Disclaimer: I work for Microsoft as the Product Manager for Online Services (as well as SharePoint). Below is my opinion only, it is not sanctioned by Microsoft and it may or may not be right.
A few weeks ago I spoke at the Auckland Cloud Camp (and did a repeat last week at the Wellington cloud camp). You can watch my Auckland presentation from a bad angle here, and the Wellington one here; but I promised Mr Ben Kepes that I’d write up a blog post summarizing my thinking too. This is that blog post.
As I write this, I’m writing it from the perspective of someone who looks at the entire Microsoft Partner ecosystem from the perspective of both an outsider (I don’t work for any of them) and an insider (I have a vested interest in their success and work with many of them on a daily basis). Your perspective will make a big difference to how you see things panning out. For example, my basis is that a large % of the IT industry in discussion below deals with the sale, deployment and maintenance of Microsoft based systems. I’m not really considering specific vertical applications (e.g. Xero) and the impact these apps have on the industry, nor am I consider companies that focus on non-Microsoft technologies – although I’m sure they too will be affected.
Business models possible with the cloud
The cloud is potentially one of the biggest changes to the IT ecosystem since the mainframe market transitioned to the PC market of today. I don’t think it will be quite as big as the mainframe/PC transition, but I expect it will be the biggest change since then.
Lets start by looking at business models that are possible with the cloud. There are six business models available today in the cloud.
1. Consumer. A consumer is one who creates an application that makes use of cloud based technology to create a solution that customers will pay for. Again the example I’ll use here is Xero. This is the focus of many entrepreneurs, so I’m not going to focus on it here beyond saying that I think this is where the cloud provides the most opportunity.
2. Consultant. The consultant I have in mind here is someone who is familiar with the cloud, the opportunities and cost savings it affords and who sells their time to customers to help them figure out a cloud strategy. In my opinion this is a role that provides value while there is uncertainty and a general lack of understanding of the cloud and what it means. But in two years time the raw consultant will be extinct. Why? Because once there is a general understanding of the cloud in the IT industry, the consultant won’t offer particularly good value for money as in house architects, etc will know enough to do this themselves, and given companies that have architects that are more likely to hire consultants, this opportunity will go the way of the dinosaur.
3. Reseller. The reseller is someone who sells the services of a a cloud provider. Think of this role as the cloud equivalent of the procurement specialist. While this would seem like an obvious place to get involved, there isn’t much money in it. For example Google apps resellers make US$10 from every US$50 per year sale. Thats a nice healthy 20% margin. For comparison, Microsoft resellers make 18% in the first year and then 6% for the subsequent two years.
This may (or may not) sound pretty good. But how many users do you need to sign up to make a living of say US$40k off this. With Google apps that means you need to sign up 4000 individuals. Don’t jump out and write a business case to start a business just yet though, because this market is a commodity market. Expect prices (and your commission dollars) to decrease year on year, meaning your customer acquisition requirements will increase significantly over time.
4. Implementer. Implementation of cloud based solutions is a twofold opportunity. Firstly there is the opportunity to migrate customers from on-premises solutions to cloud based solutions. This would be from (for example) Microsoft Exchange 2003 to Microsoft Exchange Online (or a similar competing offering). The second implementation opportunity is to migrate from one vendors cloud based solution to another. While we at Microsoft are seeing quite a bit of interest in this space at the moment, generally I would expect that once in the cloud it will be hard to migrate a customer out of it – even to a competing solution.
What this means is that if you are implementing solutions and your business depends on this, you shouldn’t generally plan on getting repeat business. Once a customer has been migrated they don’t need any server maintenance, license or hardware procurement and won’t be asking you to upgrade them to the next version because that happens at the providers end – without a System Integrator needing to be involved. In other words, once a customer is migrated to the cloud, that’s it – find another customer because this one now has much lower IT requirements and is largely independent of the implementer – at least for that solution.
5. Provider. A cloud provider is a company that hosts servers and provides an product based service to customers. Google is a cloud provider, as is Microsoft, Amazon and a number of other companies. Today, there are a number of companies that have invested in data centres and will use those data centres to become a cloud provider. This is a logical way to take advantage of existing investments and capacity.
However, this option doesn’t look great in the long term. Being serious about being a cloud provider means significant risk and cost. Providers pay for the power, cooling, space, hardware, software and technical gurus that keep their solution running – all the things your customers used to pay for. Don’t forget to multiply that because providers need multiple redundant options – if the data centre in one location is destroyed due to an earthquake, fire, or other disaster, your customers are still going to want to keep running their business – bearing in mind they may be in a different city or even country and not at all affected by the disaster. In addition to this, the pricing for cloud based solutions will be driven by the companies that have the scale – the big companies – and the direction of that pricing will be down. So, over time the costs of being a provider will increase (as your customer base increases), but the revenue per customer will decrease. This will affect all providers – small and big alike, but the bigger providers will be more able to weather the storm to market maturity.
Over time, I believe that being a cloud provider for core solutions such as email, collaboration, infrastructure and platform will become a “big boys” game. That is to say that I expect there will ultimately be a small group of big companies providing 80-90% of these cloud solutions – leaving a declining market share for competitors to play in. This will be driven by the scale these companies are able to achieve and the expansion of data centres by the current big players to meet their customer distribution.
Today there is a short term opportunity in verticals like government, but in time the big companies will dominate these areas too (Microsoft recently announced a government only option in the US for example).
I expect smaller cloud providers will either be purchased by the bigger companies (primarily for their data centre presence or customer base) or will struggle and perhaps in some cases fail.
6. Combinations. For companies that provide consulting, procurement and solution implementation of on-premises software today, it would seem obvious to expand this into the cloud. In reality, this is the only way to embrace the cloud as IT services firm. However, as (for example) email increasingly takes to the cloud, the services revenue from this part of the IT industry will dwindle. The same will happen as more and more workloads head to the cloud.
Companies that specialise in cloud based services today, will in the long run either be purchased by a bigger IT services provider (assuming they can hit critical mass with customers) or struggle to come up with a value differentiator from their larger competitors and disappear except for some niche areas.
Conclusion.
The obvious prediction here is that as the cloud takes off and gains customer acceptance, and more and more of today’s on-premise applications go into the cloud, the IT services industry will also shrink as illustrated for email and collaboration in the two charts below. 
The end result is that I think business consulting (rather than just IT consulting) will be a increasingly important part of the business model of the current IT Services company. In time we may see consolidation between consulting firms like of Deloittes and PWC with today’s System Integrators, but then there are challenges there too.
The end is not yet nigh though. This transition will take several years to make, and there is plenty of opportunity to make sure you work through this transition. Not only this, but there are some scenarios where customers simply won’t make the jump to the cloud for various reasons. This blog post is really a call to think through your business and how you can embrace the cloud and start you thinking about the opportunities and change that is coming to the IT industry.
In the future I’ll post a number of other posts that will give you further food for thought.