Blended forms of learning have become increasingly popular. Learning activities within these environments are supported by a large variety of online and face-to-face interventions. However, it remains unclear whether these blended environments are successful, and if they are, what makes them successful. Studies suggest that blended learning challenges the self-regulatory abilities of learners, though the literature does little to explain these findings; nor does it provide solutions. In particular, little is known about the attributes that are essential to support learners and how they should guide course design. To identify such attributes and enable a more thoughtful redesign of blended learning environments, this systematic literature review (n = 95) examines evidence published between 1985 and 2015 on attributes of blended learning environments that support self-regulation. The purpose of this review is therefore to identify and define the attributes of blended learning environments that support learners' self-regulatory abilities. Seven key attributes were found (authenticity, personalization, learner-control, scaffolding, interaction, cues for reflection and cues for calibration). This review is the first to identify and define the attributes that support self-regulation in blended learning environments and thus to support the design of blended learning environments. This study may serve to facilitate the design of blended learning environments that meet learners' self-regulatory needs. It also raises crucial questions about how blended learning relates to well-established learning theories and provides a basis for future research on self-regulation in blended learning environments.