Working the Phone as a CRM Tool (1,572 words)
3. Adjust auto-dialer and script screen speeds to reflect the desired tone of the presentation.
As with all aspects of P/CRM, success comes to those who attend to detail. The prolonged, conversational, interactive approach that is often so effective for low-response/high-value prospects will be undermined if the auto-dialers and script screens remain on a rapid cycle. A relaxed ambience cannot be created unless the technical infrastructure is tuned accordingly.
4. Adjust compensation structures.
Consider a call center environment where compensation is based on the number of sales generated per hour. Under these circumstances, it is in the financial interest of the TSRs to be assigned to the high-response/low-value group. Those who are not will quickly become disenchanted. Only by developing multiple compensation structures, each tailored to the specific objective, can such tensions be resolved.
5. Understand the respective roles of strategies that are reactive/rules-based versus anticipatory/predictive.
Sophisticated call center P/CRM requires statistics-based predictive models to drive contact strategies. However, it is important to recognize the limitations of such models. Despite the claims of some practitioners, they are not the solution to every problem.
Consider the challenge of predicting who is going to voluntarily cancel his or her participation in an ongoing service. This is called "attrition" in the financial services world and "churn" in the telecommunications realm. Much effort has been focused on identifying those customers who are on the verge of voluntary cancellation.
The unfortunate truth, however, is that many of these efforts have met with at best limited success. This is because, while behavior that is driven by interactions between customers and service providers generally is conducive to predictive methodologies, the same cannot be said for behavior that is the result of aggressive competitive thrusts. This is particularly true for commodity or near-commodity services.
Data-mining techniques, for example, generally can identify and quantify a rise in cancellations in the wake of a decline in usage, or an increase in defection subsequent to the onset of service problems. However, they can be essentially useless in the face of sudden and bold competitive attacks. Often, such external influences are so strong that they create a tsunami of sorts that overwhelms all existing predictive models.