Method for preparing an optimal alternative billing plan for mobile telephony users managed through a call center

• US 8,738,482 B2
• Filed: 12/02/2010
• Issued: 05/27/2014
• Est. Priority Date: 12/02/2010
• Status: Expired due to Fees

• Current Assignee: Telefonica SA
• Original Assignee: Telefonica SA
• Inventors: Martinez, Enrique Frias, Leo, Juan Moises Pascal
• Primary Examiner(s): Iwarere, Seye
• Application Number:
  US12/958,983
• Publication Number:
  US 20120143735A1
• Time in Patent Office: 1,272 Days
• Field of Search: 705/34
• US Class Current: 705/34
• CPC Class Codes: G06Q 30/04 Billing or invoicing
• CPC Class Codes: H04M 15/43 Billing software details
• CPC Class Codes: H04M 15/51 for resellers, retailers or service providers
• CPC Class Codes: H04M 15/58 based on statistics of usage or network monitoring
• CPC Class Codes: H04M 15/80 Rating or billing plans; Tariff determination aspects
• CPC Class Codes: H04W 4/24 Accounting or billing

Abstract

The invention relates to a method for preparing an optimal alternative billing plan for mobile telephony users managed through a call center, which is applicable to the recommendation of an optimal telephony plan to the customers of an operator, considering the consumption of the customer, the risk of the customer migrating to another and the market strategy of the operator. It comprises the estimation of the real bill of a customer during a time period covering the last M months, a simulated estimation of the bill of the customer using any generic plan based on the mean traffic of the last M months, applying a simulation algorithm and the estimation of the churn risk of a customer according to the value of the said customer, the value of the operators on the market, number of better offers, number of the customers which he calls per operator and the simulations performed with the generic plans, applying an estimation algorithm.

Claims

1. A method for preparing an optimal alternative billing plan for mobile telephony operators managed through a call center, based on processing by at least one first service server of information relating to billing of different telecommunication services used by a customer, including duration and time slot of use, the method comprising:

(a) estimating a real bill of the customer during a selected time period covering M preceding months, using a processor of the at least one first service server, by processing accumulated mean traffic, extrapolated to the M preceding months, obtaining a mean consumption of the customer in the selected time period;

(b) simulating estimating a bill of the customer using at least one generic plan based on the accumulated mean traffic of the M preceding months, by using the processor to apply a simulation algorithm;

(c) estimating a churn risk of the customer, using the processor, according to a value of the customer, a value of operators on the market, a number of better offers, a number of customers called per operator, and the simulations performed with the at least one generic plan, applying a churn risk algorithm; and

(d) selecting an alternative plan and presenting to the customer the selected alternative plan in the form of a recommendation,

wherein the (a) estimating the real bill, the (b) simulating estimating the bill, and the (c) estimating the churn risk are performed by the processor using data supplied by the first service server.

2. The method according to claim 1, wherein in (a), the mean consumption of the customer in the selected time period is acquired considering dimensions comprising:

consumption scenario;

different possibilities of destinations of calls that a customer can make, dependent on a regulatory framework, from a group comprising local calls, national calls, calls to other mobiles, incoming calls, or outgoing calls;

day of the week;

differences of use dependent on day of the week; and

time slot;

consumptions of the customer are stored according to several time slots considered.

3. The method according to claim 1, wherein in (b), a value of the bill of the customer during the selected time period of M months is first calculated applying a current billing plan at a time of the estimating, and the estimating applying other plans will be performed according to whether or not there is a deviation of ±

5% from the real bill with respect to the simulated bill in each month.

4. The method according to claim 1, wherein the estimating of the churn risk of the customer comprises an algorithm operating a programmable risk function formed by variables comprising a number of billing plans which reduce cost with respect to a current billing plan, a number of operators which have better billing plans, customer value, and an operator to which the customer makes a majority of calls, and wherein the churn risk of the customer is expressed as a score per customer, wherein a comparatively higher value indicates a comparatively greater perception of calculated risk.

5. The method according to claim 1, wherein in (a), data for estimating the real bill of the customer comprise a first piece of information about bills in a form of structured data including a customer number, a telephone number, a month and year of analysis, and a bill charged in the month and year of analysis and a second piece of information in a form of structured data relating to a current offer of the customer, including, in addition to the customer number and the telephone number, a code indicating a contracted offer at a time of performing the estimating, as well as a detail of all additional services contracted by the customer in an estimation period.

6. The method according to claim 5, wherein in (a), the estimating also takes into account fields comprising:

consumption scenario;

day of the week;

time slot; and

indicator of frequent numbers.

7. The method according to claim 1, wherein in (b), data for the simulating estimating comprise a piece of information which details an aggregation of consumption information for other plans and which comprises an evaluation of numbers susceptible of being added as “

frequent”

numbers, “

time slots of greatest use,” and

contracted “

service packages.”

8. The method according to claim 1, wherein an estimation algorithm comprises accumulated traffic information of each of the customers distributed over a three-dimensional matrix which considers dimensions of call destination/day of the week and time slot for each offer associated with a customer.

9. The method according to claim 8, wherein there is introduced in the three-dimensional matrix a correction according to packages contracted (amount of minutes or money units which have already been paid in a monthly fee) by a determined customer according to selected destinations (SDs).

10. The method according to claim 9, wherein the package is distributed percentage-wise among the SDs using a correction factor according to the following formula:

11. The method according to claim 1, wherein a risk function f assigned to a subscriber is given by parameters comprising:

Customer Value;

Value expressed per unit which indicates for each value segment, defined by the operator, an importance value (VC) is assigned;

Delta Function (DELTA);

Range of difference as a percentage which indicates a difference between the current offer of the subscriber and any simulated offers, an importance value will be assigned according to the percentage, i.e., a comparatively greater difference is associated with a comparatively higher imputed value;

Better Offers (OF);

Value assigned to a number of offers better than the current offer after having performed the simulation, and which takes from 0 convenient plans (none) to N convenient plans;

Frequent Operator (NF);

Value assigned to an operator most called by the subscriber and which will correspond with a value 100, and 0 for other operators;

Number of Best Plan Operators (OP);

Value assigned to a number of operators with which the customer may have had better offers, a minimum value being 1 and a maximum value (a number of operators which operate in an analysis framework); and

Operator Value (VLOP);

A value between 0 and 100 is assigned, 100 being a maximum value, according to an importance on the operations market,

creating the risk function f with the parameters

ƒ

(Ci)=VC*(DELTA+OF+NF+OP+VLOP)

which measures the churn risk of the subscriber in the analysis framework.

12. The method according to claim 11, wherein once the risk function f for a customer has been calculated, the churn risk of the customer is calculated using risk levels comprising (1) very low, (2) low, (3) medium, (4) high, and (5) very high, the risk levels being defined as ranges of values of the risk function f, and a maximum obtained for each of the simulations is selected.

13. A non-transitory computer readable medium storing a program causing a computer to execute a method for preparing an optimal alternative billing plan for mobile telephony operators managed through a call center, based on processing by at least one first service server of information relating to billing of different telecommunication services used by a customer, including duration and time slot of use, the method comprising:

(a) estimating a real bill of the customer during a selected time period covering M preceding months, using a processor of the at least one first service server, by processing accumulated mean traffic, extrapolated to the M preceding months, obtaining a mean consumption of the customer in the selected time period;

(b) simulating estimating a bill of the customer using at least one generic plan based on the accumulated mean traffic of the M preceding months, by using the processor to apply a simulation algorithm;

(c) estimating a churn risk of the customer, using the processor, according to a value of the customer, a value of operators on the market, a number of better offers, a number of customers called per operator, and the simulations performed with the at least one generic plan, applying a churn risk algorithm; and

(d) selecting an alternative plan and presenting to the customer the selected alternative plan in the form of a recommendation,

wherein the (a) estimating the real bill, the (b) simulating estimating the bill, and the (c) estimating the churn risk are performed by the processor using data supplied by the first service server.