In an ERP System, the core functionality has been well defined since the 90's. Some companies might need a different distribution module or a \n\nfancier scheduler algorithm, but MRP is pretty much MRP. An accounting system? You'd better not have a lot of creative requirements.\n\n\nCRM Definition and Solutions\n\n\nBut a CRM system is used by those right-brained types who bring you revenue, and there is significant variation in functional requirements from \n\ncompany to company. Even the precise definition of "CRM" can be debated if you get enough consultants in the room. So it's all too common to have \n\nwide-ranging discussions about marketing automation, call center features, SFA, forecasting, order entry, e-commerce, customer support, and customer \n\nportals. This makes for a very long feature list to be evaluated, ranked, and budgeted for.\n\n\n[ For timely data center news and expert advice on data center strategy, see CIO.com's Data Center Drilldown section. ]\n\n\nWhile these discussions about requirements are important, they also distract management from the issues that matter. Take a look at why.\n\nSecret #1: Features are less important than User Adoption\n\nA CRM system without users is just a database schema with a user interface. So getting users to get on the system early is a key success factor: user \n\nactivity is what starts the virtuous cycle of CRM (more users mean more relevant information which makes it easier to attract more users). Of course \n\nfunctionality (and, let's face it, some cool bells and whistles) will help attract users. But the early adopters won't need that much functionality to get \n\ngoing.\n\n\nAs I've written endlessly in this column, deploying features in a "big bang" release is not the right approach for CRM projects. Agile techniques and \n\nincremental deliveries let you optimize for ease of adoption and early business value. So instead of focusing on one feature list, work the problem as a \n\nseries of small but coherent feature sets that each solve a contained business problem for the users.\n\nSecret #2: Features are less important than Data Credibility\n\nA CRM system is only as good as the data asset it holds. The credibility of CRM data comes from its business relevance (timeliness), accuracy \n\n(cleanness), and scope (number of users and systems feeding the database). Yes, functionality contributes to the data credibility, but it's far less important \n\nthan:\n\nThe care taken in doing data migration and imports\nThe quality of the external system integrations (watch out for systems that corrupt fields, or worse, create duplicate records)\nThe tightness and consistency of semantics for key fields (such as "sales stage" or "case resolution")\nThe precision and realism of business process definitions surrounding the system.Secret #3: Features are less important than Platform\n\nThere's an old saying that the best CRM systems are built, not bought. While it might be more precise to say that the best CRM systems are assembled \n\nfrom subsystems that are bought, for sophisticated customers there is no off-the-shelf system that can check all the boxes.\n\n\nSo the key issue in evaluating CRM systems is, how malleable are they? Can their user interface and object model be configured to meet your needs \n\nwithout coding? When development is required, does the system have the richness and stability of APIs that makes custom coding a good decision? \n\nDoes the platform provide the two-way data access that will enable solid, real-time integration with outside systems? Is the system architected as a series \n\nof Web services?\n\n\nAn attribute of the platform is the number of third-party extensions and products. The more there are, the less likely you'll have to dedicate \n\ndevelopers to create features. If your developers want to do a PHP deepdive, I can think of nothing better than SugarCRM. If they have deep C++ or \n\nC# expertise, they can have fun with Microsoft Dynamics. If you want the smallest project size, Salesforce.com and its APEX language is probably the \n\nbest path. With any of these, you have to ask: Does the platform have good development, testing, and deployment tools? Does the platform facilitate \n\nincremental deployment?\n\nSecret #4: Features are less important than Reliability\n\nThink about reliability in a new way: Is the system good enough that you can rely on it to make key business decisions? Remember the "-ilities" from \n\nenterprise software? Thanks to Web services and SaaS, they've been expanded for modern CRM systems:\n\nIs the system reliable, accessible and responsive during 99.99% of your business hours?\nIs system stable from release to release, both in terms of API compatibility and data preservation?\nIs the system presenting a credible picture of the business relationship, or is the data fragmented or even contradictory?\nDoes the system provide fine-grained security, access controls, and audit trails?\nIs the system usable, even for impatient sales reps and executives?\n\nOf course, when you're going for budget and making project plans, a well organized feature list is indispensible. But know that the four secrets \n\ndescribed above will mean more for your success than any item of CRM functionality provided by the vendor.\n\n\nDavid Taber is the author of the new Prentice Hall book, "Salesforce.com Secrets of Success" and is the CEO of SalesLogistix, a certified Salesforce.com consultancy focused on business process improvement through use of CRM systems. \n\nSalesLogistix clients are in North America, Europe, Israel, and India, and David has over 25 years experience in high tech, including 10 years at the VP \n\nlevel or above.\n\n\n Follow everything from CIO.com on Twitter @CIOonline.