The end of Exchange? Probably….
I, like many of you, follow prominent voices in the Exchange community through blogs and other outlets. One recent post by Tony Redmond, MVP and technologist, caught my interest more than usual. His recent post ‘The end of Exchange? I don’t think so’ was a response to a blog post by Scott Cameron. Tony makes many good points, and lays out some thoughtful counter-points to Scott’s premise. While, like Scott, I also work for Microsoft – so as Tony point out (rightfully so) you should take the vendor’s bias into consideration, I would like offer an additional insight that I think both have missed, and why I do think the end is near (oohhh and aaaah).
To this point, the service and on-premises offerings have more or less been close in feature parity. So, Scott’s arguments make sense..it is largely a cost/benefit analysis that customers make to decide if Office 365 is right for them. Given the extremely low cost of the service, sometimes there are a few other levers that come into play. For example, whether or not a customer can accept the evergreen model, or the other “excuses” (in Scott’s words) also play a factor in the decision process. That’s pretty much where we have been from the last few years till today.
Now, the game changer. There is and will be more functionality, and more innovation in the service. There will be features, that will never make it to the on-premises product. Not because Microsoft doesn’t want to offer them, its because these are features you can only get when you have the entire platform and can do more interesting things farther up the stack. Let me give a few examples. First, we recently released multi-factor authentication (MFA). In just a few clicks, you can, for free, enable MFA for your administrators. No infrastructure investments needed..super easy to enable..enhanced security beyond most typical IT shops (I am in our services group, I see a lot of IT shops). For some nominal fee (judgment call), you can enable this per user / per application (I’m just learning this – so I think I have my facts straight). I have my MSDN subscription, that includes Azure, and I also have a trial Office 365 tenant that I enabled this on. So here is just one example on how the service continually is enhanced and can provide new features that would be rather difficult to deploy in a typical enterprise.
I’ve also been playing with Azure RMS. For anyone who has tried to deploy this on-prem, it couldn’t be easier to do with Azure RMS. All that is required is installation of a RMS connector (on-prem) and all of your on-prem services (like Exchange) are tricked into thinking you have deployed an on-premises ADRMS. Again, simplicity and integration makes cloud services like this incredibly compelling.
Ok, so that’s two examples of where deploying a service is just easier..what else? As I alluded to at the start – the gap between features will grow. One example is Microsoft ‘Oslo’. At this point, I’m assuming Oslo will only be a feature available in Office 365. Another announcement was made around new group functionality – bringing together Yammer and Exchange integration, which I personally think will be the lynch pin to finally realize the power of enterprise social and collaboration. These types of solutions require bringing together information or integration across multiple applications. For an enterprise to be able to deploy and maintain would be difficult. I also can see a future when new features are developed from having access to a large user base (not saying access to specific tenant data, Im talking about telemetry data and possibly anonymized meta-data). These applications may just not be easily productized.
There you go, that’s my bet on the future. Customers will move away from cost decisions to choosing software and services that help them get their work done, sell more widgets, or collaborate more efficiently. People will likely be able to try and buy, or use a freemium model – where services are used because they fulfill a demand. The back-end will make it easy to consume and use. Consumerazation of IT has already started breaking down traditional barriers. Look at Box, or Yammer where individuals can use a service for free. Enough people use it, IT can choose to buy it and enforce controls. People purchasing their own software on their own hardware that interop with enterprise services (like EAS clients and Exchange). Need more storage – don’t bother going to IT, just click and add to your cloud storage for $1.99, oh and its automatically highly available with business continuity. This future is why Exchange’s future (and other traditional on-prem applications) have a horizon – and I’m going to bet that its not that far off.
Am I biased, probably. Not because I work for Microsoft – but because I live in that world today. I’ve moved all of my files to OneDrive for Business after my laptop HD died. This move also let me access my files across my devices (I have a Surface RT, Dell Venue Pro 8, Laptop, Desktop, and a Windows Lumia 1020 phone). Also, I can share my files more easily, which means I can collaborate with less friction. This also enabled multi-author editing, which is the greatest feature ever! The line between internal and external collaboration is blurred. You can reach me on Lync, get me on email, or Yammer. I can secure my documents with RMS – even with non-Microsoft people. If I need to electronically sign documents – I can purchase an Office Application from DocuSign. All of these things add up to the fact that the service model empowers me to live and work better..
Its my job as a Solution Architect at Microsoft to help customers realize this vision. My current role is largely working with customers who are deploying on-premises Exchange. This shift affect me, just like the customers I help on a daily basis. But, I’m excited about this future because I do enjoy helping customers solve business problems, and I see the IT industry shifting in this direction. Only time will tell 
p.s. These are solely my opinions and do not represent Microsoft or anyone else’s. There is no insight to inside information, or customer information implied.