I don’t believe that shared hosting is dead or will be decimated. In my 12 years of hosting industry experience I’ve heard that shared hosting was to die many times before, and yet, shared hosting is still a thriving business today. I guess shared hosting is a cat with nine lives. Let’s look at some of the past death-threats.
Free Hosting Will Kill Shared Hosting
In the early days of hosting, a new ad-driven free model for hosting arose. Many were claiming that free hosts like GeoCities were going to kill subscription-based shared hosting. Well, fast forward to 2009 and Yahoo just pulled the plug on GeoCities. R.I.P. GeoCities – shared hosting is still alive.
ISDN/DSL Will Kill Shared Hosting
Back in the day when most users were using modems and when ISDN/DSL was starting to grow in adoption, there were claims that fast internet connectivity would kill subscription-based web site hosting. It was argued that if a user could get fast connectivity to the internet, why wouldn’t they run their own web server at home or their business office? Well, in the early wild wild west days of hosting, that’s exactly what the shared hosting entrepreneurs did. They ran their business and web servers out of their garage/apartment. Fast internet connectivity didn’t kill shared hosting – it fueled the business.
Domain Registrars Will Kill Shared Hosting
After ICANN opened the domain business up for competition, many new domain registrars came on the scene and started to diversify into offering hosting as a value-add. Because of the domain registrars huge market share in domains, there were many predicting that they would be the death of shared hosting. Sure, domain registrars may have many shared hosting customers, but that did not kill the shared hosting business. In fact, many hosts partnered with domain registrars, some also became domain registrars themselves, and still other shared hosts provided private label hosting for the very domain registrars that were going to do them in.
Functional Hosting Will Kill Shared Hosting
With the rise of Web 2.0 social media, we’ve seen the rise of many new “functional” online services like Hotmail, MySpace, Blogger, Flickr, and YouTube that allow users to maintain email, personal portals, blogs, share pictures and videos – all for free. I heard proclamations that shared hosts were going to die because of these functional hosts. The way I see it, functional hosts pushed many new users who didn’t use shared hosting before into the world of having a web presence. No longer was the web the sole domain of web designers and developers who knew how to program or use WYSIWYG authoring applications. Now a complete newbie without any experience could start a blog and publish their own content live onto the web. Functional hosting fueled new entrants to get a web presence, which in turn fueled growth of the shared hosting business. The more sophisticated and experienced the user became with their free functional host, the more they wanted increased control and custom templates which is supported by more mature and easy-to-use CMS applications on shared hosting platforms.
Cheap Dedicated Hosting Will Kill Shared Hosting
Dedicated hosting used to be really expensive. But then the budget dedicated hosts came onto the scene with cheap dedicated boxes. With competition, prices dropped even further. Again, I heard calls for the death of shared hosting. Why would anyone want to share when they can get a cheap box for themselves? What cheap dedicated hosting actually did was lower the barrier for shared hosting. You actually saw a rise in the number of shared hosts as the high price of hardware was not a barrier to entry anymore. In fact, coupled with the availability of canned hosting automation and control panel software, the number of shared hosting companies exploded.
SaaS Will Kill Shared Hosting
A few years ago, I sat in the audience at the Tier1 Research’s Hosting Transformation Summit where Dan Golding, Tier 1 Vice President and Research Director, declared the death of shared hosting by the movement toward software-as-a-service, managed hosting, virtualization and other pressures on the shared hosting business. Fast forward three years and shared hosting is still here. Even today, Tier1 projections still show growth in the shared hosting sector. I see the SaaS growth as tapping into yet another new market just like functional hosting, which will probably help fuel the shared and other hosting industry segments. Some shared hosts have become SaaS providers themselves or added SaaS options to their hosting offers.
Now: The Cloud is Going to Kill Shared Hosting
The new shared hosting killer is the Cloud. To me, this type of talk is just more hype. Yes, the Cloud is important and will be useful to many, but not everyone. Even Tier1 Research’s Phil Shih, Research Analyst for Mass-Market Hosting, during his presentation on “Following Web Hosting Trends” at HostingCon 2009 stated that their research shows that the Cloud’s potential revenue projection is actually a lot less than what is being stated by the rest of the Cloud industry at large. You also have roll-up companies like Endurance International, still actively targeting and acquiring shared hosting companies. Clearly these investors do not believe that their business model is doomed.
In fact, as I stated before, I think the Cloud, with its promise of elasticity, resources on-demand and pay-as-you-go revenue model, will impact the dedicated hosting business. With a dedicated server, you have to be able to handle your web site’s peak activities, which means that most of the time you are paying for server resources that are not being used. With the Cloud, you would pay for what you use and nothing more. So in this sense, it would seem like the Cloud would have a disproportionate impact on the dedicated hosting business. So it does make perfect sense for RackSpace, a renowned dedicated hosting provider, to move into the Cloud computing space. To join the Cloud movement rather than try to fight it. And lately, RackSpace has been pushing the concept of “hybrid hosting” – customers using both dedicated and the Cloud. RackSpace will need to demonstrate case studies and success stories of this hybrid model and hope that this is the way the Cloud is adopted.
So if none of these “shared hosting killers” have been the death knell of shared hosting, what effect do they have?
A Healthy Ecosystem of Hosts
With the rise of new business models and new hosting technology innovations the only thing that changes is the number of choices for the end user – which is a good thing. Increased choice affects all hosts and they need to adapt and they need to innovate to stay relevant. I believe that there will be a healthy ecosystem of hosting providers of all types (shared, VPS, dedicated, cloud) and that will each be optimized for different workload needs of the end user. No one hosting provider or hosting segment will meet all the needs of every end user.
Example: Different Workload Needs of the End User
This is a great example: Rackspace recently announced dropping support for Full Trust in their .NET Windows Hosting Cloud environment (This is listed as a caveat in Wikipedia with a schedule of Full Trust deprecation). There are many components and applications that will not work in Medium Trust. Here at DiscountASP.NET we do support Full Trust in our shared hosting platform. On our Windows 2008 Hosting platform, we have delegated the trust level control so that customers can control the trust level with their web.config file. So depending on the workload of the end user, a different hosting environment will be suitable.
When all is said and done, I think I’ll have to agree with what Nietzsche said about shared hosting; “That which does not kill us makes us stronger.”
Takeshi Eto
VP Marketing and Business Development