Preventing surrender by holders of life insurance policies Dr. Gerhard Braas Hamburg-Mannheimer Group Design and implementation of the project Customer Relationship Management within Hamburg-Mannheimer sales and marketing Field test for implementation of a system for early identification of cancellation potentials using latest data processing and analyzing methods (Data mining process) SAS user conference SeUGI 19 - Florence, Italy, May 31, 2001 /G/Projekte/Datamine/Test26 SAS SeUGI Florenz 310501
Business in force in 2000 Hamburg-Mannheimer Group number of contracts premium income 57% 7.32 million 5.47 million 43% 78% 22% 5,560 million DM 1,530 million DM 12.79 million contracts 7,090 million DM life non-life business (property damage/liability/motor/legal defence)
Overview of sales and marketing organizations HMS DANV/ OAB OVG HMI coop. DKV coop. DreBa brokers sales approach agency business target group business S/AFE (financial planning) target group business multi-level sales system agency business target group business target group emloyees self-employed/ free-lance customers customers with existing contracts lawyers tax consultants university graduates civil servants senior citizens (no specific target group defined) (no specific target group defined) (no specific target group defined) (no specific target group defined) plus customer service organization (Kundendienst) dealing with reversals of cancellations and follow-up of applications
A total of 23,880 agents are working for the Hamburg-Mannheimer Group full-time/ main-occupation agents main and secondary occupation agents 14,961 8,919 3,235 4,716 968 employees agencies (incl. trainees) self-employed agents main-occupation agents secondary occ. agents as per 12/00
Project: Customer Relationship Management within Hamburg-Mannheimer sales and marketing Data mining process: Customers with a major risk of cancellation can be identified with high accuracy well before a cancellation actually occurs Field test in different sales channels: Will latest analyzing methods (data mining) help to - reduce cancellation rates and stabilize policies in force? - increase new business and create additional growth? Project target: System for early identification of cancellation potentials aiming at avoidance of cancellations and higher customer loyalty
Application test Area: South-West Germany and Saxony/Thuringia Selection: Capital life insurance policies Participants: HMS (Main Organization) HMI-Agenturvertrieb (agencies) Kundendienst (KD - customer service) Sample: Target of about 20,000 customer contacts Duration: March 11 to August 21, 2000
Area: Saxony/Thuringia South-West Germany
Data transfer and report on customer contact Provision of data - Data mining process with precision at beginning of test Transmission of data to agents - Specific customer service instructions Test CRM Questionnaire - Report on customer contact and discussion
Für G 43 KD-Auftrag wegen Test Kundenbindung vom 01.03.00 Blatt 1 Anfordernde Stelle: VM-BSK ** Buch-Nr. 6586123-18 ** Ausgabe am an Agentur-Nr. 1616 Beitragszahler Helmut Henkel ** Alle Beträge sind in DM ** Kirchweg 44a Buchungsdatum 17.12.99 77716 Fischhausen Fälligkeit 1.00 180,50 Nass Spk Konto ausgeglichen Zahlungsweg: Z-Nr. 2 Abruf BLZ 500 505 15 60016 Frankfurt, Main Konto-Nr. 224004823 Letzte Zahlungsbuchungen Verarb.Dat. Buch.Dat. Lastschrift Gutschrift 27.07.99 01.08.99 Abruf 180,50 26.08.99 01.09.99 Abruf 180,50 27.09.99 01.10.99 Abruf 180,50 26.10.99 01.11.99 Abruf 180,50 25.11.99 01.12.99 Abruf 180,50 14.12.99 01.01.00 Abruf 180,50 KD-Aufträge und KD-Berichte in den Vormonaten: KD-Auftr. vom 08.01.98 27.05.97 KD-Auftr. wegen ANF.G 43 KUEND. ausgegeben an KDM KDM KD-Lst. Vertr. 43 43 KD-B erfaßt 01.98 05.97 KD-Auftrag vom 08.01.98 27.05.97 Art d.berichts KUENDIG. KUENDIG. Bewertung VOLL.ERF K.BEARB. Kennziffern 212 400 330 400 461 690 Mitarbeiter BRAUN,PETE BRAUN,PETE GST/Vertr. 85 85 Kurzangaben zu den Versicherungen dieser Bnr Vsnr PZ Tarif Vers.Se Beitrag V.Beg B.End Prog Name 21444134-36 3 9694 38,10 07.80 07.03 102 Sigrid Henkel VN 21444135-37 3WZA 7017 39,65 07.86 07.03 102 Sigrid Henkel VN 30689820-34 3 5927 28,45 08.83 08.03 102 Helmut Henkel VN 35798723-46 2M 10000 64,10 07.94 02.00 102 Helmut Henkel VN
E R H E B U N G S B O G E N Z U M T E S T K U N D E N B I N D U N G BNR: AD-Nr.:71 G st.-n r.: Agt.-Nr.: A. K U N D E N K O N T A K T 1. Hat ein ausführliches Gespräch mit dem Kunden stattgefunden? ja n e in M a r k t f o r s c h u n g B. p e rsö n lich e r B e s u c h weiter w e il K u n d e n ich t e rre ic h b a r weiter T e le fo n a t mit B w eil ich abgew iesen w urde mit C w eil Kunde storniert hat G E S P R Ä C H S B E R I C H T 2. W ann wurde der Kunde zuletzt persönlich 2000 1999/98 1997 vor 1997 w eiß nicht von einem HM-Vermittler besucht? 3. Ist bei diesem letzten Besuch ein Vertrag ja n e in w e iß n ich t abgeschlossen oder erhöht worden? 4. Konnten Sie feststellen, daß der Kunde innerhalb d e r le tzte n 1 2 M o n a te p e rs ö n lic h v o n e in e m ja n e in Verm ittler eines W ettbew erbers besucht w urde? 5. Hat der Kunde geäußert, daß er finanzielle ja n e in Schwierigkeiten hat? 6. Konnten Sie feststellen, daß der Kunde unzufrieden mit der Ham burg-mannheim er ist? ja Grund: nein 7. Hat der Kunde eine Kündigungsabsicht geäußert? ja Konnten Sie eine erfolgreiche Stornobekäm pfung durchführen? ja, durch eine Vertragsänderung ja, d u rc h N e u a b sch lu ß ja, durch Argum ente/erläuterungen n e in, S to rn o b e k ä m p fu n g w a r n ich t m ö g lich nein 8. Haben Sie Neugeschäft abgeschlossen bzw. steht Neugeschäft in Aussicht? ja, ich habe Neugeschäft abgeschlossen (Neuvertrag/Erhöhung) ja, N e u g e s c h ä ft s te h t in A u s s ic h t (F o lg e te rm in ) nein, m om entan steht kein Neugeschäft in Aussicht 9. W ie ist Ihre persönliche Einschätzung bezüglich dieses Kunden: Handelt es sich um einen ja n e in stornogefährdeten Kunden? C. S O N S T I G E S 10. W eitere Bem erkungen Bitte zurück an G-Stelle und von dort an HV TG-D Bei Fragen zum Ausfüllen des Bogens: Hotline von Tel. 040/6376-4245
Information and accompanying measures Regional kick-off meeting to begin test Information of sales organizations Internal publications within Hamburg-Mannheimer Motivational letter by chairman of the managing board Telephone hotline
Test method: customer contacts Testing of policies with high cancellation risk and randomly selected policies No hint at cancellation potential of the specific contract Test method: drawer Will the data mining model be verified? Will the measures taken by the sales organizations result in higher customer loyalty? Customer contact Drawer 1 14,000 risk of cancellation risk of cancellation 14,000 3 14,000 random selection random selection 14,000 2 4 Success Data Mining Model: comparison 3 to 4 Success HM sales force: comparison 1 to 3
Test results: Highly motivated sales power 17,046 attempted contacts by about 600 agencies, other agents and customer service agents 13,242 detailed discussions with customers 3,045 customers signed additional new business 4,591 new contracts Additional 345 contracts within four weeks after end of test Nearly 40% of all detailed discussions with customer resulted in new business
Test results: Additional new business Success in new business does not depend on whether customers - had been scored as high cancellation risk customers in the data mining process; - or belonged to the randomly selected group representative of Hamburg-Mannheimer s customer base. In only 78 cases new business was done over the phone. Personal visits by agents still are the most efficient method of successful acquisition.
Test results: Higher stability of policies in force 1,099 customers expressed their intention to cancel their contract to Hamburg-Mannheimer agents. Agents were successful in preventing the impending cancellation in 50 % of all cases through: - 174 changes or rearrangements of contracts - 14 additional new contracts - 358 single convincing discussions between client and customer
Test results: Higher stability of policies in force Shortly after the end of the application test the positive effects of the sales activities began to show. The customer visits during the test resulted in a decrease in cancellation rates by 25 %. Thus our specific methods developed to prevent cancellations are much more efficient than usual customer service approaches. These clearly positive results could be seen in the high-risk customer group as well as in the randomly selected group representative of Hamburg-Mannheimer s customer base.
Test results: Cancellation rates in life insurance 7 months later Detailed discussions with customers: method of preventing cancellations Drawer as control group: usual customer service approach 1 risk of cancellation confirmed Index: 110 Index: 147 risk of cancellation confirmed 3 2 random selection Index: 77 Index: 100 random selection 4 representative of HM customer base
Data mining - Advantages for sales and marketing Stabilization of policies in force as an active and preventive measure Customer contacts without potential for conflict or time pressure Reduced costs for efforts to reverse cancellations Higher net growth and higher cross selling rate Safeguarding of income from policy service Opportunity for additional business
Data Mining - Advantages for Hamburg-Mannheimer Growth: Reduction of cancellation rates in new business and existing policies Extension of policies in force through additional new contracts Higher customer satisfaction Returns: Higher profitability through employment of latest technologies Reduced costs for efforts to reverse cancellations
Contact details Dr. Gerhard Braas Quality Management () Hamburg-Mannheimer Versicherungs-AG Überseering 45, D - 22297 Hamburg Germany Tel. (++49) (0)40 6376-4163 Fax. (++49) (0)40 6376-3384 E-mail: gerhard.braas@hamburg-mannheimer.de