A series of trials were conducted in different experimental and feeding conditions to evaluate a purified and enriched lignin extracted from Eucalyptus grandis (LIPE®) as a marker to estimate faecal output in sheep fed temperate and tropical grasses. Two indoor trials were conducted in metabolism cages: the first one (Trial I) was conducted with sheep fed Italian ryegrass (Lolium multiflorum Lam.) pastures, which was evaluated in two different periods [Period I (n = 14), and Period II (n = 15)], and the second one (Trial II) was performed with sheep fed pearl millet [Pennisetum americanum (L.) Leeke] pastures (n = 16). Grazing trials were performed with sheep grazing Italian ryegrass [Trial III (n = 62)] and pearl millet [Trial IV (n = 34)]. According to the manufacturer's statements, sheep were dosed once a day with capsules containing 250 mg of LIPE® for a period of seven days, beginning two days before total faecal collections started and finishing on the last faecal collection day. To measure the observed daily faecal output, sheep were harnessed with faecal bags that were emptied once a day at 0800 h for five days. Faecal subsamples from indoor and grazing trials were taken to determine the LIPE® content by infrared spectroscopy. Regression analyses performed between estimated and observed faecal output (g DM/kg LW/day) by sheep were not significant (p > 0.05) in three of the four evaluated trials. Faecal LIPE® recovery from the total data of the four trials showed a statistical difference when compared to the reference value (100%). In conclusion, our results show that the LIPE® marker presents low accuracy to estimate the daily faecal output of sheep fed temperate (Italian ryegrass) and tropical (pearl millet) grasses.