Purpose The purpose of this paper is to identify actual (as-is) patient pathway variation among breast cancer patients and to investigate the relationship between pathways and the cost incurred by patients. Design/methodology/approach Both quantitative and qualitative methods were employed to analyze data from four Swedish hospital groups. Quantitative methods include event-log data mining and statistical analyses on the related patient cost from the Swedish breast cancer quality registry and case-costing system. Qualitative methods included collaboration with and interviewing domain experts. Findings Unique pathways, followed by only one patient, were generally costlier than the most and less frequent pathways. Earlier study findings are confirmed for mastectomy patients, with more frequent pathways having a lower cost, whereas contradicting and inconclusive results emerged for the partial mastectomy patient groups. Highest variation in pathways was identified for patients receiving chemotherapy. Practical implications The common belief – if one follows a standardized patient pathway, then the cost will be lower – should be re-examined based on the actual pathways that occur in reality. Originality/value The relationships between patient pathways and patient cost allow more complex insights, beyond the general causal relationship between successfully implementing a “to-be” care pathway and lower cost. This highlights data-driven research’s importance, where actual pathways (as-is) provide more useful information than to-be care pathways.