Low-literacy populations in Kenya face significant challenges accessing digital services effectively. A mixed-methods approach was employed, combining surveys with qualitative interviews to gather insights from users and experts. Users preferred graphical interfaces over text-based ones, indicating a preference for visual cues (direction: 60%, proportion: 2/3). Accessible user interfaces significantly improve digital service accessibility for low-literacy populations in Kenya. Design guidelines should prioritise the integration of visual elements and simplified language to enhance usability. User Interface, Accessibility, Low-Literacy, Digital Services, Kenya Model estimation used $\hat{\theta}=argmin_{\theta}\sum_i\ell(y_i,f_\theta(x_i))+\lambda\lVert\theta\rVert_2^2$, with performance evaluated using out-of-sample error.