In horses, the quantity of faeces and the faecal concentrations of plant and synthetic alkanes are inconsistent throughout the day. The estimation of feed intake and digestibility can additionally be limited by irregular and incomplete faecal recovery of alkanes that are used as dietary markers. The correction of alkane concentrations minimizes the bias of estimates, but requires the determination of faeces quantity by total collection. However, in consideration of the dynamics of alkane concentrations in faeces, sampling at selected timeframes throughout a day may be useful in avoiding such correction. Five adult horses were fed a hay-based diet offered three times a day in equal amounts. Horses received a bolus with similar quantities of n-octacosane (C28), n-dotriacontane (C32) and n-hexatriacontane (C36) synthetic alkanes twice a day. Total faeces were quantified over 3 consecutive days. Dry matter intake (DMI), output (DMO) and digestibility (DMD) were determined from the total collection trial and additionally estimated for each of 12 equal timeframes throughout the day. The diurnal patterns of the single faeces quantity (SFQ) and faecal alkane concentrations were similar between horses and were repeated from day to day. The intra-day dynamic of SFQ was pronounced. The dynamic of the faecal concentration was much more pronounced when the alkane was administered twice instead of three times a day. The faecal recovery of alkanes that has been calculated from the total collection trial ranged from 82 ± 4·1% for C36 to 108 ± 11·1% for C28. Measured DMI was 12·0 kg/day, measured DMO was 5·9 kg/day and measured DMD was 0·51. Reliable estimates were obtained for DMI with 12·3 ± 0·79 kg/day for the combination of n-nonacosane (C29) and C28 and 12·1 ± 1·01 kg/day for the combination of n-tritriacontane (C33) and C28 at 2 h after administration, and 12·1 ± 0·96 kg/day for the combination of n-hentriacontane (C31) and C32 at 2 h prior to the morning meal, which included the first bolus administration. When calculated from DMO and DMD, DMI was 12·2 ± 0·89 kg/day for C29 and 12 ± 1·0 kg/day for C33 between 5 and 6 h after the morning meal. Estimates of DMD were unbiased between the 3rd and 4th hour after the morning meal with 0·52 ± 0·014 for C29 and 0·51 ± 0·021 for C33, respectively. The DMO was 5·7 ± 0·34 kg/day and 6·1 ± 0·43 kg/day when estimated 3–4 h after the 2nd meal, or prior to the 2nd bolus administration, using the product of SFQ and the daily defecation frequency or the synthetic alkanes, respectively. Knowledge of defecation dynamics might be helpful for simplifying experimental trials. They specifically followed intake dynamics, which can prospectively be used to select sampling timeframes. Based upon current results, a selection of two to three spot samples of faeces that are evenly distributed between 2 h before and 6 h after the morning meal, which was the time of bolus administration, allows for the greatest reliability. Defecation dynamics are probably less influenced by ration/bolus type, rate of exercise, or gut peristalsis, which nevertheless can result in individual shifts of optimal timeframes.