Factors Driving Duration to Cross-Selling in Non-Life Insurance: New Empirical Evidence from Switzerland
Open accessThe DOI matches its Crossref record; 39 of the 40 references it cites were found in a registry or an index, and 1 are not checked yet; 3 of the 19 sources its claims cite were checked for support.
Abstract
Customer relationship management and marketing analytics have become critical for non-life insurers operating in highly competitive markets. As it is easier to develop an existing customer than to acquire a new one, cross-selling and retention are key activities. In this research, we focus on both car and household-liability insurance products and consider the time a customer owning only a single product takes before buying the other product at the same insurer. Based on longitudinal consumer data from a Swiss insurance company covering the period from 2011 to 2015, we aim to study the factors driving the duration to cross-selling. Given the different dynamics observed in both products, we separately study the car and household-liability insurance customer cohorts. Considering the framework of survival analysis, we provide descriptive statistics and Kaplan–Meier estimates along major customer characteristics, contract history and distribution channel usage. For the econometric analysis of the duration, we compare the results from Cox and accelerated failure time models. We are able to characterize the times related to the buying behavior for both products through several covariates. Our results indicate that the policyholder age, the place of residence, the contract premium, the number of contracts held, and the initial access channel used for contracting influence the duration to cross-selling. In particular, our results underline the importance of the tied agent channel and the differences along the geographic region and the urbanicity of the place of residence. By quantifying the effects of the above factors, we extend the understanding of customer behavior and provide a basis for developing models to time marketing actions in insurance companies.
Verified 2026-11-19 · 1 reference not checked yet
Overview
CoChat Summary
## TL;DR
The paper studies how long it takes Swiss non-life insurance customers to cross-buy the second product (car or household-liability) after first buying one product in 2011, using five years of longitudinal data from one insurer. It finds that cross-selling duration differs by cohort and is significantly related to age, residence, contract premium, number of contracts, and access channel, with tied agents generally associated with faster cross-buying and brokers with slower cross-buying.
## Key findings
- The dataset contains 60,752 customers: 28,388 CA customers and 32,364 HL customers, observed from 2011 to 2015.
- Portfolio outcomes over 2011–2015: the share of customers holding both products rises from 6% in the first year to 15% in 2015, while lapsing customers reach 22% in 2015.
- Right-censoring is substantial: after five years, 81% of the CA cohort and 83% of the HL cohort remain single-product customers.
- The paper reports that the same core covariates are significant in both cohorts: place of residence, number of contracts, access channel, age, and contract premium; it also states that tied agents are associated with shorter durations and brokers with longer durations.
- Specific descriptive quantiles include, for the CA cohort, T5% = 250 days and T10% = 781 days for ages 41–65, versus T15% = 349 days for ages 18–25; for the HL cohort, rural customers reach T10% = 289 days versus 849 days for urban customers.
- Model fit for the AFT distribution is best with log-normal in both cohorts, with BIC values of 106,306 (CA) and 107,652 (HL), lower than exponential, Weibull, and gamma alternatives.
## Methods
- The study uses a longitudinal dataset from a Swiss insurer covering all new customers who bought car (CA) or household-liability (HL) insurance in 2011, followed for five years.
- It applies survival analysis to inter-purchase times, using Kaplan-Meier estimates plus separate cohort analyses for CA and HL because the cohorts show different dynamics.
- For econometric modeling, it estimates Cox models and accelerated failure time (AFT) models, optimizing via BIC and using evolutionary trees to bin continuous variables such as age and premium into categories.
## Limitations
- The paper explicitly notes that generalization is limited because the data come from one single insurer, and observed dynamics may reflect the insurer’s unobserved cross-selling strategy and marketing actions, which the study cannot control for.
- It also states that the five-year observation window produces substantial right-censoring, and that cross-buying at competitors is unobservable.
- Evident from the text: the study analyzes only customers who first bought CA or HL in 2011, so the findings are specific to that cohort and period.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures










Verification
Claims the paper makes
2 Valid · 2 Not valid · 14 Not checked · 83 Own results and interpretations
Every source the paper cites for this claim was checked and supports it.
In the paper
“In line with the results of Staudt and Wagner (2018), brokers have a negative influence on cross-buying since independent brokers propose the most adequate product from any company to their customers and do not give a particular weight to earlier purchases at a given firm.”
What the sources say
- Supportedcites Staudt and Wagner (2018) matched to “What policyholder and contract features determine the evolution of non-life insurance customer relationships?” (2018) source
The Regression results (§5.2) report that customers buying through brokers are less likely to cross-buy than those using tied agents. The Conclusion (§7) repeats that comparison and notes that brokers, unlike tied agents, are not tied to a single insurer. (LLM, conf=0.91)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on an assumption not shown here.
- Supported
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Thereby, our numerical results support, among others, the relevance of the tied agent channel, the differences along geographic regions and the importance of the urbanicity of the place of residence.”
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Our econometric study shows that the place of residence, the number of contracts held, the access channel, the age and the premiums of the contracts held are relevant factors that significantly relate to the duration to cross-selling.”
boostedstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Thereby, our numerical results support, among others, the relevance of the tied agent channel, the differences along geographic regions and the importance of the urbanicity of the place of residence.”
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The share of right-censoring is lower in the case of the CA cohort with 81% remaining single-product customers after five years. This share is 83% for the HL cohort.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on an assumption not shown here.
63 more claims
Create a free account to see all 101 claims and every source verdict.
- Probably not a substantial claim. 24
- It may state two separate claims as one. 14
- The quote only partly states this claim. 2
Citations
Cites
Estimating the Dimension of a Model
Schwarz, G. (1978) · The Annals of Statistics · doi:10.1214/aos/1176344136 · Cited by 47,033
Generalized Additive Models: An Introduction with R
Wood, S. N. (2017) · doi:10.1201/9781315370279 · Cited by 13,852
An Introduction to Statistical Learning
James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013) · Springer Texts in Statistics · doi:10.1007/978-1-4614-7138-7 · Cited by 12,106
The Statistical Analysis of Failure Time Data
Kalbfleisch, J. D., & Prentice, R. L. (2002) · Wiley Series in Probability and Statistics · doi:10.1002/9781118032985 · Cited by 3,219
Supervised and Unsupervised Discretization of Continuous Features
Dougherty, J., Kohavi, R., & Sahami, M. (1995) · Machine Learning Proceedings 1995 · doi:10.1016/b978-1-55860-377-6.50032-3 · Cited by 1,936 · cited 3×
“Dougherty et al. (1995) and Ohlsson and Johansson (2010) propose to 353 group consecutive values to create a category.”
BackgroundDefensive Marketing Strategy by Customer Complaint Management: A Theoretical Analysis
Fornell, C., & Wernerfelt, B. (1987) · Journal of Marketing Research · doi:10.1177/002224378702400401 · Cited by 1,395
Survival and Event History Analysis
Aalen, O. O., Borgan, Ø., & Gjessing, H. K. (2008) · Statistics for Biology and Health · doi:10.1007/978-0-387-68560-1 · Cited by 813
Modeling Customer Lifetime Value
Gupta, S., Hanssens, D., Hardie, B., Kahn, W., Kumar, V., Lin, N., Ravishanker, N., & Sriram, S. (2006) · Journal of Service Research · doi:10.1177/1094670506293810 · Cited by 724
flexsurv : A Platform for Parametric Survival Modeling in R
Jackson, C. (2016) · Journal of Statistical Software · doi:10.18637/jss.v070.i08 · Cited by 558
Defensive Marketing Strategy by Customer Complaint Management: A Theoretical Analysis
Fornell, C., & Wernerfelt, B. (1987) · Journal of Marketing Research · doi:10.2307/3151381 · Cited by 542
Kamakura, W. A., Wedel, M., de Rosa, F., & Mazzon, J. A. (2003) · International Journal of Research in Marketing · doi:10.1016/s0167-8116(02)00121-0 · Cited by 336
The impact of satisfaction and payment equity on cross-buying
Verhoef, P. C., Franses, P. H., & Hoekstra, J. C. (2001) · Journal of Retailing · doi:10.1016/s0022-4359(01)00052-5 · Cited by 276 · cited 2×
“Research by Ngobo (2004); Verhoef and Donkers (2005); Verhoef et al. (2001) among others has been conducted to statistically describe and determine the factors influencing cross-buying in non-life insurance.”
BackgroundInvestigating Household Purchase Timing Decisions: A Conditional Hazard Function Approach
Jain, D. C., & Vilcassim, N. J. (1991) · Marketing Science · doi:10.1287/mksc.10.1.1 · Cited by 270
Applied Survival Analysis Using R
Moore, D. F. (2016) · Use R! · doi:10.1007/978-3-319-31245-3 · Cited by 270
Non-Life Insurance Pricing with Generalized Linear Models
Ohlsson, E., & Johansson, B. (2010) · EAA Lecture Notes · doi:10.1007/978-3-642-10791-7 · Cited by 260 · cited 4×
“%) have no declared damages (see, e.g., Denuit and Lang 2004; Klein et al. 2014; Ohlsson and Johansson 2010).”
BackgroundInfluentialJanssen, J., & Manca, R. (2006) · Springer eBooks · doi:10.1007/0-387-29548-8 · Cited by 239
The effect of acquisition channels on customer loyalty and cross-buying
Verhoef, P. C., & Donkers, B. (2005) · Journal of Interactive Marketing · doi:10.1002/dir.20033 · Cited by 235
Grewal, R., Mehta, R., & Kardes, F. R. (2004) · Journal of Marketing Research · doi:10.1509/jmkr.41.1.101.25090 · Cited by 229 · cited 2×
“Studies in the marketing literature primarily use demographic and economic variables as risk factors (Grewal et al. 2004).”
BackgroundFitting Cox's Regression Model to Survival Data using GLIM
Whitehead, J. (1980) · Applied Statistics · doi:10.2307/2346901 · Cited by 208
Cross-buying in retailing: Drivers and consequences
KUMAR, V, GEORGE, M., & PANCRAS, J. (2008) · Journal of Retailing · doi:10.1016/j.jretai.2008.01.007 · Cited by 184
Drivers of customers' cross‐buying intentions
Valentin Ngobo, P. (2004) · European Journal of Marketing · doi:10.1108/03090560410548906 · Cited by 173
Next-product-to-buy models for cross-selling applications
Knott, A., Hayes, A., & Neslin, S. A. (2002) · Journal of Interactive Marketing · doi:10.1002/dir.10038 · Cited by 161
evtree: Evolutionary Learning of Globally Optimal Classification and Regression Trees in R
Grubinger, T., Zeileis, A., & Pfeiffer, K.-P. (2014) · Journal of Statistical Software · doi:10.18637/jss.v061.i01 · Cited by 159
Klein, N., Denuit, M. M., Lang, S., & Kneib, T. (2014) · Insurance Mathematics and Economics · doi:10.1016/j.insmatheco.2014.02.001 · Cited by 107
Kamakura (2007) · doi:10.1300/j366v06n03_03 · Cited by 85
What Policy Features Determine Life Insurance Lapse? An Analysis of the German Market
Eling, M., & Kiesenbauer, D. (2014) · Journal of Risk and Insurance · doi:10.1111/j.1539-6975.2012.01504.x · Cited by 81
Investigating purchasing-sequence patterns for financial services using Markov, MTD and MTDg models
Prinzie, A., & Van den Poel, D. (2006) · European Journal of Operational Research · doi:10.1016/j.ejor.2004.05.004 · Cited by 75 · cited 3×
“…to apply Cox and accelerated failure time (AFT) models, while Cox models are typically used for such study (Ansell et al. 2007; Prinzie and Van den Poel 2006), we compare our results with AFT models that provide a direct relationship between the explanatory variables and the duration (see, e.g.,…”
BackgroundMethodInfluentialNon-life rate-making with Bayesian GAMs
Denuit, M., & Lang, S. (2004) · Insurance: Mathematics and Economics · doi:10.1016/j.insmatheco.2004.08.001 · Cited by 75 · cited 2×
“%) have no declared damages (see, e.g., Denuit and Lang 2004; Klein et al. 2014; Ohlsson and Johansson 2010).”
BackgroundIdentifying cross‐selling opportunities, using lifestyle segmentation and survival analysis
Ansell, J., Harrison, T., & Archibald, T. (2007) · Marketing Intelligence & Planning · doi:10.1108/02634500710754619 · Cited by 60
A data driven binning strategy for the construction of insurance tariff classes
Henckaerts, R., Antonio, K., Clijsters, M., & Verbelen, R. (2018) · Scandinavian Actuarial Journal · doi:10.1080/03461238.2018.1429300 · Cited by 56
Prinzie, A., & Van den Poel, D. (2007) · Decision Support Systems · doi:10.1016/j.dss.2007.02.008 · Cited by 44
The Need to Monitor Customer Loyalty and Business Risk in the European Insurance Industry
Guillen, M., Nielsen, J. P., & Pérez-Marín, A. M. (2008) · The Geneva Papers on Risk and Insurance - Issues and Practice · doi:10.1057/gpp.2008.1 · Cited by 36
Analytics of Insurance Markets
Frees, E. W. (2015) · Annual Review of Financial Economics · doi:10.1146/annurev-financial-111914-041815 · Cited by 35
Mau, S., Pletikosa, I., & Wagner, J. (2018) · International Journal of Bank Marketing · doi:10.1108/ijbm-11-2016-0180 · Cited by 23
Self-citationStaudt, Y., & Wagner, J. (2018) · International Journal of Bank Marketing · doi:10.1108/ijbm-11-2016-0175 · Cited by 22 · cited 6×
“As discussed in, e.g., Staudt and Wagner (2018), relationship management should integrate the whole customer lifecycle.”
BackgroundResultInfluentialSelf-citationEmpirical Findings on Motor Insurance Pricing in Germany, Austria and Switzerland
Laas, D., Schmeiser, H., & Wagner, J. (2016) · The Geneva Papers on Risk and Insurance Issues and Practice · doi:10.1057/gpp.2015.30 · Cited by 21 · cited 1×
“We use the covariates contained in our data, i.e., include customer information recorded at contract inception, to identify customers with similar buying dynamics (see also Frees 2015; Laas et al. 2016).”
MethodSelf-citationDuration of long-term care: Socio-economic factors, type of care interactions and evolution
Fuino, M., & Wagner, J. (2019) · Insurance Mathematics and Economics · doi:10.1016/j.insmatheco.2019.11.004 · Cited by 20
Self-citationPredictive Modeling Applications in Actuarial Science
(2016) · Cambridge University Press eBooks · doi:10.1017/cbo9781139342681 · Cited by 14
Mau, S., Cvijikj, I. P., & Wagner, J. (2015) · Zeitschrift für die gesamte Versicherungswissenschaft · doi:10.1007/s12297-015-0310-1 · Cited by 8 · cited 1×
“…have developed, analyzing the factors driving the Risks 2022, 10, 187. https://doi.org/10.3390/risks10100187 https://www.mdpi.com/journal/risks Risks 2022, 10, 187 2 of 20 customer journey (see, e.g., Staudt and Wagner 2018) and studying the next purchases (see, e.g., Mau et al. 2015, 2017, 2018).”
BackgroundInfluentialSelf-citation
Research works printed but not linked: 2
- While we are able to quantify the impact of the selected factors for cross-selling, our findings provide a first step for getting a better understanding in insurance customer dynamics. In fact, this is particularly relevant for practitioners as an intermediate step for timing marketing actions. As discussed in, e.g., Staudt and Wagner (2018), relationship management should integrate the whole customer lifecycle. We identify several directions for further research. For example, the inclusion of data about the cancellation of a contract and other products would allow to explain in more detail the customer journey. Calibrating semi-Markov models on historical longitudinal data could provide further probabilistic insights and the relevant factors for retaining existing customers (Janssen and Manca 2006). Moreover, the inclusion of more detailed risk factors and socio-economic variables may improve the modeling. Finally, the prediction power could be enhanced with a longer dataset and the application of more sophisticated methods like random forests. Author Contributions: Conceptualization, Y.S. and J.W.; methodology, Y.S. and J.W.; investigation, Y.S.; formal analysis, Y.S. and J.W.; data curation, Y.S.; writing-original draft preparation, Y.S.; writing-review and editing, Y.S. and J.W.; supervision, J.W. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. The APC was funded by the University of Lausanne. Data Availability Statement: Not applicable. Acknowledgments: Both authors are thankful to the insurance company providing the data for the analysis. Conflicts of Interest: The authors declare no conflict of interest.
- Staudt, Yves, and Joël Wagner. 2018. What Policyholder and Contract Features Determine the Evolution of Non-life Insurance Customer Relationships? A Case Study Analysis. International Journal of Bank Marketing 36: 1098-124. https://doi.org/10.1108/IJBM-112016-0175. - Staudt, Yves, and Joël Wagner. 2019. Comparison of Machine Learning and Traditional Severity-Frequency Regression Models for Car Insurance Pricing . Working Paper. Lausanne: University of Lausanne. · doi:10.1108/ijbm-112016-0175
Cited by
Parametric Survival Model on IPB University’s Graduation Data
Nashiruddin, M. A., Sumarno, H., & Budiarti, R. (2023) · JTAM (Jurnal Teori dan Aplikasi Matematika) · doi:10.31764/jtam.v7i2.12681
MODELLING THE BANK CUSTOMER ACTIVITY DURATION BASED ON THE COX ECONOMETRIC SURVIVAL MODEL
Piskunova, O. V., Klochko, R., Білик, Т. О., & Frolova, T. (2023) · Financial and credit activity problems of theory and practice · doi:10.55643/fcaptp.6.53.2023.4244
“This is often achieved by combining several existing methods, which complicates the process of building and evaluating models (Medina-Olivares et al., 2023; Mia et al., 2022; Staudt and Wagner, 2022).”
BackgroundMethodActuarial Analysis of Survival after Breast Cancer Diagnosis among Lithuanian Females
Levickytė, J., Skučaitė, A., Šiaulys, J., et al. (2024) · Healthcare · doi:10.3390/healthcare12070746
Versicherungsmarketing als betriebswirtschaftliche Funktion im Versicherungsunternehmen
Köhne, T. (2024) · Springer eBooks · doi:10.1007/978-3-658-44216-3_2
Sigorta Sektöründe İlişki Pazarlamasının Yapısal Özellikleri
BOYACI, G. E. (2024) · Güncel Pazarlama Yaklaşımları ve Araştırmaları Dergisi · doi:10.54439/gupayad.1566985
Metadata
- Authors
- Staudt, Y., & Wagner, J.
- Affiliations
- University of Applied Sciences of the Grisons (CH)
- University of Lausanne (CH)
- Corresponding author
- Wagner, Joël
- Venue
- Risks
- Publisher
- MDPI AG
- Published
- 2022-09-27
- Volume
- 10
- Issue
- 10
- Pages
- 187
- ISSN
- 2227-9091 (ISSN-L)
- OpenAlex
- W4297550056
- Cited by
- 7
- References
- 40
- Citations by year
- 2026: 1 · 2025: 1 · 2024: 3 · 2023: 2
- FWCI
- 1.20
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Risks (publishedVersion, cc-by)
- SERVAL (Université de Lausanne) (publishedVersion, cc-by)
- IRIS (submittedVersion, cc-by)
- RePEc: Research Papers in Economics (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- University of Lausanne
- Topics
- Social Sciences › Business, Management and Accounting › Marketing › Customer churn and segmentation
- Social Sciences › Economics, Econometrics and Finance › Economics and Econometrics › Insurance and Financial Risk Management
- Social Sciences › Business, Management and Accounting › Marketing › Customer Service Quality and Loyalty
- Keywords
- cross-selling, non-life insurance, automobile insurance, customer retention, distribution channels, urbanicity, geographic region, customer buying behavior
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-01, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-01 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.