Insomnia affects 30-60% of cancer patients, thus making it one of the most common disturbances in this population. When untreated, which is the rule rather than the exception, insomnia often becomes chronic. Chronic insomnia is associated with numerous negative consequences (e.g., increased risk for psychological disorders, health care costs). A large body of evidence supports the efficacy of cognitive-behavioral therapy for insomnia (CBT-I) in cancer patients, but CBT-I is still not offered routinely in cancer clinics. Self-administered CBT-I (e.g., video-based intervention) has been developed to increase patients' access to this treatment. However, results of clinical trials have suggested that these minimal treatments would be better used as a first step of a stepped care model. In stepped care, patients receive only the level of intervention they need. Generally, the entry level is a minimal, less costly, intervention (e.g., self-help intervention) followed by a more intensive form of treatment if needed (if the patient is still symptomatic). The investigators have recently assessed the efficacy of a stepped care model to administer CBT-I in cancer patients, which includes a web-based CBT-I (called Insomnet) followed by up to 3 sessions with a psychotherapist if the patient is still symptomatic. Results of this study suggest that this model of care is non-inferior to a standard face-to-face treatment (Savard, Ivers, et al., in revision), while being more cost-effective. A stepped care CBT-I could therefore be offered in routine cancer care clinics. This project will assess the feasibility and effectiveness of implementing a stepped care CBT-I in real-world cancer clinics, using a non-randomized stepped wedge design to compare the effects of our program (active phase) with a passive phase. The program is called Insomnia in Patients with Cancer - Personalized Treatment (IMPACT). The stepped care CBT-I (active intervention) is being implemented sequentially in the four participating hospitals over a number of equally spaced time periods of 4 months (wedges), for a total of 5 time points, over a period of 20 months.