INTRODUCTION: The Philippines, which has the fastest rising HIV epidemic globally, has limited options for HIV testing and its uptake remains low among cisgender men who have sex with men (cis-MSM) and transgender women (TGW), especially amid the COVID-19 pandemic. As HIV self-testing (HIVST) and technology-based approaches could synergize to expand uptake of HIV testing, we aimed to evaluate the outcomes of a community-led online-based HIVST demonstration and to explore factors associated with HIVST-related behaviours and outcomes. METHODS: We did a secondary data analysis among cis-MSM and TGW who participated in the HIVST demonstration, who were recruited online and tested out-of-facility, in Western Visayas, Philippines, from March to November 2020. We reviewed data on demographics, sexuality-, and context-related variables. Using multivariable logistic regression, we tested for associations between the aforementioned covariates and two primary outcomes, opting for directly-assisted HIVST (DAH) and willingness to secondarily distribute kits. RESULTS: HIVST kits were distributed to 647 individuals (590 cis-MSM, 57 TGW), 54.6% were first-time testers, 10.4% opted DAH, and 46.1% were willing to distribute to peers. Reporting rate was high (99.3%) with 7.6% reactivity rate. While linkage to prevention (100%) and care (85.7%) were high, pre-exposure prophylaxis (PrEP) (0.3%) and antiretroviral therapy (ART) (51.0%) initiation were limited. There were no reports of adverse events. Those who were employed, had recent anal intercourse, opted for DAH, not willing to secondarily distribute, and accessed HIVST during minimal to no quarantine restriction had significantly higher reactivity rates. Likelihood of opting for DAH was higher among those who had three or more partners in the past year (aOR = 2.01 [CI = 1.01-4.35]) and those who accessed during maximal quarantine restrictions (aOR = 4.25 [CI = 2.46-7.43]). Odds of willingness to share were higher among those in urban areas (aOR = 1.64 [CI = 1.15-2.36]) but lower among first-time testers (aOR = 0.45 [CI = 0.32-0.62]). CONCLUSIONS: HIVST could effectively reach hard-to-reach populations. While there was demand in accessing online-based unassisted approaches, DAH should still be offered. Uptake of PrEP and same-day ART should be upscaled by decentralizing these services to community-based organizations. Differentiated service delivery is key to respond to preferences and values of key populations amid the dynamic geographical and sociocultural contexts they are in.