Getting a “sweet” deal: does healthfulness of a sub-brand influence consumer loyalty?

Abstract

Purpose Increasing and maintaining the population’s consumption of healthful food may hinder the global obesity pandemic. The purpose of this paper is to empirically test whether it is possible for healthful sub-brands to achieve higher consumer behavioural loyalty than their less healthful counterparts. Design/methodology/approach The study analysed three years of consumer panel data detailing all purchases from five consumer goods categories for 15,000 UK households. The analysis uses best-practice techniques for measuring behavioural loyalty: double jeopardy, polarisation index, duplication of purchase and user profile comparisons. Each sub-brand’s healthfulness was objectively coded. Findings Despite the level of healthfulness, all sub-brands have predictable repeat purchase patterns, share customers as expected and have similar user profiles as each other. The size of the customer base, not nutrition content, is, by far, the biggest determinant of loyalty levels. Research limitations/implications Consumers do not show higher levels of loyalty to healthful sub-brands, or groups of healthful sub-brands. Nor do they buy less healthful sub-brands less often (as a “treat”). There are also no sub-groups of (health conscious) consumers who would only purchase healthful options. Practical implications Sub-brands do not have extraordinarily loyal or disloyal customers because of their healthfulness. Marketers need to focus on growing sub-brands by increasing their customer base, which will then naturally grow consumer loyalty towards them. Originality/value This research brings novel evidence-based knowledge to an emerging cross-disciplinary area of health marketing. This is the first study comparing behavioural loyalty and user profiles towards objectively defined healthful/less healthful sub-brands.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 19 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.