Exocomets scattered by planets have been invoked to explain observations in multiple contexts, including the frequently found near- and mid-infrared excess around nearby stars arising from exozodiacal dust. Here we investigate how the process of inward scattering of comets originating in an outer belt, is affected by the architecture of a planetary system, to determine whether this could lead to observable exozodi levels or deliver volatiles to inner planets. Using N-body simulations, we model systems with different planet mass and orbital spacing distributions in the 1-50 AU region. We find that tightly packed (Δap < 20RH, m) low mass planets are the most efficient at delivering material to exozodi regions (5-7% of scattered exocomets end up within 0.5 AU at some point), although the exozodi levels do not vary by more than a factor of ∼7 for the architectures studied here. We suggest that emission from scattered dusty material in between the planets could provide a potential test for this delivery mechanism. We show that the surface density of scattered material can vary by two orders of magnitude (being highest for systems of low mass planets with medium spacing), whilst the exozodi delivery rate stays roughly constant, and that future instruments such as JWST could detect it. In fact for η Corvi, the current Herschel upper limit rules our the scattering scenario by a chain of ≲30 M⊕ planets. Finally, we show that exocomets could be efficient at delivering cometary material to inner planets (0.1-1% of scattered comets are accreted per inner planet). Overall, the best systems at delivering comets to inner planets are the ones that have low mass outer planets and medium spacing (∼20RH, m).