At Home or in the Clouds-and in Open Spaces Between
In the CRM world, the argument has been and will continue to be the argument over what’s better, software as a service, which is evolving into computing in the cloud—or is it;or the on-premise deployment of CRM, which means the deployment of CRM for you by you at your business. The simplest part of the argument has always been around the “who owns the data” controversy—though it is driven by the vendors who have a bit of “nyah nyah nyah” embedded in the ongoing battle. Should it be you who will license the software, provide the servers, and supply the remaining necessary hardware, software, and personnel to maintain the system so that your business can control the data directly? Or should it be allowed to fall into the hands of a host who will provide all that hardware, software, and personnel overhead to install and maintain the system for the price of a subscription—but, horrors, in their own environment—meaning the data is in their hands, with, of course, potentially nefarious results.
Ah, if it were truly that simple. There is a simple answer to the onpremise versus on-demand (another name for it) contention—it depends. But that, of course, is what makes it more complicated.
There is another factor that makes this even more interesting. The development of open source CRM showcased by the ubiquitous presence of SugarCRM. While a seemingly additional headache, it actually is a very important part of the discussion as we move to Social CRM.
Just to set some expectations, I’m not going to jabber much about on-premise or on-demand—at least, not as much as I have in the past editions—since they are now well-established models that don’t need me to explain them over and over again. I’m just going to provide some basics and a comparative checklist for you on the criteria to consider for selection of one or the other. I will dwell a bit on cloud computing because it is new and becoming important and is still a source of confusion for most people. I’m also going to clarify the relationship between open source and on-premise/on-demand/cloud computing, which is something like peaches to peach pits. Open source isn’t competitive with the delivery models above. It isn’t a delivery model and, in fact, as we’ll see, open source CRM (or any) applications can be delivered many different ways.
Rather than bore the living hell out of you with a long technical discourse on the differences between the varieties of on-premise software and on-demand services, let’s take a look at the basic definition of on-premise to start. By the way, technically, the actual term for this is “on-premises, ” not “on-premise.” I’m going to leave it as on-premise because, as wrong as this phrasing might be, it’s what most people already call it. I’m going to stand by it due to the Common Law Usage Rule, which says that even if it’s wrong but in common use, it’s right. By the way, I just made that up.
What Is It?
On-premise is the traditional model for software installation and administration that you probably already know something about. You license the software from a software vendor. You run an instance of the software or multiple instances at your site and store the data associated with that application on servers physically located at your site with your administration and your security. You control the data. You also hire someone to install and customize the software and then stick around to get that licensed software working, since the odds of you being able to do it yourself are about the same as the odds of the software vendors liking each other. Astronomical. Over roughly a threeyear period, you tweak the system, handle downtime yourself, and at the end of the three-year cycle, you upgrade the software to meet the requirements you’ve developed over that time.
This model has advantages for large enterprises in particular because it provides optimal control over your own feature/function/ process/data destiny and it scales better than the on-demand versions. This doesn’t mean that SaaS doesn’t scale. As we’ll see, it most certainly does. But most on-premise versions that scale are built specifically for the size of the enterprise that they were built for.
A complaint about on-premise that you often hear is that implementation can take months. While this is still true for the most part, there is some potential relief in sight. For example, late in 2008, I watched a demonstration by a Microsoft Certified Partner, AlfaPeople, which can provision an on-premise or on-demand Microsoft Dynamics CRM 4.0 system so that it is up and running in 10 minutes—without customization, of course. Even so, that indicates how incredibly quickly these systems can be set up. It actually took two seconds to install and provision a Microsoft Dynamics CRM system in the demonstration I saw. While long on-premise installations are hardly a thing of the past, there is progress toward the future.
Advantages and Disadvantages
These have been talked about so frequently in the technology press that it’s reached something of a blah blah blah status. I’m listing them here so you can start asking the questions you’re going to need to ask of vendors when the time comes for you to implement some sort of CRM technology. Table has a brief description of the advantages and disadvantages of on-premise purely on its own merits, not in comparison to SaaS.
While there is no need for me to highlight any one vendor in the onpremise world, there are several who have notable products. They are, in no particular order:
- Oracle With both Siebel and PeopleSoft, Oracle stands out for the depth of the suite. When the long-awaited Oracle Fusion products emerge, there will probably be a new CRM product to contend with, but the release date for that is still fuzzy. Target market: Large enterprises.
- SAP With the release this year of SAP CRM 7.0, SAP moved into contention with a truly integrated enterprise suite. They added an enormous number of new features and functions to every part of their CRM suite—much of which was built in conjunction with their customers’ advice. For example, they’ve added Pipeline Performance Management and Territory Management to the sales force automation module. It is fully integrated with all aspects of SAP’s Business Suite. Target market: Large enterprises.
- Microsoft Dynamics CRM 4.0 has a common code base for both the on-premise and on-demand versions, making a genuinely hybrid application. It is priced to sell, as they say in the retail world. It is also a functional platform with open source components being made available by Microsoft to developers— a marked change in strategy for this rather big company. Target market: Small and midsized companies, departments of large enterprises.
- Sage CRM Sage is the multibillion European software empire that has a back and front office focused suite. While not much for supply chain management, they are focused on financials on the back end and CRM on the front end, with two products that are deliverable in on-premise and on-demand versions, though they are moving more and more to the latter. SalesLogix and SageCRM are both based on RESTful architectures, as you probably saw in “SOA for Poets”. Target market: Small and the lower end of midsized businesses.
Needless to say, this is not an exhaustive list. It simply is meant to convey the news that on-premise is far from dead. In fact, Gartner Group, in a SaaS market study they released at the end of 2008, estimated that by 2011, SaaS will control roughly 25 percent of the market—which means on-premise will continue to control roughly three-fourths of it—so it’s hardly buried yet. So if you see it walking around, it’s not a zombie, it’s alive.