I agree here. The point I was trying to make is how one improves a CRM system capabilities (evolve the functionalities). CRM is still user-driven, ie, a user formulate a query (which is pre-built by the systems designer), then submit it for interrogation. The system returns the results and perhaps displays those on a dashboard as what the user was seeking. The user will then interpret the data him/herself.

A state-of-the-art CRM will have an add-on data-driven capability which is far more superior than user-driven based system. Data-driven system shifts the burden of interpreting data & knowledge discovery from database from the user him/herself to the data itself via automated methods & algorithms. Most users have only been exposed to user-driven system. As with any piece of technology, you only think that what you’re using solves all your problem and comfortable with it, until someone introduces you so to something new that you’ve never seen before, and all of a sudden, you think, dang, this system is much better than what I had before.

UM said…
I would say that in my dealings with them they miss the essential truth of any business.

I agree here, after all academics are not business people. I meant to clarify my point a bit more. It is good to connect to a tertiary institution and use that relationship with their researchers as basically your ears & eyes to the world of new researches that keep emerging all the time. Not every new idea or some particular methodologies that designer of a software product would know immediately of how to solve or implement. If they manage to solve it in the first place, it is often the case it is the inefficient way of doing it. An example here, is the local SaaS product from StarSoft called Ezidoesit, for email scheduling task. I am pretty much sure that they’re not using proper scheduling algorithms at all. That’s fine, but if you’re complacent about your product because you’re the first to get to market, you never know some start-up somewhere can do exactly the same product as yours, but they do it much better. Anyone remember to duopoly of the mid-1990s ? Yep, Yahoo & Microsoft. Google, just appeared from nowhere and dominate. The reason was Google founders saw an efficient way of doing web-search than the incumbents of the day. Anyway, I had suggested to Dermott managing director of Ezidoesit to explore the possibility of linking up with the Auckland University researchers of the Engineering Science Department at the Engineering School to develop them (StarSoft) a scheduling API (ie, proper scheduling algorithms), since that department specialize in scheduling & optimization algorithm R&D.

If the designers are well connected to a research institution , a quick question to those researchers will definitely be replied back with the answers about those better & efficient methods. Also, researchers at a particular institution if you’re connected to them, can update you with new (or latest) info on researches done overseas which are applicable to your particular domain. So, using researchers from an institution gives you a wider perspective of things because they’re your ears & eyes to the international world of research.

I will quote the following abstracts as examples. Since Salesforce is still user-driven as I have pointed out at the beginning of this message, if they implement some data-driven capability as mentioned in the following links, it would definitely helps users to improve their experience in something quite superior in how data is interpreted, which it is still currently lacking. The following is the application of Rough Sets, Semi-supervised learning and Neuro-fuzzy in CRM.

- Rough set-based approach to feature selection in customer relationship management

- A Semi-Supervised Approach to Customer Relationship Management

- Customer Loyalty on Recurring Loans

There are many techniques which are applicable to the domain of CRM, but I picked those 3 above, since it emphasizes my earlier point is saying, that Salesforce designers are not perhaps aware of them (or know about them), but if they (Salesforce) do link up with a research institute, I am pretty much sure that their contacts at that institution might have updated them on latest research techniques which have now being made available in the literatures.

All valid points. I guess i was trying to say that for mainstream (mass) use, the core of CRM is pretty well defined and being delivered either thru core SF product or thru modules provided via appexchange.

I think you are completely correct about the benefits of research institutions in terms of thinking and breadth. I would say that in my dealings with them they miss the essential truth of any business. Doing just enough to service the market and make money (when in start up), AKA over engineering or bleeding edge or even fringe festival)

UM

#1) Tons of stuff to be added. I can quote functions that are currently not available in SF, that perhaps their top designers or architects have never thought of before. I am not sure whether SF does link up with researchers at any tertiary institutions or not, since that there are lots of research techniques available in the literatures that haven’t been incorporated into systems like SF.

Smothspan said…
I don’t think Salesforce.com, of all players, “gets” some aspects of the Business Model. If they did, they would radically alter the pricing on the Platform-as-a-Service offering, because the current pricing, even the radically revised pricing, is completely off target for a SaaS vendor.

#2) Perhaps, SF could look at dynamic pricing (also called variable pricing). Research shows that dynamic pricing tend to increase volume of sales compared to static pricing. But I seriously doubt that software architects at SF are contemplating on building one or perhaps have thought of it, or may be they have but they’re anticipating difficulties if they go ahead and implement it. It doesn’t mean that such functionality is undoable to SF, it is because perhaps, that they don’t link up with any research institutions. Research institutions know almost everything regarding software. A software system is only good of what the designers know before hand, where they implemented those functionalities into the product. Things or functionalities that they never thought of or never heard about previously (in most cases, those unheard of or unknown functionalities), would remain unknown to them, but as usual, it turned out that the unknown functionalities to a designer are often the better ones, in terms of superiority compared to competitors.

My point #2) re-emphasizes of what I stated in #1).