OBJECTIVE: To investigate labour duration in different phases of labour when adhering to Zhang’s guideline for labour progression compared with the WHO partograph. DESIGN: A secondary analysis of a cluster randomised controlled trial. SETTING: Fourteen Norwegian birth care units, each with more than 500 deliveries per year constituted the clusters. PARTICIPANTS: A total of 7277 nulliparous women with singleton foetus in a cephalic presentation and spontaneous onset of labour at term were included. INTERVENTION: Seven clusters were randomised to the intervention group that adhered to Zhang’s guideline (n = 3972) and seven to the control group that adhered to the WHO partograph (n = 3305) for labour progression. MEASUREMENTS: The duration of labour from the first registration of cervical dilatation (≥ 4 cm) to the delivery of the baby and the duration of the first and second stages of labour; the time-to-event analysis was used to compare the duration of labour between the two groups after adjusting for baseline covariates. FINDINGS: The adjusted median duration of labour was 7.0 h in the Zhang group, compared with 6.2 h in the WHO group; the median difference was 0.84 h with 95% confidence interval [CI] (0.2-1.5). The adjusted median duration of the first stage was 5.6 h in the Zhang group compared with 4.9 h in the WHO group; the median difference was 0.66 h with 95% CI (0.1-1.2). The corresponding adjusted median duration of the second stage was 88 and 77 min; the median difference was 0.18 h with 95% CI (0.1-0.3). KEY CONCLUSIONS: The women who adhered to Zhang’s guideline had longer overall duration and duration of the first and second stages of labour than women who adhered to the WHO partograph. IMPLICATIONS FOR PRACTICE: Understanding the variations in the duration of labour is of great importance, and the results offer useful insights into the different labour progression guidelines, which can inform clinical practice.