Thanks and welcome back to the CRM Products and Technology blog from Consona CRM. In my last post, I talked about my perceived disconnect in how the CRM market is viewed by different press, analysts and industry experts and called for a consolidation in their views and opinions. This post, I am going to discuss an announcement we made recently and a topic that I think is pretty substantial and important for all enterprise applications. That subject is cloud computing.
If you didn't hear about it, Consona announced plans to offer our CRM products as Cloud solutions in early 2010. We are very excited about this offering and are betting heavily that this solution stack will be highly valued for both new and existing customers. In fact, we are calling this offering the next generation of SaaS or SaaS 2.0. That's a pretty bold statement, but there are some very real and tangible reasons why this solution is truly an evolution to the first generation of SaaS CRM offerings.
First let me give a brief description of the first generation SaaS products. From an architectural prospective, top to bottom, everything is multi-tenant. This means that from organization to organization and from company to company, customers share all components. This includes the hardware, operating infrastructure and core operating systems; the database management systems; the data platforms; the application platforms; the application business logic; and the presentation layer of the application. And the vendors that supply this type of architecture typically provide a very deep configuration framework so that each customer can make business-specific operational configurations for that single application instance or tenant. Also these vendors provide an upgrade framework to allow for mostly seamless upgrades for their customers. And it should be mentioned that this type of offering is typically priced on a subscription basis, typically on a per-user, per-month type of arrangement.
There are some real tangible benefits to this architecture for some applications. And, if a high level of performance, reliability, security, portability, customization, or system-to-system integration isn't a concern (e.g., not mission critical), than you probably have a pretty good option. An expensive option, but a good option nonetheless. Now, I get beat up for this set of comments all of the time and let me explain. Yes, multi-tenant application SaaS 1.0 vendors allow some level of integration via web services to a fixed (and potentially slightly extensible) API. And this is great. But I don't think there is anyone who actually believes that web services are yet sufficient for high-load, system-to-system integrations. And, if the legacy system being integrated also has an inflexible API, well you might be stuck with exporting and importing bulk data. And by the way, importing and exporting large quantities of data isn't particularly great with these systems either. But like I said—I think some applications are well-suited for SaaS 1.0. I know that our expense reporting system is a SaaS 1.0 application, and it is fabulous.
However, there are some real limitations with this type of solution, especially for complex, enterprise-level applications. In particular, if you share all of the resources and applications with everyone else, there is a major issue with performance during peak demand hours. There are huge reliability issues in this infrastructure. If one customer goes down, they all go down (or at least a significant number). You, as a customer, are locked in to this deployment and, ultimately, this vendor, and are forced to conform to their operating model, even if your business conditions (or the vendor’s) change. Configuration is not sufficient in many, many circumstances, no matter how sophisticated the tools are. Sometimes you just need to make core business logic modifications to meet your business needs, and this can only be accomplished with ripping into the code and modifying it, something that is simply not available in any multi-tenant SaaS 1.0 application. And let's not forget, the pricing model is expensive. I know vendors that do subscription-only pricing claim that you have lower up-front costs, which is true, but do not be fooled that this is the lowest cost option in the long term. It simply isn't true for enterprise applications. This is not unlike leasing your car. Yes, it is cheaper than purchasing your car over the first couple of years. But not over the long term.
Several years ago, the idea was beginning to formulate about platform as a service. I started to think about this, and I thought to myself, "Wow, now that might get interesting." A few years later, we now have the concept of cloud computing. The definition of cloud computing according to Forrester is "a pool of highly scalable, abstracted infrastructure, capable of hosting end-to-end customer applications, that is billed by consumption." Eureka! SaaS 2.0 is born. Let's define this a bit from my perspective. First, we now have multi-tenancy. It is, however, unlike SaaS 1.0 solutions, multi-tenancy in the right area. The operating systems, the infrastructure, the networking, and so on, is all multi-tenant. And it is charged on a consumption basis, meaning that you only pay for what you use. To me, this makes sense. Your application data isn't intermixed with other data sets. Your business logic can be modified at the lowest level. You can perform large-scale system-to-system integrations by private dedicated infrastructure. You can do large imports and exports to the systems. And, above and beyond all of this, it is likely more performant, reliable and secure than your own internal network. Another aspect of SaaS 2.0 is operational portability. In other words you not only have the ability to take it in-house, but you also have the ability to go to another cloud offering if you don’t like the pricing or SLAs or security or whatever. The point is that the choice is yours. Absolutely no vendor or infrastructure lock-in.
With our announcement, I saw many tweets on our move, most of which were and are incredibly positive. I only saw one that implied the market should look down upon this type of deployment. One, supposed industry expert who is an "independent analyst focused on SME biz apps, primarily (but not exclusively) in support of Microsoft Dynamics channel partners" (by the way, I put this in here because I think it is hilariously biased) said something rather negative about our approach. He commented something to the effect that obviously, we were going to be hosting our single-tenant applications, as if this were a negative thing.
So, to this tweeter, I say proudly:
The next generation of SaaS, or SaaS 2.0, will absolutely be multi-tenant at the operating system and core infrastructure layers. It will be single-tenant at the database and application layers. It will have the benefits of better scalability, reliability, security, integration and performance than multi-tenant SaaS 1.0 applications. It will be charged for on the basis of usage and be portable for customers to do with as they will. In almost every way, SaaS 2.0 is superior to SaaS 1.0, especially when the application matters and is actually mission critical.
SaaS 2.0 is here. Get ready for Consona Support Cloud.
Disagree? Comment here or email me at firstname.lastname@example.org.