Summary1. Mutualisms are only rarely one-to-one interactions: each species generally interacts with multiple mutualists. Exploitation is ubiquitous in mutualisms, and we would therefore expect that each mutualist interacts with multiple exploiters as well. Exploiter species may also interact with one another. For example, the action of one exploiter species might open the opportunity for exploitation by a second species. 2. Exploitation is common in many plant-pollinator mutualisms: 'primary' nectar robbers feed through holes they make in flowers, which can be subsequently used by 'secondary' nectar robbers unable to create holes themselves. The overall effect of nectar robbing on plant fitness is often (although not always) negative. No study has separated the effects of interacting with primary vs. secondary robbers. 3. Here, we examine the effects of primary vs. secondary nectar robbing on pollinator visitation rate and female fitness in Ipomopsis aggregata. Manipulating the type of nectar robbing that flowers experienced, we found that secondary nectar robbing inflicted fitness costs to plants beyond that inflicted by primary robbing alone. Secondary nectar robbing significantly reduced pollen receipt to flowers, as well as fruit and seed production. Although the causes are elusive, the effect may be attributed to changes in pollinator behaviour at these plants. 4. Synthesis. Our findings provide evidence that interacting with multiple exploiters can lead to increased negative effects for mutualists, and highlight the importance of incorporating multiple exploiters into the conceptual framework of mutualism.