Pollution of the environment by metals and organic contaminants is an intractable global problem, with cleanup costs running into billions of dollars using current engineering technologies. The availability of alternative, cheap and effective technologies would significantly improve the prospects of cleaning-up metal contaminated sites. Phytoremediation has been proposed as an economical and ‘green’ method of exploiting plants to extract or degrade the contaminants in the soil. To date, the majority of phytoremediation efforts have been directed at leaping the biological, biochemical and agronomic hurdles to deliver a working technology, with scant attention to the economic outlook other than simple estimates of the cost advantages of phytoremediation over other techniques. In this paper we use a deterministic actuarial model to show that uncertainty in project success (the possibility that full clean up may not be realized) may significantly increase the perceived costs of remediation works for decision-makers.