Abstract Objectives Previous reports established the feasibility of a telehealth model for delivering speech–language therapy via Internet videoconferencing, which connects individuals with primary progressive aphasia (PPA) to an expert speech and language pathologist for treatment. This study reports feasibility of the same telehealth intervention in a larger set of progressive aphasia participants and explores factors potentially influencing functional intervention outcomes. Methods Participants with PPA or progressive aphasia in the context of a neurodegenerative dementia syndrome and their communication partners were enrolled into an 8-session intervention, with 3 evaluations (baseline, 2 months, and 6 months postenrollment). Half of the participants were randomized into a “check-in” group and received 3-monthly half-hour sessions postintervention. Mixed linear models with post hoc testing and percent change in area under the curve were used to examine communication confidence over time, as well as the influence of check-in sessions and the role of communication partner engagement on communication confidence. Results Communication confidence improved at the 2-month evaluation and showed no significant decline at the 6-month evaluation. Item-level analysis revealed gains in communication confidence across multiple communication contexts. Gains and maintenance of communication confidence were only present for the engaged communication partner group and were not bolstered by randomization to the check-in group. Discussion Internet-based, person-centered interventions demonstrate promise as a model for delivering speech–language therapy to individuals living with PPA. Maintenance is possible for at least 6 months postenrollment and is better for those with engaged communication partners, which supports the use of dyadic interventions.