The objective of this randomized controlled trial was to assess whether making HIV selftest kits available as a testing option at North Star Alliance roadside wellness clinics in Kenya and advertising their availability via SMS (text message) would increase HIV testing rates among truckers and sex workers who were irregular testers, compared with sending text messages about HIV testing in general. A sample of 2,262 male truckers and 2,196 female sex workers registered in the North Star Alliance electronic medical record system and who, based on these records, were irregular HIV testers were randomized to one of three study arms in which they were sent text messages about the HIV testing available to them at North Star Alliance Kenya clinics. Specifically, the different arms received a text, either: (1) communicating the availability of HIV self-test kits at all North Star Alliance Kenya clinics, which was sent three times (intervention); (2) about the availability of HIV testing in general at North Star Alliance clinics, sent three times (enhanced standard of care (SOC)); or (3) about HIV testing at North Star Alliance clinics in general, sent once (SOC). When arriving at a North Star Alliance clinic in Kenya, those in the enhanced SOC and SOC groups were only offered one HIV test, the standard provider-administered rapid finger-prick blood HIV test. Those in the intervention group, when arriving in a clinic, were offered a choice between (1) the SOC HIV test; (2) a rapid oral HIV self-test for self-administration in the clinic with supervision; or (3) the rapid oral HIV self-test kit to take for home use with phone-based post-test counseling. We followed each sample for two months and looked at differences in HIV testing using logistic regression. Truckers and sex workers in the intervention arm were significantly more likely to test for HIV during follow-up compared with those in the enhanced SOC (OR = 2.7, p = 0.007 and OR = 1.9, p = 0.001 respectively). There was no difference in HIV testing between those in the enhanced SOC and the SOC arms for either sample. The intervention proved to be more cost effective than either the enhanced SOC or the SOC, even though the HIV self-test kits cost much more than the standard provider-administered HIV test because of the higher HIV testing rates in the intervention arm. Offering HIV self-testing to high-risk populations such as truckers and sex workers and advertising its availability via text message may be a cost-effective way to increase HIV testing rates in these important groups.