There is a slight balance and a primary strain in predictive dialing between minimizing dropped call rates and maximizing representative talk time. If the pacing algorithm is too hard-hitting, there won't be a representative available on the other end when a live prospect answers. When this happens, the call is likely "dropped" (hung up o by the dialer) or abandoned (the prospect hung up) because the dialer's estimate was wrong and no representatives were available to be connected to the person called. However, if the pacing algorithm is too cautious, representatives will have more inactive time and will be waiting for calls as the dialer checks for live prospects. The trick to maximizing efficiency and best fixing this tension is to scale the dialer to maximum levels so it provides the pacing algorithm with the most statistically valid data going by the largest possible statistical sample.

The Scalability Challenge

For numerous larger businesses, the hard part is that conventional dialer solutions a lot of times don't scale past 300 voice channels. Since several calls are usually made to find live prospects, the number of seats that can be supported by these systems in the real world is usually quite a bit less than that. Consequently, most large businesses have to run several dialers, usually in the same locations, to reach their high-volume needs.

Conventional dialers don't permit multiple dialing nodes to work together and share data for maximum efficiency that is the problem. There are several technology vendors that can combine administration and reports across dialer nodes, but they usually can't allow multiple dialer nodes to feed common data to a shared pacing algorithm; that is, one that can be shared across various dialer nodes. Therefore, each of the dialer nodes has to make its own "estimates" going by the division of the statistical data specific to the individual dialer node.

To say it differently, mulitple nodes to be joined so that it runs on a common statistical sample is not allowed with traditional dialers. So, all dialer nodes are not able to make predictions on the cumulative sampling of all the nodes, and every dialer is on its own that is restricted at making predictions going by only its own limited data sample.

The Benefits Of A Network-Based Software Architecture

Given the prediction efficiencies of forcing the biggest achievable statistical sample, the goal of larger businesses running predictive dialers at scale is to have a single decision-making matrix control all dialing nodes, with each dialing node sending its outcomes to that matrix in real-time so that dialing decisions can be made taking the entire data from each of the dialing nodes into account. This is the advantage of a network-based software design in the context of predictive dialing. Companies that run dialing campaigns at scale on several stand-alone dialers can go through wonderful efficiency and productivity benefits by moving toward network-based dialer solutions.