Assessment of Marketing Communications and Customer Relationship Practices in Insurance Companies: Channel Responsiveness, the Limit of Contact and What Recovery Repairs

Insurers measure their communications by how much they send. This article measures what happens at the other end. It combines a mystery shopping audit, 240 scripted enquiries across sixteen insurers, five channels and three auditors, with a survey of 800 customers recording contact frequency, relevance, complaints and purchases. Three results stand out. The channel that answers best is the oldest: a branch replies in 1.2 hours with an accuracy of 72 of 100, against 29 hours and 61 points for a website form, with messenger and the mobile app between them; the three auditors agree closely on the ranking of insurers (Kendall W = 0.87), so the differences are not an artefact of who shopped. Contact has a limit. The relationship between the number of contacts a year and the probability of buying a second product turns down at 10.7 contacts, with a bootstrap interval of 9.3 to 14.4, and the squared term is significant (t = minus 4.06); each additional contact also raises the probability of opting out of communication by 0.90 points, while each point of the relevance index lowers it by 0.20. Ten more points of relevance are worth 5.6 points of cross-selling, against 2.4 points for one more contact at the market average of 7.3. The service recovery paradox is not supported: customers whose complaint was resolved well remain 0.44 points less satisfied than customers who never complained (t = minus 4.07) and 12 points less likely to intend renewal, although they are 0.77 points more satisfied than customers left unresolved. Across insurers, the audit accuracy score correlates with customer satisfaction at 0.677, so what an auditor observes in fifteen enquiries tracks what customers report.