This paper aims to identify empirically how household structure and other socioeconomic characteristics affect the lifecycle consumption profile, exploiting data available for Spain for the last 15 years. Given the nature of the data, we build a pseudo-panel of cross-sections. After reviewing the previous literature, adjusting cohort averages to time-invariant and time-variant household characteristics, we propose employing the multilevel modelling approach WBRE (Within-Between Random Effects model). Results confirm the relevance of income as a predictor of household consumption while unveiling some interesting insights. Woman-headed households have a higher consumption profile than their man-headed counterparts after controlling for income. We observe a negative effect for number of toddlers when this characteristic is considered as a lifecycle predictor, while its impact is positive when comparisons are made between cohorts. Changing status from renter to owner has a positive within effect (lifecycle predictor), while the effect of owning on consumption is negative when contextual comparisons are made.