Why not have the courage of our convictions! Why should champions of BPM wilt when confronted with one of the most interesting, common and worthy opportunities for BPM?
I submit that CRM is amenable to BPM, and that BPM is an ideal platform for CRM.
First, what is CRM? (Given the question, I feel obligated to defend the original proposition!)
CRM is the ERP for part of the fuzzy space between organizations. Thus while ERP manages the domain of well-defined financial, accounting and resource management inside the organization, between CRM, sales, marketing and procurement, we have systems enabling controlled execution of engagements between agents of different organizations. (The new term, for part of this activity at least, is "systems of engagement" (SOE) as opposed to the traditional inside-the-organization "systems of record" (SOR).)
As we've all agreed elsewhere, BPM itself is both a methodology and a technology. I have also claimed that BPM is *the* technology of *work*, given that BPM explicitly surfaces work-related abstractions and symbols, for direct usage by managers and designers. Being able to directly model, execute and manage work with the assistance of technology is the goal of all technology; it is the labour-saving purpose of technology.
So we have a domain of work, "CRM", and a technology suited to assisting with work. The big question then is "Is the work of CRM suitable to BPM"?
And the answer is yes, "the work of CRM is suitable to BPM", with a caveat. (The caveat is only because both the technology of BPM and the modeling of the work of sales and marketing are not mature.)
There are three main components to the work of sales and marketing: (1) structured communications transactions, most of which are not complete, which can also be sequentially grouped in "campaigns", (2) "narratives" or "stories" which are threaded communications transactions and (3) evolving "relationship states" between organizations, which are human- or machine-harvested semantic tokens extracted from narratives or during discussions by reps.
Of course the most desired state changes we are looking for as a result of a sales process is "dealState>>closed" and "customerState>>customer". One advantage of the CRM system I've outlined here is that in directly supporting what sales reps do, more deals should result. More communications, better communications, more mini-campaigns, more stories, more story progression, more story conclusions as a win, all faster and clearer. As a goal, nothing is quite so clear as a "closed deal"!
There is growing research on all three areas required for BPM-supported CRM. The communications transaction ("I leave a message for Nandini", "Nandini returns the call" are examples of communication transactions) is well known, and highly structured. Computer-assisted narrative management research and even product deployment is growing rapidly. And semantic technologies are beginning to go mainstream.
In the context of the suggestion that customer relationships are about emotional states and that in that context the use of BPM would be a bad idea, it's worth briefly mentioning the question of language. We all speak language, at home, in business, at community events. All these venues naturally arouse our emotions. But that doesn't stop language researchers from discovering powerful insights concerning how language works. Words and language can be used to express anything expressable; in the same way, the "language of work", BPM, can usefully express anything required where technology is used to support human work.
Most CRM systems support communication transactions in greater or lesser degree, although mostly with some awkwardness. New CRM products are beginning to show some of the capabilities described here; especially so-called "social CRM" products often show a narrative thread. On the other hand, some of the most popular CRM products are what I call "narrative fragmentation machines" which are disliked by sales reps, because it takes so long to pull together everything you need to know for a call. All these things could be done better if built on a technology that supports work directly, including the work of the sales rep. (Per the comment above about BPM technology not yet being mature, full support for flexible sales situations via BPM would require capable case-capable BPM.)
In summary, CRM is amendable to BPM. This becomes clear when you unwrap what CRM is all about. The easy dismissal of BPM for CRM is only an artifact of insufficient analysis of what constitutes CRM. (The fact that there is comparatively little research on what could be called "theory of sales" is a compounding factor here.)
The work patterns which define much of the work supported by CRM (communicating, story telling, state tracking) are all types of work. Add BRM to BPM, case capabilities (for the fuzzy part), plus database and various interface portals and you will have the "BPM-based CRM that eats the world".