Eradication is a management strategy that can provide substantial ecological and economic benefits by eliminating incursions of pest organisms. In contrast to eradication efforts that target other pests, weed eradication programs can be very protracted owing to the presence of persistent seed banks and difficulties in detecting the target. Hence there is a need to develop criteria to assist in the evaluation of progress towards eradication. Knowledge of the extent of a weed incursion (the ‘delimitation’ criterion) is considered fundamental for eradication success, as an incursion will progress from any infestations that remain undetected and thus uncontrolled. This criterion is examined with regard to eradication programs targeting Bassia scoparia L. A.J. Scott [= Kochia scoparia L. Schrader], Chondrilla juncea L. (both in Western Australia) and Orobanche ramosa L. in South Australia. The B. scoparia incursion, which has been eradicated, was largely delimited within 12 months of the inception of its eradication program. In contrast, the Western Australian C. juncea incursion has never been delimited, owing to insufficient investment in surveillance during an eradication program spanning 30 years. An exponential decrease in the detection ratio (infested area detected/area searched) over time suggests that delimitation has been approached within 6 years of the inception of the eradication program for O. ramosa. An effective surveillance program is essential for achieving delimitation of a weed incursion.