will happen and, obviously, I describe which segments of that extremely broad industry will be the first to slip – “… and the specialist WordPress hosts may be the first disappear …” – and I go on to explain why they will be particularly susceptible.
I then reiterate, at the end, that “.. this disruption will hit specialist WordPress hosting first, but will continue on to wipe out the entire Web hosting industry as we know it today …”.
It’s actually the opposite. Google, etc. might easily replicate simple hosting and probably do it more cheaply. What they can’t replicate, is the value-add and service the better managed WP hosting companies provide.
I don’t fundamentally disagree with you on this, but I don’t think it’s currently as easy as you describe or likely to play out quite the way you envision.
To deal with any kind of server there is substantial conceptual knowledge required to understand what you are even doing and “where your stuff is,” to say nothing of handling version control, backups, migrations, major OS upgrades, and all levels of dependency management. Paying others to worry about this and handle it is ultimately about delegating stress and saving time for what you want to be doing instead. ServerPilot is very far from being able to take on that role, and the mass market is very far from understanding the value or even the need. Try to get ServerPilot to say what they will do with customers who don’t self-upgrade when their OS reaches EOL, and how long they will support old versions of PHP.
It’s hard for me to envision one of the cloud giants getting all the major market segments covered in an interface or series of services that cater to , more or less, everybody. Premium specialty managed hosts like Pantheon and Flywheel with very specific niche markets make sense as customers of the cloud giants, or maybe as subsidiaries, but not as services to duplicate and compete with on price. At the end of the day their unique value is to know a particular type of customer really well and serve them really well. I don’t see that ever being replaced to more than a minimally adequate level by Skynet. Mass market services and products always tend to be minimally adequate at best because they are maximally faceless and impersonal. Most people may accept that, but not all, and no one will love it.
Thanks for the great comment Dan.
Obviously, none of us can guess exactly how it will all play out, and, yes, ServerPilot is a good example of a product that solves a complicated problem with an almost-as-complicated solution. My hunch is, however, that, over the coming decade, significant segments of the market will turn out to be less interested in the “personal touch” than the option to cut costs, and mass market players such as Google have already shown the patience, determination and technical talent to make serious inroads into areas as vital as paid email hosting.
Probably so. 😐 So what’s the upside?
I’m not sure how, exactly, the upside will manifest itself, but I am pretty sure there will be one. Anytime technology lowers barriers and enable more people to do more things, unexpected benefits emerge. Many people are still upset that the Web has more or less wiped out local newspapers, but I think what we have now is far better overall.
But even with that, local news papers still exist. Not always do things completely disappear but they do drastically change. This change has been evolving for a very long time. It is fun to watch.
The specialised knowledge required, is to know that Digital Ocean, Linode etc are the ones to go for.
Of course a *true* techie such as myself will use another free service such as serverpilot.io to set up a highly optimised server instead of the default 1-click install. Maybe I should write a tutorial…
Very good point, most normal folks would not know which provider to use … but that changes radically once one of the big guys decide they have got their service to the point at which it is ready for the mainstream and start to advertise it during the Super Bowl etc.
I have used ServerPilot and, although a lot slicker than CPanel, it is fundamentally the same thing, so, I don’t see the point of paying $120 a year per VPS for that as opposed to $25 per year for Webuzo. My hunch is that an increasing amount of that functionality will be integrated into the VPS dashboards anyway.
Your article is predicated on that belief that service is meaningless, and that customer service response time in the 24 hour or longer range, if at all, is perfectly acceptable for most people. Email only support is simply unacceptable for many as well.
As long as folks are using content management systems like WordPress, and others, there will be a need for expert and responsive support. The fastest web server on the planet with the most amount for free space is not going to help a business whose site is down due to a network or server glitch–and no one available to help in a responsive manner.
Not everyone is an expert…
No, my article is predicated on the belief that a major part of the industry’s value proposition is about to be swallowed by upstream providers.
There will always be a market for people with skills to sell services to people who don’t, but the dirty secret of the Web hosting industry is that MOST companies turn out to be not all that skilled when things actually do go seriously wrong. I am often asked to help rescue people who thought they were paying for “WordPress management” by “WordPress experts”, and paid top dollar for years only to discover, when they actually needed real expertise, it simply was not there. I have dealt with all the major hosts, have seen the mistakes they make and am all too aware that, with rare exceptions, they don’t particularly care, that isn’t really what their business is about.
In the same way that we are already seeing the phasing out of server admins as a distinct entity in businesses, I expect to see the emergence of a new category of expert who incorporate much-simplified server and network management into their skillset, as an enhancement to their primary role as Web Designer or App Developer or Ecommerce Advisor etc. Bear in mind that people naturally enhance and broaden their skills as their careers develop. The cloud giants are pushing hard to make it easier for users to route around the middleman, some will choose to do so.
OK, I’m a bit confused now. You seem to be talking about all the hosting providers that just provide space and bandwidth for cheap, but have no real expertise. As far as that goes I agree. But, I guess the part I’m missing is how the big players are going to do any different.
People don’t just need help installing WordPress. While a majority of people might not even attempt that unless it’s just a click of a button… after that step is completed, the fun has just begun. Who is going to help out after that? Certainly not those big players you speak of.
If your thesis is just that some big players are about to compete and replace the useless bottom end smaller server-space providers… then I agree, I guess, but that doesn’t require much predictive capability, just time to happen.
If you’re talking about them replacing companies like WP Engine or Pagely (i.e.: managed WP hosting) and/or Website design services, then I’d say absolutely not.
Much of the reason for less local server admins, is that people tend to use services and cloud storage instead of a local server today. But, I don’t think that analogy holds true for Website hosting.
The complexity level and kind of thing we’re talking about there is just way higher. It’s kind of like saying desktop publishing software killed media designers. No, it just killed off stagnant services who refused to adapt to that change. It did mean people with no knowledge could easily do a crumby job. (Just like a push-button WordPress install on some new Google service might make LAUNCHING a WordPress site simple.) But, design knowledge and fundamentals… and the fact that businesses with money hired the expertise didn’t really change. Website design and hosting are more in that category.
We tried siteground but we were not very happy. We do not like godaddy either but siteground response time was worse than godaddy for the EXACT SAME WEBSITE. You can see the screen shot at;
Yes, Web hosting is still a complicated and wildly variable industry. The silver lining in my article is that speed, reliability and cost will all improve as the cloud giants compete to provide the best end-to-end service. We really will look back with horror upon the mess that is today’s Web hosting industry.
Kind of irony to see a short hosting ad promoting SiteGround and WP Engine after reading your article 🙂
Lame joke aside, I have to agree with your view in this article.
I had exactly the same feeling when I heard Google keyman Urs Holzle that they are giving out $20 mil of free infrastructure credits to startups – “these big players are coming after hosting companies and would kill the industry soon”.
Get (really) big or get out – I guess. And I believe soon Google, Microsoft, and some other big names will be giving free, decent hosting accounts to everyone who wants it – just like free emails that we are enjoying right now. It’s sad to see big players monopolying almost every aspect in our online activities these days but like most things in our capitalism world – it’s inevitable.
I miss the good old times when we have more smaller players in every field.
Yes, Jerry, the irony is not lost on me 😀
Unfortunately, that is just one example of how dependent content websites are upon the advertising and affiliate schemes. To their credit, WPMayor are remarkably good about not allowing commercial considerations to affect editorial decisions. I was allowed to write on any subject and there was never any suggestion that I should alter my content to avoid embarrassing advertisers.
With regard to the perception that the big players are taking over, it is important to remember that, even today, all Web hosting companies build their service upon someone else’s infrastructure. For instance, WP Engine use Linode in the same way that anyone else can. My argument is that the industry, as we know it today, is all about adding value by bridging the gap between the abilities of their clients and the skills required to use the underlying infrastructure. Unfortunately for them, that gap is rapidly being closed by the infrastructure giants.
I guess I have to disagree on your last point. It was WAY easier 10 years ago to run a site than it is today if you want to rank in Google and keep it secure. While a manual install of WordPress takes a bit more learning than a control-panel-click, it’s AFTER that point where things get complex.You can’t just install WordPress and go anymore. And that is where all the added value comes in. Besides simple things how much data and how loaded the servers are, that aspect is what really separates one host from another.
Very informative. Thank you for your insights and view of the future.
Hi,
Thoroughly enjoyed reading your piece. Certainly interesting and in no unclear terms very fascinating. Marketing rules the roost right? In fact it has never been out of focus! Thanks a lot for your insightful article.
sundar
“Anyone can now go to Linode or Digital Ocean and fire up their own VPS (Virtual Private Server)” Totally not true. Not everyone, in fact, only a few have the specialized knowledge it takes to do that.
Honestly, the real barrier for most people is technical confidence, not competence. I have trained hundreds of people on how to spin up a VPS, point a domain and create a WordPress site from scratch – the ones who enjoy it most are the ones who know so little that they are not aware it is meant to be difficult.
Yes, they need to take notes and, yes, I need to explain a lot of basic concepts, but I have yet to come across anyone who cannot grasp it and, more importantly, who doesn’t enjoy the thrill of creation. They almost all turn to me, afterwards, and express surprise that it was so straightforward, so simple.
And then what happens?
Thanks to everyone who has commented so far, I appreciate the kind words and encouragement 🙂