Apple Telecom Bedienungsanleitung PDF herunterladen (Seite 23)

21 | What customers really want – A customer-centric strategy for telecom operators

If a customer stops making some of his or her high-volume calls within the

system or if one of his or her call parties leaves the system, the customer is

considered to be at risk. To design retention campaigns based on the churn

prediction model, the telecom operator can address the customer at risk pro-

actively via outbound calls. All calls are recorded and evaluated by a voice

analysis system. The system identifies key words to evaluate customer

preferences, which are then used to help the agent define the offer for

the customer.

The results of the campaigns and the customer preferences are played back

to the churn prediction and centralized data warehouse team. They can

(pre-)classify the whole customer base and apply the classification in other

functional units, e.g. the shops. The units involved in the process work to-

gether interactively, and communication with the customer is an interactive

process, too.

The entire organization has embarked on a continuous learning process.

To get on this trajectory, companies are often organized by segment and

set up special coordination units.

An example taken from the Russian market illustrates the churn prediction

approach described:

Challenge: Russia's mobile operators are facing low customer loyalty with

high churn and multi-SIM usage in a fully saturated prepaid market. A spe-

cific firm is aimed to identify at-risk customers in real time to set up counter-

action measures when customers start to use a competitor's SIM card.

Solution: To detect at-risk customers, social relationship analysis of call

data records (CDRs) for on-net and off-net calls is used. When identifying

a customer who is likely to start using a competitor SIM, the customer is

contacted through text message campaigns and calls. Obviously, this can be

done only if the customer is still using his original SIM, either to spend the

remaining prepaid balance or as a multi-SIM. When a customer logs onto the

operator network, the company tracks it in real time using home location

prevent churn. In order to enable this approach an interaction between the

units for network, data mining and call centers has to be established with

fast reaction times.

1 2 ... 18 19 20 21 22 23 24 25 26 27 28 ... 31 32

Keine Kommentare

Apple Telecom Bedienungsanleitung Seite 23